Citation
The effects of cognitive style and gender on verbatim and gist memory for rapidly-presented montage video

Material Information

Title:
The effects of cognitive style and gender on verbatim and gist memory for rapidly-presented montage video
Creator:
Kenny, Robert F
Publication Date:
Language:
English
Physical Description:
ix, 140 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Analysis of variance ( jstor )
Cognitive style ( jstor )
Educational research ( jstor )
Learning ( jstor )
Memory ( jstor )
Montage ( jstor )
News content ( jstor )
Summarization ( jstor )
Viewers ( jstor )
Visual perception ( jstor )
Dissertations, Academic -- Teaching and Learning -- UF ( lcsh )
Teaching and Learning thesis, Ph.D ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 2002.
Bibliography:
Includes bibliographical references (leaves 124-139).
Additional Physical Form:
Also available online.
General Note:
Printout.
General Note:
Vita.
Statement of Responsibility:
by Robert F. Kenny.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
028813101 ( ALEPH )
50510693 ( OCLC )

Downloads

This item has the following downloads:


Full Text












THE EFFECTS OF COGNITIVE STYLE AND GENDER ON
VERBATIM AND GIST MEMORY FOR
RAPIDLY-PRESENTED MONTAGE VIDEO





By

ROBERT F. KENNY







A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY




UNIVERSITY OF FLORIDA


2002














ACKNOWLEDGMENTS

Special appreciation is extended to Jeffry Hurt, David Miller, and David Ostroff, members of my doctoral committee, for their support in this undertaking, and all their advice and counsel as I moved through the process of completing my degree. I took classes from all of them. With each course, I became more inspired and awed by their insight, knowledge, and enthusiasm for their profession. They have inspired me to be just like them.

I would especially like to thank Lee Mullally, my committee chair, for his

direction and support, and especially his determination not to let me take any shortcuts, or not to give my best efforts in the many classes and seminars I took from him, the reviews and edits for the textbook I wrote, and especially this study. He has my unending appreciation for his extraordinary patience and ability to keep me focused.

I am grateful to Chuck Braverman, whose kinestasis film was used in this study. I saw it almost 30 years ago and knew then it would somehow become a part of my life. I am particularly grateful to Randolph Wright from Pyramid Media, copyright holder of American Time Capsule, who graciously granted me permission to use it in this study. I am also thankful to Mitchell Stephens, whose book, The Rise of the Image, the Fall of the Word became the catalyst for this study. Other producers and researchers in new media like Jeff Scher and Annie Lang have also inspired me with their insights into visual perception, fast-cuts, and montage video. I am happy and lucky to have met them.





ii








I am grateful to all the many teachers and students who helped with the review of the test instruments, pilot test, and test administration, in particular Rachael Maland, James Keith, and Marshall Broitman. I am particularly grateful to Tom Wilson and Alan Smolowe, who were there with me every step of the way, with a constant watchfulness, keen insights, special interest, and concern for what I was doing.

To my family, I thank my son, Rob, and daughter, Erin, and their spouses who cheered me on, often wondering in amazement why I would attempt to go down such a path at this point in my life.

Lastly, I extend a very special thanks to my wife, Sandra, to whom I dedicate this work. I am extremely grateful for her endurance, sacrifice, and constant encouragement through this odyssey that started more than six years ago. While I often have wondered where the time has gone, I never could imagine spending it with anyone else.















TABLE OF CONTENTS

pqge

ACKN OW LDEGM EN TS .................................................................................................. ii

LIST OF TA BLES ............................................................................................................ vii

ABSTRA CT ..................................................................................................................... viii

CHAPTERS

I INTRODUCTION ......................................................................................................... I

Introduction .................................................................................................................... I
Statem ent of the Problem ............................................................................................... 2
N eed for the Study ......................................................................................................... 4
Background .................................................................................................................. I I
Purpose ......................................................................................................................... 12
Research Questions ...................................................................................................... 15
Delim itations of the Study ........................................................................................... 16
Lim itations of the Study ............................................................................................... 16
Definition of Term s ...................................................................................................... 17
Sum m ary ...................................................................................................................... 20

2 REV IEW OF TH E LITERATURE ............................................................................. 22

Perception and Know ledge .......................................................................................... 22
Neurology of Vision .............................................................................................. 24
V isual Perception Theories .................................................................................... 25
Fast Seeing ............................................................................................................. 28
Encoding ................................................................................................................ 31
Relationship of M em ory and Learning ........................................................................ 35
Textual versus Pictorial M em ory ........................................................................... 36
Recognition and Recall .......................................................................................... 37
Verbatim versus Gist M em ory ............................................................................... 39
The Effect of Arousal on M em ory ............................................................................... 42
The Heuristics of Pictures ............................................................................................ 47
Pacing versus Presentation Speed ................................................................................ 50
How Learner Attributes A ffect W hat is Learned ......................................................... 53



iv









Cognitive Style ....................................................................................................... 53
Im pulsive-Reflective Style ..................................................................................... 54
Factors from Film and Television That M ay Affect Learning .................................... 58
Preconceived M ental Dem ands .............................................................................. 58
Exem plars and Other Form al Features .................................................................. 60
Symbol System s ..................................................................................................... 66
Evaluating the Educational Im pact of New M edia ...................................................... 69
New M edia and M em ory ....................................................................................... 70
Effect of Fast Cuts on M emory .............................................................................. 73
Presentation Speed versus Content ........................................................................ 73
M ontage ................................................................................................................. 75
Summ ary ...................................................................................................................... 77

3 M ETHODOLOGY ...................................................................................................... 81

Introduction .................................................................................................................. 81
Population .................................................................................................................... 82
Sam ple .......................................................................................................................... 82
Instrum entation ............................................................................................................ 83
Test descriptions .......................................................................................................... 88
Initial Developm ent ............................................................................................... 88
Pilot Test ................................................................................................................ 89
Test Adm inistration ............................................................................................... 90
Stim uli .................................................................................................................... 91
M ethodology .......................................................................................................... 92
Research Design ........................................................................................................... 93
Hypotheses ................................................................................................................... 95
Sum m ary ...................................................................................................................... 96

4 ANALYSIS OF THE DATA ....................................................................................... 98

Introduction .................................................................................................................. 98
Results .......................................................................................................................... 98
Summ ary .................................................................................................................... 106

5 CONCLUSIONS AND RECOM M ENDATIONS .................................................... 109

Introduction ................................................................................................................ 109
Findings ...................................................................................................................... 110
Discussion .................................................................................................................. III
Presentation Speed ............................................................................................... III
Gender .................................................................................................................. 112
Verbatim versus Gist M em ory ............................................................................. 112
Cognitive Style ..................................................................................................... 113
Interaction Between Speed and Style .................................................................. 114
Im plications ................................................................................................................ 114



v








Recom m endations ...................................................................................................... 116
Sum m ary .................................................................................................................... 120

A PPEN D ICES ................................................................................................................. 122

A ppendix A Screen Shot of M FFT-20 .................................................................... 122
Appendix B Screen Shot of Sam ple Test Q uestion ................................................ 123

REFEREN CES ................................................................................................................ 124

BIO G RA PH ICA L SK ETCH ........................................................................................... 140









































vi














LIST OF TABLES

Table pffe

1. 2 x 3 Two-W ay ANOVA (204) Subjects ....................................................................... 94

2. Multivariate analysis of variance (204) Subjects ........................................................... 94

3. 2 x 3 two-way ANOVA (129) Subjects ......................................................................... 94

4. Source table of analysis of score variance by presentation speed ................................. 99

5. Score means and standard deviation for presentation speed .......................................... 99

6. Source table of analysis of variance by presentation speed and gender ...................... 100

7. Score means and standard deviation of scores for gender ........................................... 100

8. Source table of analysis of variance for speed versus gender ...................................... 101

9. Score means and standard deviation for speed versus gender ..................................... 101

10. Source table of analysis of variance for verbatim score for speed ............................ 101

11. Source table of analysis of variance for gist score for speed ..................................... 101

12. Multiple comparisons between speed and gist scores ................................................ 103

13. Score means and standard deviation for gist versus verbatim test items ................... 103

14. Source table of analysis of variance by presentation speed and cognitive style ........ 103 15. Means and standard deviation for overall scores for style ......................................... 104

16. Source table of analysis of variance for verbatim and gist and cognitive style ......... 105 17. Means and standard deviation for gist and verbatim scores for style ........................ 106







vii














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

THE EFFECTS OF COGNITIVE STYLE AND GENDER ON VERBATIM AND GIST MEMORY FOR RAPIDLY-PRESENTED MONTAGE VIDEO By

Robert F. Kenny

May, 2002

Chair: Lee J. Mullally, Ph.D.
Major Department: School of Teaching and Learning


Younger students live in a media-centric world. Researchers have shown that today's youth spend more time watching television and movies than most any other leisure-time activity. Moreover, the presentation speed of passages on commercial television has increased significantly in the past fifty years. Researchers have shown that viewers automatically learn to cope with symbolic pictorial presentation methods through repeated exposure to television and its visual patterns. Increased exposure to rapid sequence and presentation speed brought on by rapidly-presented montage found in television programs aimed at youth raises the question as to whether these individuals remember more from this type of presentation techniques than can their adult counterparts. Recent advancements in technology have changed the way viewers look at and interpret video media. Rapidly presented video media place additional attention




viii














demands on their viewers. Yet these viewers have been found to be able to perceive and remember content from these messages.

The purpose of the study was to investigate whether youthful viewers are able to perceive video images that are presented very rapidly and then immediately remember verbatim information and assimilate the contextual gist of the overall passages. The study also attempted to determine whether or not there are differences in the way males and females process these images. In addition, subjects were categorized as being either impulsive or reflective to determine whether cognitive style has an effect on pictorial memory.

The study revealed that memory is not significantly affected by changes in

presentation speed when test results were viewed as a composite of verbatim and gist scores added together. However, when the two scores were separated, gist memories were positively affected by increases in presentation speed. The study also revealed that cognitive style had a significant impact on pictorial memory.

















ix














CHAPTER I
INTRODUCTION


Media educators and theorists for years have been analyzing Marshall McLuhan's famous quip, the medium is the message (Meyrowitz, 1985). In some regard, McLuhan's statement may be considered a pre-emptive retort to later critics of educational media like Richard Clark (1983) who claimed that media are "mere vehicles that deliver instruction but do not influence achievement any more that the truck that delivers groceries causes changes in nutrition" (p. 445). Communications theorists like Walter Ong (1982) not only agreed with McLuhan, but also extended the meaning of his message to imply that the types of media people use define the way they think. Ong's notions bring to mind possible questions as to whether today's media-centric youth perceive differently than previous generations, with implications as to the kinds of mediated instructional strategies that might be successful in motivating them to learn as well as providing perceptual stimuli for recognition and recall.

The predominant types of production techniques used in today's electronic media are changing. Encouraged by the successes of early pioneers of rapidly presented music video montages on networks like MTV, VH 1, and Nickelodeon, and helped by rapid advances in technology, today's television producers regularly communicate very complex messages using fast-cuts and video montage (Stephens, 1996). Significant increases in viewership of these programs may be providing fertile opportunities for today's youth to practice their ability to receive and assimilate this fast-paced message



I





2


delivery approach. According to Seward-Barry (1977), sleeping has become the only activity that occupies children's time more than watching television or playing video games.

For these reasons, a change may be warranted in the way in which educators view the role video media play in initiating learning opportunities and conveying more complex or abstract thought. This thinking appears to be incompatible with earlier communications theorists like Edgar Dale (1969), who felt that television finished somewhere in the middle of a twelve-point influence scale in its ability to convey contextual ideas. New digital media that employ visual imagery "supplemented by quick motion, sound, and computer editing" (Stephens, 1996, p. 69) may be ready to take on the same predominance as a communication medium in only fifty years that it took Guttenberg's great invention five hundred years to accomplish. Corcoran (198 1) defined intelligence as a skill in a particular medium and suggested that the symbolic codes used in that medium that serve communication purposes and are internalized by a receiver become an authentic tool of thought. Research studies in which technology (i.e., media) was successfully used as cognitive mind tools appear to back up this thinking (Hokanson, 2000; Jonassen, 1996).

Statement of the Problem

Researchers and educators looking into the intrinsic instructional value of video media have presented conflicting views on the role visual perception plays in attention, motivation, and recall. There have been several studies that have looked into the potential affect mediated coding systems have on cognition (Davis, Scott, Pair, Hodges, & Oliverio, 1999; Nugent, 1982; Seidman, 198 1; Walma van der Molen & Van der Voort,





3


2000). While many theorists have looked into the use of video, most of them had little good to say about its ability to bring anything new to the table with regards to using it as a communicative medium (Berlo, 1960; Calvert & Scott, 1989; Ide, 1974; Kozma, 1986). This may have been due to the limitations imposed by the technology in use at the time. Recent technological advances in commercial television production techniques allow today's producers to readily integrate fast-cuts and montage and provide new tools to more easily communicate complex thought using a pictorial narrative structure. A correctly constructed rapid-cut montage passage has been found to add clarity because of the phenomenon that the interpretive whole of a montage segment literally communicates more than the sum of its parts (Hitchon, Druckler, & Thorson, 1994; Stephens, 1996). In other words, it is the composite whole of all the visual images in a passage or segment when considered all at once that gives it extended meaning. In addition, newer editing techniques have evolved that emphasize the perceptual continuity of a rapidly presented image-based narrative structure, rather than the classical point of view of editing that stressed the importance of applying strict rules in order to obtain slow and smooth transitions between successive shots (d'Ydewalle & Vanderbeeken, 1990).

One of the specific problems being addressed in the current study is to update the thinking with regards to using video as an instructional medium in light of these new technological developments. A review of the literature of the past twenty years has already yielded some studies into the impact of rapid video editing in an educational setting (Stephens, 1996; Wetzel, Radtke, & Stem, 1994). However, more recent advances in technology that have superseded those studies and an increased usage of rapid-cuts and montage production techniques in commercial television programming have created the






4


need to update any earlier research that does exist. Lang has conducted more recent studies that have looked into the effectiveness of fast-cuts (Lang, 1994; Lang, 1996; Lang & Basil, 1998; Lang, Bolls, Potter, & Kawahara, 1999; Lang, Zhou, Schwartz, Bolis, & Potter, 2000), but hers deal with television in a casual viewing environment. The present study was designed to look at the effect of these rapidly presented visual passages in an educational setting. It is hoped that a new look at rapid visual processing brought on by this study might lay the groundwork for educational media producers to update their thinking about rapidly presented video montage by providing a research basis that appears to be lacking currently. Specifically, the questions that are being reviewed are:

1) Is it possible to remember content solely from rapidly paced visual montage
that is not supplemented with some form of verbal narrative?

2) Is there a difference between gist and verbatim memories for rapidly presented
videos?

3) How do individual characteristics such as gender and leaming/cognitive styles
affect a student's ability to process fast-cuts/montage video presentations in a
classroom setting?

Need for the S!qdy

As has been stated already, most of the previous studies into the impact of fastcuts/montage have concentrated on commercial television viewing (Bryant & Rockwell, 1991; Lang & Basil, 1998; Lang et al., 1999; Lang et al., 2000; Zillman, 1991). There has been one attempt to look at the use of fast cuts/montage in an instructional setting (Keller, 1976). However, that investigator looked at viewers as a collective whole with little regard for individual differences in personalities and capabilities such as cognitive style. Further, this study is more than twenty-five years old and its subjects were college-level students enrolled in communications classes, rather than school-aged students studied in





5


their formative educational environments. The current study attempts to resolve some of these contradictions to see the effect fast video presentations have on immediate memories for their content.

Educational media researchers have had to contend with a conflicting view that holds that the relatively fast presentation speed of television programs creates an environment that may be detrimental to attention and recall (McCollum & Bryant, 1999; Neuman, 1976). To the contrary, Intraub's (1999) studies into conceptual masking have shown that humans are able to recognize and recall pictorial presentations when a minimal amount of lag time separates individual images. In addition, some of the more recent successes with children's programs have shown educational successes that are due to their rapid and rhythmical presentation speed (Anderson, Alwit, Lorch, & Levin, 1979; Anderson and Bryant, 1983; Anderson & Collins, 1988; Pearl, 1982). Still others have proposed that presentation speed in instructional media may actually add interest to otherwise uninspiring content (Canelos, 1986; Edgar, 1997; Hawkins, Pingree, Bruce, & Tapper, 1997; Hill & Lang, 1993).

Previous examinations into the effects of casual television viewing have not

translated too well into an educational setting (Salomon, 1994). As Kozma (1986) stated, "viewership should not be confused with learning" (p. 14). However, once one delves deeper into these studies, four possible reasons for this lack of transfer become apparent. First, the reputation that commercial television has held for being nothing more than an entertainment device has hurt its standing in educational circles and has caused detrimental pre-conceived notions about how viewers are to be properly introduced to televised content (Wetzel et al., 1994). Second, many of the schemes used in commercial





6


television are considered by some to be nothing more than prototypical, trite, and overlyfamiliar formulas that reduce attention and concentration because they have been overlearned (Anderson et al., 1979; Langer & Imber, 1979). This schema also "leads one to revert to a mindless routine in which the material is ignored or receives a low level of attention" (Wetzel et al., p. 169). This alleged over-familiarity with format has lead many educators to believe that viewers will have difficulty responding appropriately to educational televised presentations, unless some form of outside instructional intervention is also inserted (Wetzel et al.). The third reason is that many of the studies into casual viewing tended to lump all viewers into a single category (Lang and Basil, 1998; Lang et al., 1999; Lang et al., 2000; McLuhan, 1964; Neuman, 1976; Tyner, 1998; Walma van der Molen & Van der Voort, 2000). Classical instructional models tend to validate the value of classifying learners by their individual differences (Gentry, 1998; Joyce, Weil, & Calhoun, 2000).

The fourth and possibly most important reason for the lack of correlation between studies of casual viewing and those performed in an educational setting is that the leading environment is thought to present a different set of circumstances--a unique view as it were. This is based on the importance placed on the medium to be evaluated as to its unique ability to bring about some type of alteration of intellectual behavior or thinking process. Although many of the symbolic (i.e., intellectual) combinations found in nonverbal endeavors such as music, painting, and dance can be displayed directly on television, there has been some question as to the extent to which transformations in the thinking process are actually created by viewing the medium, and whether any changes that might occur are of any significance (Ide, 1974). Television has not been given credit





7


for yielding any new intellectual construct of its own. Previous studies into the value of using television as a medium for intellectual change have demonstrated mostly negative progress in that the results showed television did not actually interfere with learning, or that it was not less effective than other forms of media (Thompson, Simonson, & Hargrave, 1996; Wetzel et al. 1994). The current study attempts to clarify where rapidly presented montage might fit into the overall instructional scheme.

Previous attempts at investigating rapid video and montage presentation were

limited by some incorrect assumptions as well as limitations on technology. For example, Intraub's (1999) experiments that demonstrated weaknesses in an individual's ability to understand and remember briefly-glimpsed images dealt with pictures that were not or only very loosely related. The proper use of montage implies that the pictures included are at least conceptually related. Intraub indicated that subjects might be able to hold more than one picture at a time in a conceptual buffer, so long as the "series was not too long" (p. 57). Keller (1976) tried to fix this shortcoming in his dissertation but ran into some serious technical flaws in his overall design. Intraub even admitted that his experiments might have been more successful if they included pictures that were related to one another. His intuition may have been correct. It appears that humans may have the ability to construct meaning from these types of presentations through the use of interpretive coding (i.e., the process by which meanings are put together from specific parts of visual communications). In describing their research in teaching Native Americans how to use film to communicate meaning, Worth and Adair (1997) noted that the process of coding has been neglected in the study of most of the fine arts, including film. Their comments alluded to the fact that theform of a medium might be what carries





8


its meaning. In an earlier attempt to extend this notion to television, Pearl (1982) discussed the relationship between form and content and admitted that it is theform (that is, the way it uses verbal and linguistic codes) --not the content-- of television that makes it a unique communicative medium. However, she also cautioned that form and content cannot always be distinguished "no more than grammar and meaning in any verbal language can" (p. 24). She went on to say that some forms are unique to a particular medium and apply syntactical meaning only in the context of that medium. For example, slow motion is not real and its meaning must be learned. Other researchers have shown that these unique formats, once learned, generally become used by people in their own thinking (Barnett, 2000). That is perhaps why, for example, when one applies slow motion to a video message, it generally carries some contextual or emotional connotation each time it is used. The current study tries to apply the same logic to fast cuts and video montage to see if this presentation format can be interpreted in such a way as to portray some implied meanings of its own.

Previous research into using increased or compressed presentation speed and movement in multimedia has been the subject of controversy in the literature. Some studies of commercial broadcast video have shown that the relatively rapid presentation speed of programs may have a deleterious effect on recall (Alwitt, Anderson, Lorch, & Levin, 1980; McCollum & Bryant, 1999). Other research has shown that the instructional value of video is aided by a systematic presentation speed of information as a presentation strategy (Comstock, Chafee, Katzman, McCombs, & Roberts, 1978). Still others have shown that presentation speed and rhythmicity in leisure-time media can actually heighten enjoyment, enhance motivation, and can "play an important part in





9


determining the affective or emotional response of message receivers" (Seidman, 198 1. p. 49). Lang et al. (1999) and Lang et al. (2000) looked into presentation speed as its own construct, comparing/combining it to/with arousing content. They discussed the effect of adding interesting and arousing content as having a positive effect on cortical arousal. and therefore, recall and recognition. Lang et al. (2000) also alluded to future research that should continue to probe the shape of the relationship between presentation rate and recall and test even faster rates of edits to determine whether there is a point at which memory begins to decline. They suggested that producers who want their messages to be remembered should create arousing messages that are presented slowly or at medium speed, or calm messages that are presented at medium or rapid speed. They concluded that producers should not create messages that are either calm and slow or arousing and fast.

The effect of integrating increased message presentation speed directly into

instructional messages in an educational environment whose content might be considered by some students as less than arousing will be examined in the current study. It should also be noted that Lang et al. (2000) considered cuts to beAst if they changed at a top rate of eleven to twelve per thirt-y-second segment (i.e., one every 2-3 ) seconds). The current study looks at a top rate of one cut every one-third to one-half of a second. or roughly ten times as fast. The slow messages will run roughly equivalent to Lang's top rate.

It should also be noted that the term cuts. as it was used in connection with

commercial television research, differs from what how it was used in the current study. in those studies, a cut might simply mean a wipe or a swipe, or a change from one camera





10


angle to another. The content may not change at all, only the viewer's perspective. In the current study, a cut always signifies a change in content. One picture or image was replaced with an entirely new one.

Lang et al. (2000) indicated that under very broad interpretation, the speed of message presentation of motion and edits/cuts can be considered a form of cueing. Downs (1989) also referred to presentation speed in a similar way, referring to it as a message's domain "attribute" (p. 3). Salomon (1979) acknowledged that media attributes are those that are "within the mediated stimulus, possibly shared to some extent with other forms of [sic] media' and make "the presented information more comprehensible or better memorized by learners of particular characteristics" (pp. 5-6). Under Salomon's definition, symbols include "most objects, marks, events, models, or pictures" (p. 29). It is assumed in this study that the rhythmic patterns afforded by fast-cuts are an event. Where the current study varies from previous research is that it takes the interrogation of symbols and attributes to another level. The current study aims to show that rapid presentation speed (also referred to as fast-cuts) may be considered an invaluable communicative attribute of media and is, therefore, capable of being studied separately to discover its contribution to learning. This study aims to determine to what extent that rapid presentation speed of video images either aids or interferes with recognition and recall (specifically, gist and verbatim memories), taking into consideration the changes that appear to be taking place in an ever-increasing media-centric society. Even if fast, moderate, and slowly presented visual stimuli are all shown to be equivalent in their ability to stimulate recognition and recall, educational research might benefit from discovering whether visual stimuli presented at an increased delivery pace can provide








similar instructional outcomes in a more efficient and stimulating manner --something found by some researchers to be a more practical outcome measurement for using video media than achievement alone (Kini, 1994; Peck, 1987).

Background of the SqLdy

Younger students live in a media-centric world. Researchers have shown that

youth today spend more time watching television and movies than most any other leisuretime activity (Pearl, 1982). In addition, the presentation speed of passages on commercial television has increased significantly in the past fifty years (Stephens, 1996). Researchers have shown that viewers automatically learn to cope with symbolic presentation methods through repeated exposure to television and visual patterns (Abelman, 1995; Bargh, 1988; Carr, 1982). Because of their increased exposure to rapid sequence and presentation speed brought on by fast-cuts/montage found in television programs aimed at youth, it may also be assumed that these individuals might be able to comprehend these messages on a much wider scale than can their adult counterparts. One cannot continue to assume that exposure to video, regardless of presentation speed, is simply a passive viewing activity. These advancements in media technology are changing the way viewers look at and interpret video media. In addition, the widespread availability of production techniques provides easy access to capabilities that allow people to use video media to easily create their own content. It has been widely shown that these acquisition (i.e., production) opportunities also increase exponentially one's ability to comprehend content delivered in like form (Tyner, 1998).





12


PgWse of the SgLdy

This study was designed to determine if there are differences in the way visual montage messages that vary only in their presentation speed (i.e., fast, medium, or slow) are perceived and immediately remembered in an educational setting. Recent reports found in the literature, (Brainerd & Reyna, 1990; Brainerd & Gordon, 1994; Reyna & Kiernan, 1994) have suggested that researchers have been able to successfully parse memory into verbatim (i.e., precise/literal) memory for specific details and gist memory (i.e., contextual or contextual remembrances similar to that which is assessed in reading comprehension tests). This study looks at immediate memory as a whole and separately, using these same categorizations. The educational setting was determined to be several ninth grade classrooms that were categorized by gender and also ftu-ther delineated by the subjects' individual cognitive/learning style. The overall sample set for this study came ftom that population group. For purposes of this study the identification of cognitive style was limited to the reflective-impulsive scale, as originally developed by Jerome Kagan (1965; 1966), and later refined and re-catalogued by Cairns and Cammock (1984). Identical videos that varied only in their speed of message delivery were presented to three randomly assigned groups of students. The procedural requirements for implementing the cognitive style test instrument indicated that the analysis was to be performed on a smaller subset of the sample base to see if the variance in delivery speed also affects gist and verbatim memory, and if there might be any interaction with their prospective cognitive styles.

Previously, presentation rate had been found to affect immediate memory in examinations of viewers in a commercial viewing environment that compared several





13

contrasting theories about memory and cognition (Lang et al., 1999; Lang et al., 2000). These investigations concluded that individuals have limited capacity for processing information and indicate that a rapid influx of visual images maintains an individual's orientation reflex and stimulates cortical arousal. Unlike other forms of arousal, cortical arousal has been determined to maintain a person's ability to pay attention (Pearl, 1982; Reeves, Thorson & Schleuder, 1986; Zillman, 1991). The earlier studies also alluded to conflicting findings about the impact of the content on attenuation and arousal (Anderson, 1986; Archer, 1965; Basil, 1994), maintaining that arousing content combined with too fast a pace/message delivery would normally result in loss of retention and recall. Unlike these previous studies that included occasional highly arousing content, the current one used an educational topic that might not be considered by many students as arousing (i.e., American history) in an authentic classroom setting.

The current study also attempted to improve upon some of the various

shortcomings found in the one other earlier study (Keller, 1976) that used very similar but out-dated methodologies. In his doctoral dissertation, Keller used a self-created kinestasis film (i.e., rapidly cut video montage) in an attempt to define what effects on recall are realized by speeding up transmission or information to less than one second per fi-ame. Standards previous to Keller's study deemed a presentation to be rapid if cuts occurred every five to ten seconds. Keller also endeavored to derive empirical data to determine how well the kinestatic process worked on recall and how efficient visually compressed presentations might be for instruction, trying to find out whether the kinestasis technique was more than a simple artistic endeavor. However, by his own admission, Keller's research came up short because some of his methods may have been




14


lacking in rigor. For example, he utilized random pictures as opposed to those with similar contextual content. As schemata have been long shown to be of importance to leading (Bartlett, 1932) studies looking into recognition and recall need to use a single context. Additionally, most of the problems Keller ran into appeared to be related to production and processing techniques that reflected the state of the art of video production at the time (Keller). His difficulties ranged from dissimilar content, picture size that was too small, and brightness that caused a deviation in results based on the seat location of his subjects. Due to technological shortcomings, he utilized what are known in the industry as wipe-onsldowns as his motion technique. Wipes are production techniques utilized in television broadcasting in which one picture is replaced by another that is gradually wiped across the screen. An edit or cut, on the other hand, is where one picture immediately and completely replaces another. In this technique, the succeeding picture frame partially masked, or interrupted the previous one and vice versa, causing a potential confound. There has been some reference in the literature (Goodglass, 1971) that indicates that visual masking introduces a new, unnecessary variable into studies on visual processing. Keller also had problems in his test phase because he could not cause the pictures to replicate the movements that occurred in those same pictures when he presented them originally. He admitted that this approach was flawed in that the whole picture could not be viewed until the end of the procedure, causing a degradation of recognition and recall. The current study attempts to clarify some of those issues. Keller referred to the need to "find another motion procedure" (p. 90). The current study used a commercially-made video montage that refers entirely to the same subject matter, for





15

which permission has been granted by the copyright holder to modify for use in this study (R. Wright, personal communication, March 21, 2001).

Finally, Keller's (1976) subjects were college level communications students who were segregated by grade point average. He identified the need to find a more rigorous randomization variable than GPA, noting that there was a significant interaction effect among the subjects, who accounted for almost 24% of variance in picture memory recognition, while the treatment effect only accounted for 6%. The current study attempted to look at an additional classification method (i.e., cognitive style) that had been identified and validated in research subsequent to Keller's study (Green, 1985; Kagan, 1965; Okun, Callistus, & Knoblock, 1979; Ridberg, Parke, & Hetherington, 1970; Salkind & Wright, 1977; Witkin & Goodenough, 1981).

Research Questions

Following are research questions raised in this study:

I Is there a difference between the amount remembered in terms of verbatim
(specific details) and gist (contextual content) memory when changes in
presentation speed occur?

2. Are there gender differences in one's ability to remember information presented
in rapid video presentations?

3. Are verbatim and gist memory affected by a learner's cognitive learning style? These specific questions are intended to lead to an over-riding question as to whether if the amount of information retained through fast-cuts/montage is found to be equivalent to more moderately paced instruction, is a faster delivery more educationally efficient, as determined by the fact that the same amount of material can be successfully presented in a shorter time intervaI9





16


Delimitations of the Stud

This study was conducted with certain limitations. In previous normative data validation and reliability studies for the cognitive style testing instrument being used in this study, subjects were limited to ninth grade students from a specific geographical area. Only those students who agreed to participate and whose parents signed a consent form are included in the study. The study only looked into the amount of information that could be immediately remembered from the video passages. While immediate memory has been deemed an important prerequisite to learning, this study makes no representation with regards to whether the material is subsequently learned (i.e., retained for longer periods of time, or able to be utilized in subsequent tasks) by the subjects. The fact that learning may or not actually take place implies that additional pedagogical and cognitive techniques are subsequently employed, and is beyond the scope of this study.

Limitations of the Study

This study contains limitations and assumptions that may affect the

generalizability of the results. The study is only generalizable to the population from which the sample is taken. In this case, subjects came from one high school of a mixed ethnographic and demographic make-up in North Central Florida. While it is understood that students who have previous knowledge of the topic (i.e., American history) might be more likely to recognize and recall the pictorial representations in the videos, it is also assumed that this knowledge base was most likely equally distributed throughout the sample set because subjects were selected at random. Also, as the students were randomly assigned to the three presentation groups, it was determined that any previous history knowledge would not materially affect the results.





17


It also should be noted that this study has limitations with regards to its ability to predict actual learning. The study was designed to only define certain rules about presentation speed of an initial presentation that may become one component of an overall instructional strategy. By itself, the study does not attempt to make any predictions or value judgments about those strategies. Rather, it simply attempts to support the validity of fast-paced video segments that might become building blocks in sound pedagogical patterns, as delineated in instructional models that might be subsequently proposed by future educational research.

Definition of Terms

Following are specific definitions of terms used in this study:

Cognitive Lqle is the stable ways in which persons differ in perception and

encoding of information (Wittrock, 1979). Specifically, it is Kagan's scale (1966) that was later modified by Cairns and Cammock (1984) that is used as a basis for the current study.

Concep!LW masking is concept that describes the phenomenon that occurs when a new visual image is introduced into temporary memory and thereby occludes the previous one (Intraub, 1999).

Gist Memory has been successfully used as a treatment effect in several studies looking into prose memory (Biederman, Rabinowitz, Glass, & Stacy, 1974; Brainerd & Gordon, 1994; Cowan, 1998; Reyna & Kiernan, 1994; Voss, Tyler, & Bisanz, 1982). Gist memory is one of two types of determinations (the other being verbatim memory) as to the amount and quality of immediate memory. In previous research gist determination required subjects to try to identify the main idea in a passage or correlate among several





18


scenes. The objective of assessing gist memory in this study was to determine whether the subjects comprehend the basic contextual themes presented on the video and to compare it to subjects' ability to immediately remember specific (i.e., verbatim) details from the video presentation.

Kinestasis is a term coined by Charles Braverman (1969) that is derived from combining the Greek words for moving and static. The term refers to a film or video process that presents still and/or moving images at speeds that create the illusion that they are moving in a continuous flow. The treatment video used in this study is a video copy of a kinestasis film developed by Braverman, and whose copyrights are owned by Pyramid Media.

The term media includes films, television, and compressed motion pictures found in digital multimedia. Previous researchers into presentation speed (Wetzel et al., 1994) and compressed motion pictures (Arnold, 1996) delved into one or more of these types of media and have referred to them as virtually interchangeable in meaning. This study also refers to the term new media. This term connotes an inference about the convergence (the merging of television, computers, and the telephone) that is taking place. The current study uses screen media and iideo media interchangeably when referring to its capabilities to present a pictorial story. It uses new media to mean traditional media that are being technologically advanced due to convergence.

Minimal unit of analysis is that product of semiotic analysis which is the smallest interchangeable element that still retains the basic properties of the whole symbol system but cannot be further subdivided without losing those properties (Corcoran, 198 1). Hill (198 1) referred to this concept as a syMbol, or the basic unitary element of intellectual





19


activity. In the assessment of reading comprehension, a word is often associated with the unit of analysis (Murdock, 1982). In this study, individual picture frames are considered as being the minimal unit of analysis. These may be equated to a paragraph (Corcoran) for purposes of comparing the results to reading comprehension-type assessments.

Orienting ReWonse (OR) is the automatic response to a stimulus that helps an individual focus attention on that stimulus and is often captured through psychometric means (i.e., an attention. blink, or a look (Lang & Basil, 1998; Lang et al., 1999; Reeves et al., 1986; Thorson & Lang, 1992). An OR can also cause an overt arousal response that may be detrimental to recall. In this study the orienting response is considered to be a potential confound and is dealt with in the methodology of the study. Subjects will be provided a short preview of the video presented at the same speed in order for them to get used to the format.

Presentation sWed is the rate of scene changes or edits of a video media

presentation and is the operative term utilized in this study. In many studies used to support the premises in this study, pacing has been the operative term used to refer to the speed of message presentation (Corcoran, 1981; Hill & Lang, 1993; Lang et al., 1999; Lang et al., 2000; McCollum & Bryant, 1999; Reeves et al., 1986; Schale, 1971, December; Wagely, 1978; Zillman, 199 1). The term pacing also carries certain preconceived connotations in educational circles that have nothing to do with the premise of the current study. In order to avoid confusion, all references to pacing were replaced in the current study with the terms presentation speed, or speed of message presentation.

Screen media are that which use controlled exposure of sequenced images on a screen (either on a video or movie screen), often in fixed relationships to speech, music,





20


and other sound (Corcoran, 1981). Much of the literature presented in Chapter 2 refers to screen media. For purposes of this study screen media and video media are considered to be identical.

Verbatim/rote memory refers to subjects being able to identify whether specific objects within a scene, or overall scenes occurred in the video that has just been seen (Brainerd & Gordon, 1994; Reyna & Kiernan, 1994). In some studies, verbatim memory also has been referred to as object recognition (Tse, Vegh, Marchionini, & Shneiderman, 1999). For purposes of this study, verbatim memory is assessed positively and falsely. Subjects were asked to select an image that was presented in the video display (positive identification) from a group of pictures immediately after presentation of the test video. Subjects were also are asked to decide whether any picture or series of pictures in a group actually appeared in the video (false identification).

Sunimpa

Meyrowitz (1985) posited the idea that new media, through its ability to merge many formerly distinct knowledge situations, appear to be "breaking down the boundaries among various disciplines, opening new dialogues, and fostering the development of cross-disciplinary areas of study" (p. 327). He also speculated that these new kinds of electronic media may be "introducing our children to a different way of thinking that involves the integration of multiple variables and overlapping lines of simultaneous actions" (p. 326). TEs assertion proposes, among other things, that electronic video media have already greatly reduced the influence that time and location used to have on what people know. It is not, therefore, unreasonable to agree with Stephen's (1996) proposal that a transition may be occurring in the way the youth of





21


today think. This appears to be at odds with the linear thinking processes associated with print media. The current study, through the questions posed in the current chapter and a review of the literature found in the next, attempts to find out how significant these changes in perception of new media are.














CHAPTER 2
REVIEW OF THE RELATED LITERATURE Perception and Knowledg

When humans learn from the world around them, they receive stimuli

(information) via one or more of their five senses: audition (hearing), gustation (taste), olfaction (smell), kinesis (touch), and vision (sight) (Murch, 1973). Processing received stimuli can be automatic or be the result of some literal attempt to react to it. For example, when one smells food, it may make him or her feel hungry. This may happen with or without that person actually realizing what is happening.

Besides being the catalyst for automatic processing of bodily functions,

perception has also been considered by educational researchers as a fundamental building block for knowledge and leading. Murch (1973) posited that humans must be able to perceive something before they are able to learn about it. This information may be firsthand, or via an artificial artifact (also known as a medium). Gibson (1969) defined perception as a process of receiving stimuli and consciously reacting to them by assigning some meaning. Accordingly, humans tend to react to perceptions individually, based on previous personal experiences, making this experience very subjective. Winn (1982) described the process of learning as a taxonomy of three procedural steps. After a person first receives information to be stored, and later when he or she attempts to retrieve it from memory, a long process of transformation, abstractions, and elaborations takes place. The first step is perception, generally considered an automatic operation, in




22





23


which features are integrated into complete visual displays. Assimilation is the next level of complexity and involves the recognizing and integrating of information into schemata for later recall. Analogy is similar to assimilation but is an even more complex process that involves the temporary assimilation of new information into abstract schemata that embody other concepts that are peripherally related to it. Explained in this way, Winn alluded to perception as a necessary but insufficient step in the whole learning process.

Bruning, Shraw, and Ronning, (1999) posited that the body holds perceptions in a series of registers called memory so that perceptual analysis can occur before that information is lost. When all is working well, perception allows humans to detect the incoming stimuli and allocate attention to them. Through pattern recognition, individuals may recognize a stimulus because of their previous experiences. However, on occasion, pattern recognition may be limited, thereby preventing any learning. Marr (1982) proposed a structural approach to the relationship between perception and intellectual endeavors. He imagined that knowledge is stored in the brain as a set of statements, templates, prototypes, or models of distinctive features of a particular object or class of objects that has been perceived. Bruning et al. (1999) appeared to agree, stating that using specific, structured descriptors as an approach to recall and recognize patterns may be very helpful for teaching because it suggests that perception can be a guided activity. Teachers can instruct students "the proper knowledge needed for an accurate structural description" (p. 27). Although the visual field may be partially obstructed, the mind is still able to make deten-ninations about what it sees, using clues and mental mappings (visual patterns) from previous experiences (Seward-Barry, 1997). Gestalt psychologists refer to this process as visual intelligence and base their theories on this emphasis of the





24


mind being able to discriminate parts to the whole and make other spatial determinations that are described further later in this chapter. But first, to understand more fully the Gestaltist view of visual intelligence and how it relates to this study, it might be best to consider some of the literature concerning the neurological process involved with vision. Neurology of Vision

According to Murch (1973), biologists define vision as a process where the eye sees the world through a biological process of photic radiation, an energy source that stimulates the eye. The process begins as reflected light that bounces off objects in the environment. This optic array is focused by the cornea and lens onto the visual field on one's retina that lines the back of the eye (Seward-Barry, 1997). These images are then transmitted to the brain through various routes, one being the cortex/limbic system. This system provides stimulus to the brain, including the amygdala: the sub-cortical region within the temporal lobe (LeDoux, 1991). According to Seward-Barry, researchers have identified four parallel systems involved in the different attributes of vision: one each for motion and color, and two for form. Goldstein (1989) points out that "perception is based not on direct contact with the environment, but on the brain's contact with electrical signals the represent the environment. We can think of these electrical signals as forming a code that signals various properties of the environment to the brain" (p. 50). In other words, perception as a biological ftinction may be described as being based on a coding system that takes in the visual signal and is interpreted by the brain. LeDoux (1991) viewed this interpretation process as a dual operation. He proposed that sensory perceptual processing takes two routes. The first travels along the amygdala and readies the body to react, even before it recognizes the need to do so. The second route is through





25


the neo-cortex where the signal can be analyzed and then sent to the amygdala and emotional response added after cognition (Seward-Barry). This dual processing theory provides an insight as to why visual perception takes on such a personal meaning. It also introduces the idea that will be further developed later in this chapter that there can be two very different emotional reactions to perception: one helpful to cognition, and one that can be detrimental.

Visual Perception Theories

For some theorists, seeing is more than a simple biological function. According to Seward-Barry (1997), visual perception is actually taking the visual image, combining it with other data from the other senses, and synthesizing it with previous experiences. Perceptual psychologist J. J. Gibson (1979) made a distinction between the visualfield, the image that appears on the retina, and a mental creation of what comprises our world, giving further credence to the fact that to an individual, reality is very interpretive and subjective. To constructivist theorists, visual reality is simply a "map-like image, the end product of a process that begins with light refraction in the environment and ends in the intricate and complex dynamics of the mind" (Seward-Barry, p. 15). Viewed this way, perception is what intercepts an outside stimulus for a person to create an individual reality. When video producers present their images in unique ways (such as utilizing rapidly presented montage) they simulate this mapping process and create opportunities for viewers to construct their own views and interpretations.

Gestalt theory, as outlined by Ellis (1938) and later by Wertheimer (1959), shares some common beliefs with constructivists. However, this theory proposes a more consistent view of perception among individuals through the concept of idea grouping





26


(Seward-Barry, 1997). Characteristics of stimuli cause humans to structure and/or interpret a visual field in certain ways or groupings. Wertheimer noted the following primary factors in deten-nining these groups:

Proximity Humans tend to group visual elements together according their spatial nearness.

Similarity Humans tend to group together items similar in some aspects (i.e., their design, size, shape, textures, etc.).

Simplicity Visual items are grouped together by humans into simpler figures, according to their symmetry, regularity, and/or smoothness. Closure Humans tend to complete on their own an incomplete picture or other visual entity.


Wertheimer was particularly concerned with problem solving. While problem solving per se is outside the scope of the current study, these ideas form an important basis for some of the assumptions about the affect that rapidly presented video montage may have on memory and leading. Wertheimer felt that the essence of problem solving behavior is one's ability to be able to see the overall structure of a problem through the use of one or more of the above classifications which all allude to the ability to fill in the gaps, incongruities, or disturbances that prevent one from seeing these overall structures and use an innate need to do so as an important stimulus for learning. According to Ellis (1938), other gestalt thinkers like Kofka and K6hler agreed with Wertheimer but extended these ideas to looking at visual patterns. The laws of visual organization found in these schools of gestalt thinking (i.e., proximity, similarity, simplicity, and, closure) form a strong theoretical basis for visual media like film, video, and television to be able to create its own unique sense of reality (Seward-Barry). The laws of visual organization





27


guide the senses to transform individual visual inputs into composite visual units made up of real and imagined stimuli, forcing one to make inferences about them (Murch, 1973).

These views of perception lend credence to a belief that a standardized or

continuous view of the perceptual process can be formulated and forms the basis to the theoretical approaches used in this study as they relate to an individual's ability to remember and aggregate more complex meaning from video montage. Studies by gestalt psychologists Richard Gregory and Sernir Zeki into concepts, such as phi phenomenon (processing movement as individual sequences separated by brief instances) and apparent movement (the impression of movement from two stationary stimuli), reveal that there might be two distinct systems of visual perception in the brain (Seward-Barry, 1997). They also show how specialized areas of the visual cortex work together to create a unified perception. One is responsible for perceiving movement between brief flashes and the other takes care of perceiving movement between long flashes. These two concepts provide further credence to the idea that the visual whole is made up of several independent parts which, taken together, can furnish additional meaning, as afforded by fast-cut montage found in new media technologies.

The visual process, as efficient as it appears to be, is not perfect (Seward-Barry, 1997). Much of gestalt theory is based on the law ofpragnanz which stipulates that psychological organization will always be good (i.e., simple, regular, symmetrical) as the prevailing conditions will allow" (Koffka, 193 5/1963, p. I 10). It is also the early twentieth century counterpart to Aristotle's concept of common sense that posited that "efficiency is achieved through simplicity, regularity, and symmetry" (Seward-Barry, 1997, p. 47). These same theoretical concepts (proximity, similarity, simplicity, and,





28


especially, closure) explain how humans can often be visually fooled by slights of hand or other artistic manipulations of visual stimuli, how they are able to see paintings differently, or be taken in by virtual reality. The mind tends to close off partially completed geometrical patterns, and/or create its own sense of reality based on what it thinks it sees (Seward-Barry). These same four laws also explain how montage video provides the continuity needed to be able to tell a story visually. Fast Seein

Evidence points to the ability of humans to be able to process visual patterns very rapidly (Coltheart, 1999; Tovee, 1998). Theory suggests that impulses processed in the visual cortex are able to process almost 87% of perceived information within 400 milliseconds. According to Tovee, clinical researchers also found that this operation is not linear. In fact, the majority of infon-nation (around 67%) is actually available within the first 20-50 milliseconds of the initial "spike train" (p. 143). The investigation of what is understood from rapidly presented sequences of visual stimuli began with the research of Potter and Levy, in the late nineteen sixties and Forster in the early seventies (Coltheart). Their studies suggest that most information encoded in an impulse is available to the brain to act on at the very beginning moments of reception. Forster developed a technique of rapid presentation of sequences termed RSVP (rapid serial visual presentation). Potter and Levy provided a comprehensive review of RSVP methodology, which proposed that humans possessed a natural short-term memory system (Coltheart). These researchers confirmed their studies by placing constraints on how long the visual system has to process stimulus in both forward and backward masking functions in which the masking stimulus follows very quickly the onset of the





29


initial pattern/object stimulus. Through studies on rhesus monkeys, researchers have been able to point out which cells within the brain are responsible for certain functions such as pattern recognition. Once identified, these brain cell studies were translated to human subjects, who were tested to determine how fast the brain is able to react, recognize, or otherwise discriminate between faces of individuals. Researchers found that after only a 20-millisecond delay, test subjects could recognize and discriminate face identity, indicating that the minimum amount of time that neurons need to be active in order to mediate recognition and discrimination is in the 20-30 millisecond range (Tovee). The next stage to processing this information is the brain reacting to these stimuli and initiating reactionary motor commands. Independent follow-up research, based on clinical tests performed on epileptic patients in preparation for brain surgery, indicates that this process takes somewhere in the 150-200 millisecond range (Allison, et al., 1994). These researchers confirmed that the whole process of pattern recognition and reaction takes in the neighborhood of 400-500 milliseconds, indicating that, not only can the visual system process information rapidly, it can also rapidly update and modify responses, based on previous experiences (Tovee). For example, people can rapidly and accurately reconstruct meaningful objects out of fragmentary or ambiguous evidence (Ramachandran, 1994). These findings appear to denigrate traditional assumptions about perception that awareness must mediate between stimulus and response. In fact, researchers now know that many responses are automatic and emotional, and may bypass cognitive processing all together (Seward-Barry, 1997). The fact that initial responses can precede cognitive processing has caused a renewed interest in subliminal research, which deals with subconscious message registration. However, results from research in this area





30


have been inconclusive due to the inability to date of researchers to quantitatively delineate and measure perceptual responses (Seward-Barry).

Most previous research into limits on perception and recognition of rapidly presented stimuli dealt with verbal information (Coltheart, 1999). However, Intraub (1999) worked extensively with scene recognition and memory. Citing experiments of Potter, she suggests that abstract representations of scenes are used in the interpretation of pictures and scenes. Using a technique called conceptual masking, she was able to show that the attention. demands of new and meaningful pictures interfere with one's ability to continue processing their predecessors. Perhaps, sharing the gestaltist view of closure, Coltheart further demonstrates this phenomenon, referring to it as "boundary extension" (p. 6). Even in rapidly presented stimuli, subjects indicated that they tended to remember information that was not shown but was likely to have been present outside the picture's boundaries. Further, Intraub produced experiments that showed that photographs, when shown in rapid succession, (at a rate that mimicked one's ability to scan them) were unable to be recognized by their viewers, even a few moments later. These studies seem to report limitations on memory, rather than perception, shown with Intraub's studies of a series of rapidly presented pictures that were separated by blank screens of varying lengths of time. It appears that the longer time lags between meaningful pictures permitted viewers to properly process, categorize, and consolidate their meaning. Potter argued (Intraub) that a picture is identified within about 100 milliseconds. Until the item is identified, it is "vulnerable to visual masking by a new visual event" (p. 52). Once identification is complete, however, the pictorial representation is maintained in shortterm memory for a few hundred milliseconds, long enough for it to be consolidated and





31


encoded into memory. Intraub reported that the process of stabilizing integration errors can take place when pictures are presented at a rate of about three to four pictures per second. This is roughly equivalent to the presentation rate of the pictures presented to the fast-paced group of the current study.

Encodin

Perception might have been conceived of as a totally individual operation was it not for the fact that there exists some type of standardized analytical coding thesis on which to base a more stable view of the process. In reality, humans process information by breaking perceptual units down into sub-units called symbols and icons (Neuman & Prinz, 1990). Corcoran (198 1) proposed the idea that symbols and icons comprise the very basic subset of cognition. He referred to them as "the minimal units of perceptual analysis" or "that product of serniotic analysis which is the smallest interchangeable element that still retains the basic properties of the whole symbol system but cannot be further subdivided without losing those properties" (p. 118). He identified several forms of icons and symbols, including (but not limited to) words, pictures, and auditory patterns. He also referred to rhythmic patterns and control as symbols of perception that can be translated to enhance learning but he did not limit his conceptualization to audition. He also claimed that visual rhythmicity is a form of symbolic representation that can be learned as a cueing system for attention and recall (Corcoran).

Humans process symbols and icons on more than one plane, depending on the structure of the input. Words, pictures, sounds, etc. are perceived and treated in similar, but distinct ways. Paivio (1986) proposed three different procedural methods: representational, or the direct activation of verbal (i.e., text-based) or non-verbal (i.e.,





32


pictorial) representations, referential, the activation of the verbal system by non-verbal system or vice-versa, and associative, the activation of representations within the same verbal or non-verbal system. These differences in method of processing of stimuli contrast in many ways. While researchers like Carr (1982) believe that a common semantic system exists that provides similar representations for words and pictures, others have suggested that differences in processing exist or that one form of symbol takes precedence over the others. For example, the structural organization of a perceived sequence can play an analogous part in its perceptual segmentation (Corcoran, 1981). Corcoran (198 1) also proposed that verbal and non-verbal forms of information are processed at different rates. Still others (Archer, 1965; Gummerman & Gray, 1972; Huba & Velletino, 1980; Rettenbach, 1999) posited the idea that increasing the speed in perceptual encoding might be a learned activity that varies based on age and/or sex.

Perhaps the most significant debate with regards to the role perception has on memory has taken place between those who believe in the superiority of verbal coding over non-verbal, or vice versa. In their studies comparing the two, Smith and Haviland (1972) introduced two complimentary assumptions. The influence hypothesis argued that the perceptual superiority of words is due to a subjects' opportunity to use the redundancy in words for making perceptual inferences. The unitization hypothesis proposed that words are more perceptible because there are fewer units to process in a word than a non-word. Other researchers concluded that visual coding takes precedence. Pezdek (1977) performed an experiment in which the sentence the bird is perched atop of the tree was shown repeatedly at the same time as a picture of an eagle atop of a tree. After time, the participants responded to a follow-up questionnaire that the sentence





33


actually was that an eagle is perched atop of a tree. In this case, visual perception changed the actual perceived wording of the sentence. His studies were later confirmed in additional studies performed by Gwen Nugent (1982) who also found that visual images are recalled more readily than verbal texts. Her studies have been confirmed several times by those evaluating the so-called Stroop effect where subjects incorrectly name a text message by the actual font color of the text, rather than the color spelled out by the text message (Shalev & Algom, 2000). Plass, Chun, and Mayer (1998) were less conclusive. They developed a generative theory stated that learners actively selected relevant verbal and visual information, organized the information into coherent mental representations, and then integrated with one another these newly constructed visual and verbal representations. Mayer, Bove, Bryman, Mars, and Tapangco, (1996) reinforced this idea in studies of science curricula and found that there existed a coherence between the two, so long as they complimented each other and that each was specific to their purpose. Another way to look at the coding requirements between verbal and non-verbal representations is that they are simply different from one another, with neither taking precedence. In his studies, Singer (1980) found that verbal coding is a sequential process, whereas the everyday sights and sounds of everyday life is a parallel operation that encompasses a range of events within the same time. While its strengths make verbal encoding more efficient for memory, they also limit its functionality in non-textual (i.e., non-reading) experiences.

Another comparison in the research of relative strengths or weaknesses of the various encoding methods occurs between vision and audition; with neither being the unanimous choice of researchers. Nazarro and Nazarro, (1970) tried to equate aural and





34


visual stimuli. They suggest that short-term memory is based on auditory encoding even when the stimuli are presented visually. They also found that learning (i.e., memory) is reduced if the presentation relies mostly on visual inputs alone. Further, many researchers (Cooper, 2000; Davis et al., 1999; Flannagan, 1998; Flowers, 1995; Moreno & Mayer, 2000; Sterrit, Camp, & Lipman, 1996) have shown how audio tracks can aide visual encoding, especially where the fidelity of the visual display suffers.

Perhaps the most conclusive studies about the relatively stronger effects of visual over aural stimulation have been offered by Gavriel Salomon (1979; 1994). He sites many studies where visualization was more effective in recall, and concluded that visual actions are often remembered over auditory have been "primarily due to the salience over action" (Salomon and Cohen, 1977, p. 29), especially moving images. Further, he cites some investigations in which visual tracks were successfully added to supplement the auditory. He concluded that there probably is no visual dominance in children but since actions are generally more memorable, visual information is normally more likely to be recalled. However, he also cautioned that "adding pictures to an already well-fortned auditory presentation may provide some distraction" (Salomon and Cohen, p. 29).

Researchers have shown that words and pictures are processed on opposite sides of the brain and follow different psychological laws (Ederlyi, 1985). This fact had originally deprecated most traditional ideas about subliminal communication being an effective means of communication and learning. Most of the research done in subliminal perception had been done with verbal stimuli, based on the conventional assumptions about cognition having to precede response. Erdelyi's and LeDoux's (199 1 ) more recent studies using visual stimuli alone have produced more pronounced results. Although their





35


work found that the signal is sufficient to cause an emotional response, it wasn't strong enough to reach conscious thought, and that the emotional responses are autonomous and primal, similar to those that humans, have needed for survival throughout their evolutionary process (Dixon, 198 1). This inherent automatic preparation of their response to stimuli possibly explains why humans might likely be unconsciously biased towards accepting or rejecting ideas, and occasionally acting on them. It is also where researchers believe the most significant contributions to research into subliminal message processing might lie (Seward-Barry, 1997).

Relationship of Memory and LegMilIg

Studies involving learning and memory date back more than a century (Crowder, 1976). Beginning with the Ebbinhaus' preliminary studies into brain associations during the late nineteenth century, the psychology of learning principles evolved by the mid twentieth century into Kbhler's framework of analyses that addressed three processes associated with learning and memory: stage analysis, coding analysis, and task analysis. While learning may actually encompass many more of these analyses processes (such as habituation, acquisition strategies, operant training, etc.) --all of which are significant when taken in their entirety-- they are beyond the scope of the current study. This study deals with simple recognition and comprehension assessed in stage analysis, an essential pre-condition to learning, and certain aspects of coding analysis, a subset of the memory process that has been found to be an adequate premise on which to base assumptions as to whether learning may be taking place (Estes, 1994). Estes related the ability of individuals to complete the process of recalling information from memory to their ability to classify or categorize that information in a meaningful way. Under this definition, the





36


term categorization shares some meaning with the tenn perception. However, categorizing also implies a deeper process than simply detecting that some unit of information has been perceived. It also means that other related characteristics have been captured (i.e., location, color, or shape, etc.) to apply meaningfulness to them that aides the process of recalling that infon-nation later. Under this definition, immediate memory is looked upon as an important but insufficient prerequisite for Icaming. The current study utilized a cued recall performance test in order to assess the ability of the treatment to create (and the subjects to immediately remember) an initial memory trace. Textual versus Pictorial Memo

Although much of the research into memory cited in this review deals with textual information (Crowder, 1976; Estes, 1994), a great deal of it has direct relevance to the current study. In his research into memory capacity, Sperling (1963) was able to deten-nine that his subjects were able to attend to visually perceived textual characters at a rate of about 100 per second. These same subjects were only able to name (categorize) these characters at a rate of six per second. This discovery led Sperling to assume that initial information-gathering and naming (categorizing) may not be the same process, and that the task of recalling utilizes different resources and techniques for pictorial than for textual information. Durso and Johnson (1979) found that pictures benefit more when the tactic used for elaborative processing is to simply name the picture whereas words aid more when subjects perform some task to categorize them. Gaining an understanding of these differences between the textual and pictorial memory processes has served two purposes in the current study. First, it helped to emphasize the role pictorial perception can play in the learning process. Second, it reinforced the need to assess pictorial memory





37


with like stimuli, that is, to use pictures as targets in the test questions as much as possible because textual questions and answer choices introduces an additional translation process.

Recognition and Recall

Over the years, there has been a significant amount of confusion regarding the use of the terms recognition and recall as it relates to measuring how much learning has taken place (Sheppard, 1967). While very similar in nature, recognition and recall are not synonymous. Their difference in meaning suggests that we have different ways of retrieval once information has been stored. Studies have shown that recognition memory resides on the right side of the brain whereas recall (especially verbal) takes place on the left. Recognition implies a simpler intellectual task that does not necessarily require a significant amount of encoding and carries a longer-term residual. Sheppard describes studies in which even a quick glance was enough to assure some long-term recognition value. He showed objects from a Sears Roebuck catalog to Guatemalan peasants who had no personal knowledge of the items. They were able to recognize the pieces even several months afterwards. Singer (1980) drew several additional conclusions about the differences between recognition and recall. He suggested that the human brain is capable of storing a tremendous amount of visual material --much more than information received in other forms.

The ability of individuals to store so much visual information more rapidly is perhaps due to the holistic or gestalt qualities of the right brain where visual imagery is processed. Thus, material seen just a few times (i.e., presented on television) without any significant effort at learning it, can be recognized if it is re-presented later. The greatest





38


limitation of assessing learning based on recognition alone includes the fact that it is not a useful measurement when the responder doesn't have the object in front of him or her for identification. Perhaps the most significant contribution to the learning process that recognition plays relates to what Singer (1980) called the differential affect. In short, differential affect is an original startle that occurs when new and novel information is first presented. The brain processes this information on the right side of the brain in the same way it processes information to be later used in recognition activities. Singer noted that the startle effect takes some getting used to, but once the initial startle is over, the brain is better able to process subsequent information it receives. Theories surrounding the startle effect suggest that increasing recognition from stimuli used in the quick-cut imagery in today's new media is both a trainable act and is sufficient to gain viewer's attention so that they may recognize it again during future stimuli engagements. As such, it helps to formulate a theoretical basis for certain hypotheses in the current study as well as suggestions for potential future research found in Chapter 5.

Another assertion about the differences between recognition and recall is that they are basically the same with the exception that the former may be a somewhat easier process because it involves a lower capacity threshold requirement (Crowder, 1976). Crowder found that this idea about recognition and recall has been discredited by some due to its tie to a concept that recognition and recall are basically the same processes. Crowder also reports that researchers have more recently looked upon the two as linked but separate operations. Recall involves the generation of an additional process that can be tested separately. For example, a variable might improve recall while, at the same time, damage recognition. Additionally, there is evidence in experiments involving





39


intentional versus incidental learning that shows that recall improves when the subjects are warned of an ultimate memory test, where recognition does not change significantly and may even decrease. According to Estes (1994), recognition, might be better described as two distinct processes: absolute and recognition in context. The former relates to whether a subject recognizes a particular person, or he or she remembers ever encountering that person before. The current study attempts to elaborate on Estes' work, comparing the ability to remember verbatim facts with subjects' ability to remember contextual (i.e., gist) sequences. Estes further suggests that recognition does not "in general, provide a direct window to memory" (p. 230). Rather, it may be said that recognition provides "the best available basis for estimating memory storage", as long as "interpretations of data are guided by appropriate models" (p.230). It is the problems associated with these model interpretations where much of the on-going controversy about the recognition-recall debate appears to reside. Verbatim versus Gist Memo

The apparently unresolved confusion with regards to the use of the terrns

recognition and recall led to the need to find another means to assess immediate memory in the current study. A further review of the literature led to a discovery of relatively new research on two complementary memory processes that have proven useful to the current study. The foundation of these newer ideas about describing memory in terms of a dual relationship in a bi-polar representational system lies in Brainerd and Reyna's (1990) fuzzy-trace theory. Fuzzy trace is consistent with many of the ideas about childhood memories but runs counter to Piaget's developmental/structuralist approach that uses the mind as a computer comparison (Miller & 13jorkland, 1998). Brainerd and Reyna's






40


method is based on a metaphor of intuitionalism, in which people prefer to think, reason, and remember using inexact, gist-like traces rather than more precise (but also more forgettable) verbatim traces. The concept traces memory preferences through age and personality differences, with younger children preferring verbatim, exact references, which become fuzzier as they grow older. Miller and Bjorkland describe fuzzy-trace theory as a hypothesis about representation, which posits that children of different ages are disposed to use different types of representations that are available to them. What differs between this theory and other more mainstream conceptualizations is the propensity to use different types of representations to solve problems, making it more aligned with Piaget-like concepts of constructivism. According to Miller and B3jorkland, earlier misconceptions about fuzzy-trace theory caused confusion as to how it differed from other already established conventional concepts like schema theory. Gradually, however, fuizzy-trace theory began to gain acceptance with mainstream theorists, especially when it was applied to issues of "age differences in suggestibility and false memory creation" (Miller & Bjorkland, p. 188). It was later adopted "outside of the immediate Brainerd and Reyna's [sic] sphere of influence" and began to be used to explain "basic and applied phenomena on a wide range of tasks" (Miller and Bjorkland, p. 18 8).

Independently, Brainerd and Gordon (1994) and Reyna and Kiernan (1994) began to evolve their jointly-developed theories on to other memory domains, concentrating on the specific differences between gist and verbatim memory traces. The concept has been subsequently been adopted by others (Cowan, 1998; Tse et al., 1999). The verbatim-gist classification has more recently been used in a variety of ways, including to





41


counterbalance the more traditional ideas about encoding, and the verbal versus nonverbal comparisons, and more importantly for this study, in analyses of the ability of subjects to recognize and utilize information perceived in visual presentations (Haber, 1970; Loftus, 1979).

According to fuzzy-trace theory (Brainerd & Reyna, 1990), both gist and

verbatim memories are based on similar inputs, but are stored separately. Verbatim memory may be associated with higher ability students, but some researchers have shown that reasoning power can often be disassociated with strict recall of specific details. Others have indicated that gist determination is associated with reasoning powers, making it higher-level activity events (Reyna & Kiernan, 1994). It follows then that gist memory can be assessed independently without requiring that the subjects also score well at verbatim memory and vice versa. Reyna and Kiernan did point out that the only exception to the verbatim-gist independence rule is the fact that success with verbatim recall has been shown to be higher with linear text-based presentations as opposed to pictures.

Although it is a relatively new theory, fuzzy trace serves to explain an important concept of recognition and recall from non-verbal displays as used in the present study. As Brainerd and Gordon (1994) pointed out, so little is known about the developmental and functional relationships between verbatim and gist memories that it is "necessary for researchers to proceed step by step, investigating these relationships within delimited classes of inputs in the hope that general conclusions will ultimately emerge" (p. 163). This gives rise to questions that investigate the effect cognitive style and presentation speed might have on these two forms of memory determination. It is the intent of the





42


current study to address this challenge and extend their analyses into the pictorial memory domain.

The Effect of Arousal on Memo

Another, related topic on memory is the concept of arousal and an attempt to clarify some of the differences of opinion with regards to its perceived positive and/or negative effects on memory. Many of the research findings with respect to arousal and memory appear to be based on the Yerkes-Dodson law (Yerkes & Dodson, 1908) which predicted an inverted U-shaped relationship between arousal and performance. A certain amount of arousal can be a motivator toward change (with change equating to learning), whereas too much or too little can work against the learner. The Yerkes-Dodson law appears to be the basis for much of the early controversies surrounding arousal in mediated messages as to whether they impose a negative or positive effect their viewers (Lang & Basil, 1998). Neil Postman (1986) asserted that the presentation speed of media messages is a force that increases arousal to the point where most cognitive activity is negated. He relegated television viewing to pure mindless entertainment with little or no educative value. On the other hand, Zillman (199 1 ) postulated that arousal can be a unifying force that intensifies motivated behavior. Further, arousal plays a significant role in many motivational models, like the ARCS model developed by John Keller (1983). The 'A' in this acronym stands for gaining the leader's attention, which is handled through arousing or curiosity-seeking cognitive engagement.

The interactions between arousal, motivation, and attention and their affect on leading have gained considerable focus in educational research in recent years. Posner





43


(1982) described four basic ideas about attention. Each is listed below with parenthetical comments regarding relevance to the current study:

I Mental operations related to attention take time to perform and the amount
of time required will correspond to fixed qualities of the stimuli (such as
presentation speed) and the discretionary strategies of individuals (such as
cognitive style).

2. Mental events occurring closely in time are processed successively.
(Montage video attempts to replicate these closely related mental events.)

3. Internal events can be studied by observing the amount of facilitation or
inhibition they cause and this process is hierarchical. (Immediate memory
feeds longer term recollection.)

4. Attentional. processing favors stimulus change. An orienting reflex biases
people toward fresh or novel sources of stimulation. (Presentation speed
can act like an orienting reflex.)

The works of Reeves et al. (1986), Reeves and Geiger (1994), and Reeves and Nass (1996) have centered on the individuality of the attention reflex. They clearly defined attention as "a psychological cognitive process that varies within individuals over time" (Reeves et al., p. 254). These more modem theories look at attention as more of a filtering device that is not exactly an all or nothing gateway. Further, there might be different types of attention requiring different types of responses. For example, vigilance tasks require people to wait for an event and then respond quickly. They decrease over time and require individuals to attend to only one at a time. These are primitive orienting cues that stem from primal man. Attentional preparation tasks give people cues or primes before stimulus occurs. Perceptual intrusion tasks are those so arranged that people cannot avoid attending to them. Finally, attention switching tasks are those in which subjects must switch their attention from one task to another. Reeves et al. elaborated further. They posited an Orienting Response Theory (OR) that suggests, in part, that





44


attention is subject to habituation that explains, for example, how people become accustomed to living in cities and with noises and after a time becomes accustomed to them to the point of eventually being able to block them out without loss of focus on other tasks at hand. OR Theory might explain how youths of today might have become accustomed to fast-paced media so that they can able to learn from it more efficiently than their elders who are generally less exposed to these new media techniques.

These new ideas about arousal also encompass the concept that attention (and

subsequent arousal) in humans involves mechanisms that are used differently depending on the task to be performed. Lang et al. (1999) differentiated between cortical and emotional arousal in which the former can cause the body to bring more attention to a situation rather than less. However, even cortical arousal may have its limits. In reviewing Zillman's limited capacity model, Lang proffered four related principles:

I A viewer will allocate an overall level of processing resources to the
complete viewing task based on goals, interests, etc.

2. A viewer's goals influence the proportion of resources allocated to the
various sub-processes (such as storage & retrieval).

3. The structural and content features of the message elicit orienting behavior
and the automatic allocation of resources to encoding.

4 The content and structural attributes of a message can also elicit arousal,
which results in the automatic allocation of resources to encoding and to
storage.

Zillman (199 1) admitted that when there are insufficient resources available (i.e., cognitive overload) to carry out all the sub-processes, some aspects of cognitive processing will be performed less well. This idea of finding optimum levels of arousal to enhance motivation and performance was further developed by Winifred Hill (1985) who developed a series of laws and theories that included several about stimulating students'





45


interest through arousing content. Among her assumptions were that optimal arousal involves three factors: novelty, complexity, and intensity. The level of arousal (i.e., attention demands) is function of an inter-reaction among these three factors. Highly intensive and novel stimuli are more arousing than those that are weaker or more familiar. Moderate intensity levels of arousal are more reinforcing than either high or low levels of either the newness or novelty of content. According to Lang et al. (1999), it is important to find a consistency and/or balance between the level of external arousal applied by the presentation vehicle and any arousal that might be inherent in the content. If task or conceptual complexity outweighs its novelty, a more moderate stimulus would be more reinforcing for cognition. If, on the other hand, intensity and complexity are relatively equivalent in strength, a change in content (or novel form of content) is what will regain a viewer's attention. Hill summarized by stating that "a highly novel stimulus will [sic] be more reinforcing if it is mild and simple, whereas among very familiar stimuli the more intense and complex ones [sic] tend to be preferred" (p. 176).

Lang et al. (1999) and Lang et al. (2000) elaborated further on the correlation

between intensity and complexity of content, relating the former to presentation speed and the number of structural features occurring in a message. Again, referring to the limited capacity model she suggested that, as the pacing in a message increases, the number of orienting responses (ORs) elicited by that message (and the amount of information available to be encoded) will also increase. Because of these increased ORs, a viewer will automatically allocate more resources to encoding fast paced messages. She also proposed that viewers control some aspects of the resource allocation process by making decisions such as whether to watch, how carefully to watch, how hard to try based on





46


how interesting the subject is, how relevant the information is, or simply whether the they want to remember it at all. Furthermore, Lang proposed that the medium itself can control some aspect of the automatic allocation of processing resources by also eliciting it own orienting responses (ORs) from the viewer. The prime example Lang used in her research was an increase in presentation speed. Lang found that these ORs are automatic, reflexive, and attention responses to changes in the environment or to stimuli, and people have learned that they signal important information. She also found that this automatic allocation of resources is a relatively short-tern response, occurring over seconds. She concluded limited, elicited arousal results in the automatic allocation of resources to both encoding and storage.

Kahneman (1973) proposed a variable capacity model that predicts that arousal elicited by a message will increase one's overall pool of available cognitive resources. While he failed to specify some type of upper limit, he did suggest that secondary reaction time might slow down as presentation speed increases, especially if content is also arousing. Conversely, as speed increases for messages with calm content, rapidity alone might be sufficient to increase the size of the cognitive pool allocated to respond to the message. Basil (1994) proposed fixed capacity model that predicts that higher resources may be committed to the message but the overall size of the cognitive pool remains constant. Overall, arousing messages might cause slower reaction times than calm messages if the presentation speed also increases because resources are automatically allocated to the encoding sub-processes as the result of orienting response. Conversely, if more resources are committed to creating a faster response, resources available for encoding will be reduced.





47


Sheingold's (1973) work further complicates the ability to unwind these

contrasting views of resource allocation by showing that the age and sex of the viewers tend to confound things. Lang et al. (2000) later tried to unravel these variables by eliminating content as one of them. Using a measuring device that calculated skin conductance and heart rate, she gauged secondary task reaction times during several different studies and came to the conclusion that the limited capacity model most accurately reflects what occurs (at least in a casual viewing environment) with the following two hypotheses:

I Arousing content is likely to elicit feelings of arousal and measurable
sympathetic nervous system activation. As pacing increases, viewers
encode more and, therefore, recognize more of specific content.

2. Arousal results in the automatic allocation of processing resources to
encoding and storage.

Lang's conclusions suggest that media producers who want their messages to be remembered should create arousing messages that are presented slowly or at medium speed, or calm messages that are presented at medium or fast speed. These producers should not create messages that are at the same time calm and slow or arousing and fast. While Lang's work has been a watershed for casual television viewing, questions remain as to how much of their work relates to student viewers and in an educational environment where content might simply be considered boring. In spite of these reservations, the limited capacity model was used to develop a portion of the hypotheses for the current study.

The Hueristics of Pictures

To the extent that processing of viewed information is controlled by highly

varying individual past experiences, extracting information also varies greatly. While





48


there have been a few well-elaborated schema theories that "demonstrate specific influences of the effect of showing pictures episodically in the absence of personal or semantic interpretations, a number of independent investigations of perceptions formed on the basis of televised news reports demonstrate convincingly that there exists a "high degree of correspondence between subject interpretation and recall" (Zillman & Brosius, 2000, p. 3 8). Value judgments also play a role information processing. Several studies into news viewing (Bums, 1992; Hayes-Roth & Hayes-Roth, 1977) also have shown that pictorial representations have a positive effect on attention and processing. Likewise, the value of increasing the number of occurrences of pictorial displays (as opposed to simply reiterating a fact verbally) has been demonstrated in a number of studies. Zillman and Brosius showed that an increased number of occurrences of a pictorial message can also influence perception in the news. For example, in their studies, viewers' concept of how much significance or importance they should appropriate to a story seemed to rest in direct proportion to the number of times it was shown on television, to the point that where subjects applied more significance to the story than it actually warranted. Zillman and Brosius pointed the fact that repeated visual representations tended to slant the news, suggesting that there was "strong evidence that the sheer number of concrete, visible sources relating their experiences and concerns does exert an influence on issue perception" (p. I 10). They further suggested that photographs used in print media and on televised newscasts, regardless of how innocuous the content, are remembered longer and the image that was more likely to be retained was of the visualization, rather than the textual message that accompanied it.





49


In conveying any mediated infort-nation, it has long been deemed important to

separate the flow into manageable chunks and to isolate and focus on some events at the expense of attention to occurrences in between (Kozma, 1986). According to Zillman and Brosius (2000), a mediated narrative (even an instructional message) broken down into exemplifying chunks may jump from event to event, irrespective of the event's proximity in location or time, so long as it shares enough similarity that makes it sufficient to provide reliable information about other events in the group (or the group itself). A picture or series of pictures, then, may become a form of an exemplar to be used to aid recall and recognition as long as they formulate a schema for their viewers. While it is has been shown that learners remember more when they attach personal meaning to an idea or concept (Brown & Kulick, 1977; Tulving, 1972), research into casual television viewing has also demonstrated that viewer recognition and recall for content of news shows (Zillman & Brosius), music videos (Hitchon et al., 1994), and/or advertisements (Hill & Lang, 1993) can be modified, either through a form of exemplification of the information by means of lashing loosely related content or structure in associated pictorial representations. All three of these studies, taken together, show that by using a combination of stratifying content into meaningful and related chunks and employing certain cueing strategies can overcome some of the memory-related limitations imposed by presenting unfamiliar content. Mayer et al. (1996) suggest that attention and memory for pictures (especially moving pictures) can be controlled by the use of formal exemplifying features such as zooms, presentation speed, sound effects, and music. Likewise, the format of the instrument used in the current study employs many of these visual exemplification techniques (such as similar coloration, zooming, and differing the





50


length of time images are displayed) to stratify presented information into manageable chunks to aid in the recognition and recall of its viewers.

Pacing versus Presentation Speed

When referring to the speed of presentation, non-education researchers often use the terms, rhythmicity, tempo and pacing, interchangeably. However, their definitions do not always equate. Educational multimedia developers often refer to pacing as the rate that a learner proceeds from one place to the next, as in self-paced instruction (Canelos, 1986; Kozma, 1986). In this connotation, pacing is more associated with interactive video, where interaction also implies some degree of learner control of the rate and sequence of the events that take place. On the other hand, in broadcast or instructional video, where the events are presented to viewers at a pre-defined rate, pacing may be better defined as "the rate of information presentation" (Kozma, p. 14). With this latter definition, pacing is more associated with rhythmicity or tempo. It is within the latter context that the term pacing is referred to in the current study. In recognition of the fact that, in educational circles, the term pacing has long been associated with Canelos' connotation, the term presentation speed is being used exclusively in the current study, even though pacing was the operative term used most often in the literature, especially in studies conducted in non-educational settings.

There is one point in the research where both casual viewing and educational researchers appear to agree: that changing the speed in, and/or adding musicality or rhythmicity to a pictorial presentation adds to memory for its content (Flowers, 1995; Hitchon et al., 1994; Luckett, 1996; Patel, Peretz, Tramo, & Labreque, 1998; Shaffer, Greenspan, Tuchman, Cassily, Jacokes, & Stemmer, 2000; Wagely, 1978). There are





51


also indications (Corcoran, 1981) that rhythmic control can be learned and translated as a coding structure in its own right to enhance leading and can be made to act as a cueing system. Zillman (199 1) found that rapidly presented visual programs foster superior attention, and potentially superior learning. While he admitted that much of the successes of faster speeds may be attributable to arousal, he also quoted studies that showed that the creation of "transient alertness, even by primitive means such as the frequent instigation of the orienting reflex, tends to facilitate information acquisition in audiences for which high levels of attentiveness cannot be (otherwise) expected" (p. 126). Apparently, alternating the speed of an instructional message tends to increasingly arouse the cortex, which, in turn, as many researchers have suggested, tends to make a viewer more alert (Lang et al., 1999; Pearl, 1982; Reeves et al., 1986; Zillman, 199 1). Nelson (1990) quoted studies that seem to indicate that rhythm serves both organizational and expressive functions that can also be shown to significantly help to organize thinking and cognition. While investigating the effects of training using tachistoscopes Peck (1987) found that by increasing the speed of a lesson slightly, it may be possible to increase attention and, by doing so, also increase leading. Peck also found that the effects of altering the speed are more pronounced for slow than for fast readers, something later confirmed by Walters (1983). In fact, a review of the literature into the use of tachistoscopes in remedial and rapid reading training programs (Dick, 1973; Long, 1982; Scale, 1971; Sheingold, 1973; Woodley, 1984) appears to confirm the positive effect a more rapid presentation speed can have on increasing attention and cognition, especially when the content includes redundant clues and it represents conceptual information. These rapid reading training techniques have been shown to also increase visual





52


awareness of larger spans of information, and processing rates can be accelerated without loss of comprehension.

Zillman (199 1) showed that the tempo of a message can be influenced by more than its structural or production process. Message content, for example, may have a carry-over effect. He showed that the interspersion of fast-paced humor is often mediated by cortical arousal. Semi-attentive children in the audience may be made alert "by the inserted humorous tidbits, and this alertness may have carried over into exposure to the immediately subsequent educational material" (Zillman, p. 126). These content studies were not limited to humor. Other studies looked at attention reflexes caused by fireworks (Reeves & Nass, 1996; Zillman, 1991), and negative or violent content (Pearl, 1982; Reeves et al., 1986; Reeves & Nass, 1996). Hill and Lang (1993) used a combination of image complexity and the number of structural or formal features in a message to manipulate effects of pacing. Other studies looked into the interaction between pacing and content. In his studies into the levels of arousal found in fast cuts in music videos, Hitchon et al. (1994) found that fast-paced video montage did not contribute negatively to the complexity or ambiguity issues as long as the story line was not confusing or ambiguous. Theoretically speaking, if increasing the speed of message delivery does not necessarily cause attention overload, one might be able to make a stronger case by looking at a corollary question (i.e., does the slowing of the pace that an image is presented necessarily increase its retention?). In video productions, presentation speed has a direct relationship to the number of frames an image appears, with 30 frames per second being the standard (Stephens, 1996). Potter and Levy (Keller, 1976) and others (Goodglass, 1971) found a significant correlation between retention and duration





53


of images displayed for 200 to 300 and milliseconds, which equates to approximately 1/3 of a second, and those displayed for more than a second. Optimal retention occurred between 200-300 milliseconds and one to two seconds. Conversely, in other studies, it was found that images displayed for more than one to two seconds actually caused a decrease in cognition due to loss of attention (Thorson & Lang, 1992). It might be that duration becomes a "variable of lesser import as the size of the learning set increases and the number of categories decreases" (Keller, p.82). In short, this research appears to indicate that longer displays do not necessarily signify a relative increase in recall, and can actually cause degeneration. In the current study, images appear on screen for as little as 1/3 of a second to a maximum of one second, with the majority appearing for less than 1/2 of a second, well within the range indicated by Potter and Levy (Keller) and Goodglass.

How Learner Attributes Affect What is Learned Cognitive s1yle

The concept of cognitive style has endured for more than fifty years. It came out of the New Look movement in perception that was born during a symposium sponsored by the American Psychologist Association held in New York in 1949 (Witkin & Goodenough, 198 1). Participants in the New Look movement were a loose confederation of psychologists who became critical of the dominant approaches to perception then in vogue. Their main criticism was that most current approaches tended to ignore the person doing the perceiving. Out of that symposium came a flood of new studies aimed at looking at the personality traits of individuals (i.e., individual differences) during the





54


process of learning. Out of this broad context of research emerged the concept of cognitive styles (Witkin & Goodenough).

Guilford defined cognitive style as that "which conceptualizes intelligence as having a process dimension" (Green, 1985, p. 2). According to this view, learning is not merely an automatic reaction to a stimulus but a set of operational steps that varies, depending on individual proclivities. Brumby (1982) asserted the following assumptions regarding cognitive style:

One's cognitive style is singular (i.e., an individual has only one) and can
be measured on a bi-polar scale.

While an individual possesses one style, others may be present in varying
degrees.

Individuals can select an appropriate style appropriate to the task at hand.

Earl Messick (1970), a charter member of the 1949 New Look movement, catalogued nine dimensions of cognitive style, covering research that he and his colleagues performed, and that of others who came along in the years that followed. His list included scanning, breadth of categorization, conceptualizing style, levelers versus sharpeners, distractibility, tolerance for unrealistic experiences, cognitively complex versus simple, field-dependent versus independent, and impulsive versus reflective. Of the nine mentioned, meta-research (Green, 1985) has shown the latter two to be the most commonly accepted as credible sources for investigating how individuals perceive and process visual patterns. Of these two, impulsive-reflective was determined to be the most relevant to the current study.

Impulsive Reflective Style

The impulsive-reflective scale was borne out of Jerome Kagan's research work





55


(1965) with cognitive tempo (i.e., rate of cognition) in younger aged children in which their speed and attention to detail was indexed. Children deemed to be impulsive tend to react very quickly and make quick decisions (i.e., they select the first answer that occurs even though it may be wrong), while reflective tend to take more time to consider various options but are also generally more accurate with their interpretations. Contrary to many stereotypes about bright children thinking quickly, Kagan (1966) found that neither tendencies for fast or slow decision times were significantly related to verbal ability or innate intelligence. According to Block, Block, and Harrington, (1974) some have criticized Kagan's findings, noting that although response times were positively correlated with performance, overall, the correlations were often quite low (ranging from close to zero to around .45). However, according to others (Ayabe, 1973; Bridgeman, 1980) theirs and others' research appeared to ignore the converse of the measurement scale that compares slow-accurate to fast-accurate, which states that accuracy and not speed counts the most in complex problem-solving situations. Bridgeman points out that, even with timed tests, the negative correlation between speed and accuracy can also be a plus, suggesting that there are many children who "are fast and accurate" (p .212). The problem, then, appears to be taking an extremely bi-polar classification and make subjective determinations in overall psychological and/or personality evaluations that may require less polarization and more sub-categorization. These classifications (or close derivations thereof) have been used quite successfully and deemed to be valid classifications in specific educational evaluations. Campbell and Davis (1982) found that the "reflection-impulsivity style construct emerges as an ecologically valid and parsimonious descriptor of a component of student behavior" (p. 8) where it is





56


determined, for example, to hinder leading performance. Conversely, Boyden and Gilpin (1978) found latency and error rates to be independent of measures of distractibility. The relationship between impulsivity and academic achievement is not necessarily tied to aptitude or intelligence but to one's ability to attenuate and/or process specific types of inputs (Kogan, 197 1; Leino, 198 1; Messer, 1970). Cooper (1982) suggested that differences in processing speeds may well be accurate indicators of one's ability to process mostly visual information. Ridberg et al. (1970) found that cognitive style is predictive of performance in a variety of measurement tools, including those for reading recognition, secondary leading, and reasoning. Hedberg and McNamara (1985) found that when using visual information the tendency to reflection or impulsivity might be an important predictor of performance, particularly in relation to time and error under conditions of response uncertainly and time pressure. Merri nboer (1990) was able to use the classification as a predictor of academic performance and use it to pre-arrange feedback strategies to increase effective computer usage, especially in younger students.

It appears as though impulsivity is a characteristic that mitigates with age. Wright (1979) found that impulsivity lessens over time as one grows into adulthood. Okun et al. (1979) found that adults have significantly higher latencies than middle-aged and younger adults. In other words, as one grows older an individual tends to become more logical in his or her outlook, incorporating familiarization and planning into perceptual processes. However, an increased ability to exert a reflective response to stimuli can also work in reverse. O'Brien (1968) discussed effects of becoming stimulus bound, especially when exposed to substandard information. This phenomenon, which occurs mostly in adults, causes individuals to so over-structure their internal response that they





57


were unable to subsequently correctly perceive a very clear and undistorted image. Although adults tend to be more reflective in response to stimuli, these same reflective tendencies could work against them when comparing the correctness of their responses to younger individuals, especially if the younger person participates more often in visually oriented past times. There appears to be a crossover of the effectiveness of one's increased developmental reflective abilities and one's ability to effectively develop a corresponding facility to realize perceptual closure from visual data that is either less than highest in fidelity and/or is displayed at a very rapid rate.

Anderson and Revelle (1994) looked at the effect changes in daily arousal rhythm patterns play in causing similar alterations in impulsivity-reflectivity. In this context, arousal is meant to portray the processes that mediate non-specific alertness, or liveliness. Impulsive tend to demonstrate high alertness and sense of arousal, a key element in learning preparedness. Highly impulsive children have also been linked to certain attention deficit syndromes, as portrayed by their lacking in the ability to sustain longer periods of attention (Anderson & Revelle). Anderson and Revelle's research indicates that the impulsivity rates vary (i.e., are more prominent) by the time of day, with impulsivity in those individuals with impulsive tendencies being more pronounced in the morning hours.

Using impulsivity-reflection to categorize individual differences has not been

without a certain amount of controversy. While the impulsive-reflective scale can be very beneficial in predicting certain academic outcomes and has significance in timed tests, there are shortcomings in making a completely accurate independent identification with regards to whether a student fits this classification in all contexts and under which





58


conditions (Bridgeman, 1980). For sure, the impulsivity label was considered powerful by some (Kagan, 1965; Kagan, 1966) and appealing by others (Hedberg & McNamara, 1985). However, some researchers (Ayabe, 1973; Bridgeman, 1980; Campbell & Davis, 1982) found that some weakness exists in the outcome of using classification as it relates to discriminant validity and generalizability. In other words, there appears that some evidence of loss of power, due to a corresponding loss of independence of traits in the subjects tested in certain studies. The criterion used to justify discriminate validity in the identification of a classification method or methods, which measure a discrete trait to the exclusion of others (Campbell & Fiske, 1959). An additional problem in using this classification is that it appears to be more heavily weighed down by the errors subjects made than latency itself (Ayabe; Block et al., 1974). Further, there is an indication that many studies failed to yield consistent findings in an ecological (i.e., classroom) setting due to inconsistencies with the exact definition of impulsivity (Campbell & Davis).

Lastly, previous studies systematically excluded females. Some researchers went so far as to suggest that gender differences were so extreme that they would skew the results (Cairnes & Cammock, 1984). The current study looked at gender differences to see if the same rationale is still justified.

Factors From Film and Television That May Affect Learning Preconceived Mental Demands

Salomon's (1984) with mental effort appears to confirm the idea that learning that is or is not realized by a medium can be influenced by one's previous experiences with that medium by coloring student preconceptions about the mental demands being placed on them by that medium. In his studies, students reported that they exerted more effort





59


towards reading than with television, but they rated the latter as more realistic and more efficient. Greater mental effort requirements attributed to reading seemed to result in greater perceived demand (PDC) towards that medium. Conversely, as these same students perceived less mental effort was required towards television, they exerted a correspondingly less effort and paid less attention. The amount of effort exerted was also in direct proportion to preconceived notions about one's perceived self-efficacy about either medium. The more comfortable they were with the medium, they less effort was exerted. While Salomon's studies showed some overall weakness due to some discrepancies between self-reports and actual performance, his studies were backed up by several others'. For example, Ksobiech (1976) reported that pre-conceived task demands affected the proportion of time students requested to see the visual portion of a video presentation or only the audio portion, based on whether they were told that they were going to be examined on the material, or were only asked to observe it for evaluation. Students sought the video or audio source that maximized the purpose for which they were to receive the material. Krendl and Watkins (1983) found that students viewing television for entertainment value differed in the number of items they were able to recall from it from those who were asked to view a presentation to learn from it.

Accordingly, some researchers (Langer, Blank, & Chanowitz, 1978; Langer & Imber, 1979; Schank & Abelson, 1977) hold the view that most encounters with television cause viewers to generally approach it with a sort of mindlessness that can undermine its role in an educational setting. It was very important for the success of the current study that the subjects do not consider the activity of viewing the video and subsequently answering the questions that followed as an entertainment activity,





60


especially because researchers have shown that currently most of younger viewers' daytoday encounters with both computers and videos (in particular music videos) are for amusement purposes (Seward-Barry, 1997).

These ideas about preconceived demands relate directly to the procedures used in the current study. Care was taken while administering the test to 1) demonstrate the purpose of the viewing activity using a sample portion of the video and 2) instruct the subjects to treat the activity just as if they were to be graded on it, even though no intention to do so was introduced to them beforehand. Exemplars and Other Formal Features

Although many studies seem to establish television's reputation as a medium that promotes mindless non-engagement of its viewers, research has been far from definitive. As of a recent count (Abelman, 1995), over 4,000 scientific research articles and government reports have been published examining media effects with special emphasis on the impact of television on children, and still the literature continues to be imprecise as to how children learn from media interactions. On the other hand, some studies appear to be more conclusive. Anderson and Bryant (1983) produced results of studies of specific formal features of television and showed that the research can be fairly robust in that subjects were able to define and understand the internal mental processes involved with interpreting mediated messages. For example, they found that valence in children can be influenced by specific interpretations of what they see and hear. Particular (i.e., a woman's and/or child's) and/or peculiar voices, sound effects, auditory changes, and visual movements all had a positive effect on recall and comprehension, whereas men's





61


voices, long zooms, and long periods of inactivity were found to be less comprehensible and less interesting.

Other studies used foreign (Salomon, 1994) or indigenous (Worth & Adair, 1997) subjects unfamiliar with certain features and/or function common to film and television production in the United States confirmed these findings. Salomon found that Israeli children unfamiliar with the format or content were able to recognize and recall information from various episodes of Sesame Street just as well as American children who were well exposed to the shows. In teaching film-making young Navajo Indians who had never seen movies before, Worth and Adair were able to demonstrate in qualitative and longitudinal studies that certain film communication contains universals that do not appear to vary, even in light of cultural biases. The Navajo films showed, as a group, similar functional patterns in their visualizations to those of other cultures. Worth and Adair found that how we interact and become social through our manipulations of a variety of symbols and symbol forms (such as those associated with film) enables people to communicate more fully and fruitfully with each other. For example, while observing general movement patterns in a film, a viewer can also know exactly what the producer is trying to communicate. Worth and Adair (p. 8) wrote, "you can actually see what is being done -how it moves. If you write an entire book about it, then it is still. You give it to someone and he reads it and he does not get the same picture in his mind. What I want to see is how something can move in front of my eyes." Worth and Adair were suggesting that a movie is a movie, regardless of the language and culture in which the maker lives. There are subtle differences in the use of specific techniques, content, and so forth, but overall, most features and patterns were the same (i.e., the linear nature of episodic





62


sequences). While the study with the Navajo tribe dealt with making films, Worth and Adair found that much of their findings would carry over to video. In later experiments several years later, their results with video followed a parallel course. For example, they found an aversion on the part of their subjects to the use of zooming and close-ups while teaching them to use the medium. These same proclivities carried over, even with advanced technologies and easier access found on the video cameras.

Salomon (1979; 1994) suggested that a proper review of media effects of

television involves the examination of the subtle interactions among components of the medium, individual characteristics of its viewers, and the proposed outcomes for which the medium is being used. Zillman and Brosius (2000) make a further differentiation. To them, many of the effects attained by mass media are due to the redundant, stereotypical, and cyclical attributes by which viewers learn by extracting personally relevant material (i.e., episodic memory per Tulving (1972)). According to Zillman and Brosius, rather than creating a sense of mindlessness, this redundancy is actually what creates the power of the medium. They pointed out that all media (especially mass media) create learning through the use of specialized media effects (i.e., exemplars) that may be best realized when used over and over again. For example, most successful mass media producers understand that, in conveying information, it is important to cut the flow into manageable chunks and to isolate and focus on some events at the expense of attention to occurrences in between. In other words, a story line may jump from event to event irrespective of the events' proximity in location or time. It is an effective strategy so long as the succeeding events share enough similarity and are sufficient to imply that every individual event is capable of representing the group of events at large (Zillman & Brosius). In other words,





63


a scriptwriter shows a man getting into his car and driving out of the driveway. Then in the next scene he is on the highway on the outskirts of a big city. Finally, he pulls up into his parking space at work. Enough information has been presented to allow viewers to imagine that time has passed and the man has driven all the way to work and are able to fill in the information in between. Bryant and Rockwell (199 1) found that viewers wellversed in television and/or film techniques can fill in gaps of omitted information, even after missing as much as five or more minutes of a television show simply because they are familiar with the repetitive formats used by screenwriters. Abelman (1995) showed that viewers can learn how to build an internal concept map for the most often-used devices in films and teleplays (including time leaps, fall backs, dream sequences, scenes showing characters recalling past events, etc.) even without comparable real-world experiences. This appears to further reinforce the theory that television viewing is a learned activity and highlights the inter-relatedness of children's linguistic, cognitive, and perceptual skills for accurate comprehension of television's most basic narrative device: temporal sequencing. Corcoran (198 1) suggested that intelligence may be defined as a skill in a particular medium and that symbolic codes that serve communication purposes can be internalized by a receiver to serve in a cognitive capacity. In other words, filmed/televised techniques such as zooming, slow motion, or rapid montage may actually be thought of as tools of thought. Viewers appear to learn how to use these tools naturally without being trained, because these types of story-building and post-production editing devices are used redundantly in one show after another (Abelman; Arnheim, 1974; Corcoran; Jankowski & Fuchs, 1995; Zillman & Brosius).





64


Pearl (1982) discussed the relationship between form and content and admitted that her research showed that it is theform (that is, the way it uses verbal and linguistic codes), not the content of television that is unique. However, she also cautioned that form and content cannot always be distinguished --"no more than grammar and meaning in any verbal language can" (p. 24). Like grammar, some forms are unique to the medium and apply syntactical meaning only in the context of that medium. For example, slow motion is not real and its meaning must be learned. But once learned, these formats are generally used by people in their own thinking. Pearl outlined a taxonomy for this coding structure of formal features. For example, entropy reform complexity is a combination of the number of different scenes in a show, the number of characters, and the number of times the scene and characters appear. Dynamism is the rate of change in scenes and characters. In this view, content variation is merely determining which features appear together, their tempo or rate, and/or their variability within scenes.

Unfortunately, the link back to attribution or lack of perceived mental demand has been caused by television producers who tend to use these same technical features over and over again as a means of convenience in an entertainment industry that tends to value volume over quality (Jankowski & Fuchs, 1995). However, redundancy of format and content can also play a role in causing viewers to make value judgments about what they see. Zillman and Brosius (2000) cite studies showing that an increased number of occurrences sometimes influences viewers' perception of the news story. For example, if the number of occurrences of a particular story in a newscast is increased, viewers tend to be influenced by the story in that the event was probably more significant than it actually was. (i.e., it tends to slant the news). There is strong evidence that "the sheer number of





65


concrete, visible sources relating their experiences and concerns does exert an influence on issue perception" (Zillman & Brosius, p. I 10). They go on further by stating that the incidental use of image exemplars that add non-redundant, specific information to the text of a news report does influence issue perception. Specifically, the incidental nature of a pictorial supplement to a story goes unrecognized and, as a result, is integrated with the narrative information in fostering perceptions and judgment. Over time, if presented enough times, the photographs are actually remembered longer and the retained image of the incident was of the photograph, not of the text. Furthermore, if the image that supplements a news story presents a negative or positive aspect, that image can actually influence one's value judgments. This is especially so if the presentation speed of that image is varied (Barnett, 2000; Corcoran, 1981; O'Brien, 1968; Pearl, 1982).

However, presentation speed has not always been consistently viewed as a

positive media effect. Pearl (1982), for example, talked of television being a magnet on viewers that "shapes their behavior" (p. 1). In an attempt to investigate the ties to violence and television viewing she found herself in conflict with her own research. First she claimed that "the rapid form of presentation characterizing American television in which novelty piles upon novelty in short sequences may well be counterproductive for organized and effective learning sequences" (Pearl, p.20). But she then goes on to state that this may be more of a problem with younger children. Kozma (1986) appears to support this view. He points out that a fast tempo does not necessarily sustain visual attention, and that an increased tempo is not likely to provide enough time for the viewer, especially one inexperienced with television formats, to extract that which is needed bring it into working memory. In addition, others have indicated that as the amount of





66


information increases with the increase speed, short-term memory limitations are soon reached and information is lost (Lang et al., 2000). The fact that broadcast medium presents information linearly and does not permit a child to return and review content puts the onus on the producer to utilize commonly recognized techniques such as maintain scenes for longer periods, to limit the objectives of the program, slow its speed, and/or to use slow motion. Kozma points out that shows like Sesame Street accommodate these limitations by constraining the objectives of each episode and representing information in different forrnats. However, simply slowing the speed down may not be the total solution either. Cronbach and Snow (1977) point out that viewers who are familiar with the content might actually need a faster speed to maintain their attention.

Salcedo (1985) looked at such features as focal length and found that close-ups were regarded as more intimate emotionally by viewers. Other features elicited similar varied impressions (Barnett, 2000; O'Brien, 1968): time on screen (the longer characters appeared on screen, the better impression they left on viewers), types of transitions (fades are considered to be more positive or emotional than cuts), and slow motion among them. In addition to its presentation speed, the video used in the current study utilizes several of the above features to increase comprehensibility and emotional involvement of its viewers.

Symbol Systems

Symbol systems differ along several dimensions, which Salomon (1979; 1994)

believed are to be primarily those of notationality, repleteness (density) and resemblance. A notational symbol system is one in which there is a strong, clear, and consistent correlation between a set of symbols and a set of objects or concepts, such as a written





67


language, musical notes, or mathematical symbols. Repleteness or density refers to the relative richness or number of dimensions that the information conveys, like a sketch as compared to a full-color painting, a textual passage as compared to a picture (even more so if that picture is moving). Salomon distinguished among symbol systems with respect to the degree that they resemble the object or concept being represented. Symbol systems tend to vary in terms of their depictive and/or descriptive reference, from a one-to-one to a totally abstract representation. Salomon also distinguished between psychological and real resemblance. Depicting an object with a high degree of realism may or may not be required for understanding, depending on an individual's pre-conceived ideas as to that something should look like. As all cognition and learning are based on internal symbolic representations that are central to all systems of mediated communication and thinking, actual resemblance (i.e., fidelity) may or may not be required in individual cases. Gardner, Howard, and Perkins (1974) agreed with this thinking and proposed that concepts likefidelity should be viewed in its broadest context. That is, the symbolic process should be compared to how one imagines or conceives something, not necessarily how true to form it is (i.e., colors, lines etc). In other words, a medium might be said to be faithful to the real world if it presents information in a way that is true to the way one thinks (i.e., in a streaming sense).

Salomon (1979) explains that it is not the medium itself that makes the difference in message processing. Rather, he held that film and television are generally not symbol systems themselves, but rather are a place where multiple symbol systems are used. Because they manifest no one system, television and film adopt the symbol system of the content they depict. For example, television and film may use photography, print, speech,





68


dance, music, etc. and can be replete, resemble reality, and approximate notions of notational ability, borrowing their power from other media. This same power to employ a wide range of symbol systems may also define film and television's potential shortcomings, reflecting a fallacy in the belief that they always deliver to their potential (Wetzel et al., 1994). For example, if a video screen is simply placed in front of the class that displays nothing more than a talking head lecture covering an irrelevant content, there will be little hope for realizing a valuable instructional delivery. All media carry with them a systematized methodology for symbolic processing, but that varies in how they are used, in what context, and by which individuals.

Symbol systems theory plays a significant role in formulating some of the

hypotheses for the current study. First, the pictures found in the depiction of historical events are iconic in nature and, as such may be considered primary symbols. Second, although the speed of delivery is considered rapid or fast (at about 200-300 millisecond per image), the speed within this context is varied, which manipulates the amount of mental effort required to comprehend the conceptual context. As the images are presented in combination in a montage format, they represent reality to viewers in a way similar to the way they think (i.e., in a streaming sense). Lastly, as outlined by Salomon (1979), fidelity not withstanding, the images in the videos used in the current study are indeed symbols in that they represent the sole source of reality about the historical events to the viewers who were not live witnesses to those events. In addition, as per the conceptualizations of Snow et al. (1965), the amount of information the subjects are able to extract from the videos may well depend on the amount of previous experience they have with this medium (television).





69


Evaluating the Educational Impact of New Media

The technologies which are helping video and computing converge, and with

which it is now possible to more easily create visual productions full of rapidly presented montages, are often referred to as new media. In spite of its inference, the term new media is not new. Researchers have been alluding to television in this way for almost fifty years. In writing his introduction to the Seventy-Third Yearbook of the National Society for the Study of Education, David Olson (1974) noted that new technologies/media are not the panacea to all possible educational reform needs that researchers were looking for:

It would be of much greater promise to discover the areas in which
media diverge and hence serve different purposes. Perhaps, the
function of the new media is not primarily that of providing more
effective means for conveying the kinds of information evolved in the last five hundred years of a book or literate culture but, rather, it is that
of using new media as a means of exploring and representing our
experience in ways that parallel those involved in that literate culture.
(p. 8)

Olson's ideas have been backed up by other researchers. Gross (1974) suggested that new media does not reduce the vital importance of competence in the basic modes of intelligence and communication. Chu and Schramm (1968) found in earlier studies that screen size, using animation, changing the aspect ratio or size of the screen made few, if any, significant difference in leading, even though these changes all were seen to be preferred by students in follow-up questionnaires. While the current study also takes advantage of many of the features of new media, it is not new media by itself that can take credit for any positive (or negative) outcomes with regards to memory. The basic premises for how to evaluate the use of media in educational settings remain. Olson reaffirmed the need to base research of new media using a symbol systems approach. He





70


resolved that previous research showed the lack of a strong theoretical base, and that new media would not reach their potential until some scheme is shown through empirical evidence that they can re-specify information, how that this re-structuring would be influenced by the media that presents it, and what the psychological consequences of relying on that re-structuring might be. Media are similar in the knowledge they communicate, but they differ as to the skills needed and developed by using them. Evidence appears to show that educational media cannot be chosen simply because of their ability to convey certain kinds of content. Rather, they must be chosen based on their ability to develop the intellectual processing skills that they help to develop.

Olson (1974) outlined several criteria for analyzing the theoretical potential of various media for use in educational settings which became a basis some of the goals of the study and several of the recommendations in Chapter 5. To evaluate the instructional impact of media one must:

Identify the specific symbol system it uses.

Identify the specific basic skills that are required and how much of literacy
is required for use of this medium.

Identify the intellectual consequences of exposure to that medium (i.e., is
the ability to imagine an action in slow motion dependent upon having
seen slow-motion film?).

Identify the scholastic goals for which the medium is most appropriate.

Detennine how knowledge and skill (intelligence) in the medium can be
evaluated.

New Media and Memorv

New media make it very much easier to create new an interesting pictorial storyline using rapidly presented montages, as evidenced by the exponential use in





71


commercial television. Market and psychological research (Reeves & Nass, 1996; Stephens, 1996) has demonstrated its effectiveness on memory in commercials and marketing campaigns. Further, researchers (Reeves et al., 1986; Reeves & Nass) appear to concur with Olson's assertions in light of a rapidly changing technological arena. Rapid changes are also in process for the way in which researchers view media assessment. Several things gave rise to an increased interest in looking at the psychological and sociological aspects of television viewing during the late 1980s. An evaluation of the then-current research by Reeves et al. found that there was an increased understanding about attention. First, was the fact that recognition was not always present, and second, and most important, that continued attention was not always a necessary prerequisite for learning and understanding. Until that time, most evaluation of attention was limited to measuring direct eye contact. However, Reeves et al. discovered that due to increased changes in visual displays, with their rapid scene changes, quick movements, and zooming, that viewers appeared to have developed a sense of being able to filter stimuli and, in essence multi-process. At the same time, much attention was paid to the potential hann this newer, faster paced television was having on its younger viewers (increased proclivity for violence, poor attitudes and moral judgment, etc.). The question Reeves et al. began to look into was why these same effects could not be implemented into instructional programs in order to accomplish increased educational goals. The answer may or may not be obvious and requires that one take a deeper look at the aspects of the new characteristics of message delivery brought on by newer technological developments. In other words, attention is only one of the variables. Reeves et al. also found that earlier studies to determine which program attributes caused children to attend





72


to television during shows like Sesame Street or the Electric Company served to also point out that attention might be better defined as "visual selection" (p. 253). Attention was measured by a hidden observer who pushed a button when a child's eyes were directed towards the television screen. This type of eyes on screen method dominated research for years. However, more recent literature has clearly defined attention as "a psychological cognitive process that varies within individuals over time" (Reeves et al, p. 254).

Reeves et al. (1986) later began to find that perhaps, attention, as it relates to new media, is a many-fold process and that it is not necessary an all-or-nothing effect. Perhaps, different types of attention required different types of responses. Vigilance tasks are an automatic process that requires people to wait for an event and to respond quickly. This process decreases over time and only allows a subject to be able to attend them one at a time. Reeves et al. correlated this to primitive cues that are born out of human's animal instincts. A attention preparation tasks involve giving people cues or primes before a stimulus occurs. Television can be thought of as a continuous priming or cueing process. This strategy could be used to study the sequencing of various visual devices, such as pans, zooms, audio silence, screen luminance, or presentation speed and their ability to prime attention for subsequent program content. Other questions could be asked about the effects of un-cued or abrupt scene changes on attention and the circumstances under which cueing increases processing efficiency. Just as Olson viewed symbol systems, attention appears to require internal devices that are also used differently, depending on the task that is to be performed. Measuring attention can be added to





73


Olson's list of criteria for analyzing the theoretical potential of various media for use in educational settings.

Effect of Fast Cuts on Memo

Reeves et al. (1986) re-defined the result of attention changes in humans,

referring to them Orienting Responses (OR's). These types of responses help to explain, for example, how humans become habituated to living in large cities with lots of loud noises, or how one becomes so used to the common sounds found in his or her home (like a cuckoo clock) and begins to tune them out. Reeves et al. identified several characteristics of OR's that have implications for evaluating the effectiveness of fast cuts in an educational setting. Reeves et al. also related these changes to cortical arousal, in which an individual may have some type of control over and learn how to manipulate for his or her own purposes. They noted that there is some evidence to suggest that those who habituate slower perform better on audio and visual tasks than those who habituate quickly. They also documented studies that showed that high cortical arousal in adults is related to an increased ability to remember information --both of which are premises of the current study. These ideas about OR's also formed the basis of some of the recommendations found in Chapter 5.

Presentation Speed versus Content

Lang et al. (1999) more recently discussed adding interesting and arousing content as having a positive effect on cortical arousal and, therefore, recall and recognition. She also noted that too much arousal can be harmful. In describing a limited capacity model, she outlined a sequence where viewers allocate an overall level of mental processing resources to the complete viewing task based on their goals, interests, etc. In





74


turn, these goals influence the proportion of resources allocated to the various mental sub-processes (such as storage and retrieval). The structural and content features of the message elicit orienting behavior (OR) and an automatic allocation of resources. When there are insufficient resources available to carry out all the sub-processes, some aspects of processing are performed less well, indicating that too many inputs are overloading the system. Fast-cuts can be this overloading stimulus, unless the content is delivered in a structured and continuous way (i.e., it builds its story through a series of related schema).

Lang et al. (1999) further noted that it is the medium itself, by calling upon certain processing requirements (i.e., its symbol system) that can elicit an Orienting Response. The Limited Capacity Theory suggests that stimuli that elicit arousal result in the automatic allocation of resources to encoding and to storage. Thus, an increase in presentation speed by itself can increase a viewer's arousal levels and result in the need to allocate additional resources to encoding and storage. Lang and Basil (1998) and Lang et al. (1999) and Lang et al. (2000) found that content alone can also elicit arousal. Arousal will create the cognitive process to reach its top limit much more quickly if content is also arousing. On the other hand, as the presentation speed increased for messages containing only calm content, subjects were able to take on additional cognitive tasks. Lang's studies provided evidence that one should blend the types of content with the speed of the pace it is presented. Lang et al. (2000) found that memory capacity problems can be ameliorated if the content is structured so that the amount of new information provided is limited, irrespective of the presentation speed of the video presentation. Further, they found that, if the pieces of information are related to some larger overall schema, the presentation speed also did not matter as much. These concepts





75


were supported by earlier research (Amheim, 1974; d'Ydewalle & Vanderbeeken, 1990) into recalling visual sequences where individual frames of information were best remembered when they related to an overall schema of events. An important factor in memory enhancement in all these experiments was the cohesiveness of the contextual content of the messages being presented.

Montag

Mitchell Stephens (1996) elaborated on the potential for an increase in memory capacity for related schemes of events in his chronicles of the development of montage in cinema and television. He delineated the beginnings of the use of montage from its use in the early films of Sergei Einstein (1949), to Chuck Braverman's (1969) use of kinestasis editing, and its later extensive use in MTV music videos. Stephens demonstrated how rapid presentation rates have not only enhanced montage videos. He also provided several examples how producers have been able to replace textual messages as a primary form of communications. His work appears to support the same type of "technological determinism" (p. 2 1) found in McLuhan's (1964) the medium is the message principle: that the means we use to express our thoughts also change our thoughts. Stephens suggests that, conceptually, most of thinking with regards to mankind's current system of logic is based on the literacy of the written word. He even predicts that new media, with its ability to empower individuals to regularly use vast communicative strategies like montage, will create a new, unique forrn of abstract logical thought, similar to what Alan Kay (1999) had in mind.

In order to provide full understanding of the concept of montage, Stephens (1996) traces the etymology of the term. Montage was first introduced by Russian filmmakers,





76


who referred to a concept of permitting producers to "record movement in the most complex combinations and [sic] to place points wherever they wanted" (p. 102). The Russians began to produce short films that placed short cuts of still photography and pasted them together to form a surreal photomontage in which "the meaning of a shot is dependant upon the shots that surround it" (Stephens, p. 102). It is this view of montage as a collection of highly related thoughts expressed visually that provided the catalyst for its use in the current study as a way to provide gist memories in videos without the need for textual supplements.

Montage, in combination with fast cuts, has become a whole new form of

intellectual discourse that Stephens (1996) called "complex seeing" (p. 178), a technique that was not available to early television producers. Much of early television lingered much longer over scenes, allowing the narrative to play a significant role in telling the story. According to Stephens, new media technologies are changing the paradigm. "Now that the cuts and montage have increased the speed of the instances we see, we are no longer dwelling on the individual instance but, rather, are being persuaded by the onslaught and the continuity of the entire series. Not only are we affected by each individual image but their composite total" (p. 180). In essence, the ability of an individual to become accustomed to the fast-cut montage may be a newer, higher level form of persistence of vision --the gestalt-like concept that forms the basis of being able to watch films. The mind closes the gaps in between pictures in a stop-action series of movements to form one continuous motion. The current study attempts to see if montage has the potential to follow a similar track in an educational setting.





77


Summer

In researching the current trends of the so-called media effects, one thing has

become evident. It is the newer structures and the form (of which rapid presentation is a part) of the montage video segments that continue to play an ever-increasing role in how messages are delivered. This view is not a new one. McLuhan (1964) and Salomon (1979) predicted media effects long before technology caught up with televisions promise.

Flowers (1995) cited research as early as the 1 970s that contended that educational environments might be "too visuo-centric" (p. 570), indicating that looking into visual processing may have been overdone in many previous studies and that its role in learning might have been exaggerated. In some respects, Flowers might be correct. Singer (1980) admitted that many previous assertions regarding textual versus visual image processing had yet to be supported directly by research evidence, and suggested the need for further experiments on length of video sequences and its effect on retention of content (or at least intelligent comprehension of content). Intraub's (1999) experiments appear to answer that need. She found that humans possess a "remarkable capability" (p. 67) to understand scenes that are presented at a pace far more rapidly than normal scanning, so long as the scene changes overlap with at least somewhat related content.

Over the past decade, media researchers have started to look at the impact new media have on cognition in a casual viewing environment (Basil, 1994; Hawkins et al., 1997; Hill & Lang, 1993; Hitchon et al., 1994; Walma van der Molen & Van der Voort, 2000; Watt & Krull, 1977; Zillman, 1991). Some are also beginning to conduct more extensive research on the effects on memory of editing techniques and the content found






78


in rapidly presented montage in commercial programs (Lang, 1994; Lang & Basil, 1998; Lang et al., 1999; Lang et al., 2000). It is perhaps Lang's work in commercial television that provides the best backdrop for studies that investigate a corresponding effect of these new media techniques might have in the educational domain. If one accepts the contentions of Reeves et al. (1986) and Stephens (1996), it should be easy to understand how today's youth, brought up on MTV, with its fast cuts, rapid movements, and iconoclastic acoustics, could get used to, prefer, and possibly learn from similar techniques in an educational setting.

There are those who contend that the rapid fonnat of television segments takes away from attention and understanding (Anderson et al., 1979; Anderson & Bryant, 1983; Anderson & Collins, 1988). Reeves et al. (1986) demonstrated information in several research studies that contradict that belief. Stephens (1996) and Meyrowitz (1985) appear to agree. They concluded that, once its capabilities are fully implemented using techniques like fast cuts and video montage, television will provide an opportunity to transcend the time and place continuum by cutting rapidly between images taken from different contexts. In other words, viewers can find additional meaning in the relationship between numerous different scenes. Taken in this way, the video image provides opportunities for creative seeing, making the case that interpreting images requires more imagination of their viewers, not less.

This does not mean that educators or their students have no obligation with

regards to content. Lang and Basil (1998) and Lang et al. (1999) and Lang et al. (2000) have suggested that producers who want their messages to be remembered should create arousing messages that are presented slowly or at medium speed or calm messages that





79


are presented medium or fast. Producers should not create messages that are, at the same time, calm and presented slowly or arousing and presented too fast. Carrying this thought into an educational setting, Verhagen (1992) found that increasing the presentation speed of video images can be fully interpreted and remembered if learners start their viewing task with a realistic expectancy of its demand characteristics and are motivated to tune their mental effort accordingly. Further, research in tachistoscope training (Schale, 1971) showed that rapid reading training increases visual awareness of larger spans of print, and that the processing rates can be accelerated without significant loss of comprehension. As a minimum, even if comprehension is the same, or even if it suffers only slightly, fast cut montage may not be a negative influence.

Media choices may be as much about cost and efficiency as about cognition and leading (Cobb, 1997). If it is accepted that one type of efficiency is cognitive efficiency, it follows that media choices can profit from an understanding of cognitive processes much in the same way as the latter can be positively leveraged by the type of media that is ultimately selected.

In this chapter, a theoretical basis was laid out for the methodologies, procedures and analyses used in the current study. An analogy may be made to describe the way that video montage derives its ability to convey meaning through the use of closely-related pictures and rapid presentation. One should view montage similarly to the concept of chunking in speed reading of text-based passages. In montage, the individual pictures take the place of words and sentences. Several related pictures become the paragraphs. Presentation speeds may vary, just as one might vary reading rates, causing variations in the ability of individuals to remember specifics and derive contextual meaning from the





80


passages. What remains is trace memory for remembering the gist of the storyline, even if the specific words or phrases (i.e., verbatim memory for the specific pictures) cannot be recalled exactly.

The next chapter outlines the specific methodologies used in this study to

investigate some of the affects that variation of presentation speed, personal cognitive style, and gender may have on memories for information presented in the passages. These methods follow closely the ideas of looking at combined memories, and then breaking these down into verbatim and gist recollections to see if there are any differences. After analyzing the results, implications are reviewed and recommendations are made for this initial look into memories for pictorial representations.














CHAPTER 3
METHODOLOGY

Introduction

In this study the researcher attempted to elicit information about how differing the presentation speed might affect immediate memory for the pictorial representations found on the video montage. For this portion of the study, scores for verbatim and gist memories were pooled together. The researcher then investigated whether the same pooled scores differed between males and females taking the test. Next, the scores for verbatim and gist memories were segregated and reviewed in light of the differing presentation speeds. It is recognized that gist memory is a less exact indication of immediate memory about individual pictures than is verbatim memory. However, it was anticipated that a review of gist memory might provide additional overall insights to memory in that it requires viewers to draw certain conclusions about the contextual meanings much like a typical reading comprehension test of text-based materials. Lastly, the researcher investigated the effect cognitive style might have on the verbatim scores and gist scores taken together, as well as for each of them individually.

A computerized instrument was specifically prepared for this study that displayed the video at one of three different presentation speeds and then immediately tested subjects to determine what they remembered from the video. The program was encoded to score verbatim and gist memory scores separately and also combined into a single






81





82


score (referred to henceforth as the total score or total combined score) in order that comparisons may be made for speed, gender, and style.

Verbatim memory was tested in two ways. First, subjects were asked in a series of questions to pick out from groups of four the one picture that appeared in the video. They were then asked to identify the picture in each of a series of four still images that did not appear on the video. In order to test whether the subjects were able to comprehend (i.e., get the gist of) contextual meaning conveyed by the video, subjects were asked to place pictures in chronological sequence, to identify by way of multiple choice questions some of the major techniques used in the video to project moods and attitudes, and to answer some questions about the circumstances (i.e., storyline) presented in the video.

Population

The subjects for this study came from a population of ninth grade students in a high school in North Central Florida. Ninth graders were selected due to their closeness in age to the norms developed for the instrument selected to categorize cognitive style so that some historical comparisons may be made. That instrument provided normative data, which was very useful in helping to categorize and compare the current group of students as to their impulsive-reflective tendencies and current trends.

Sampl

The overall sample set consisted of 204 subjects. The entire sample was used in the analyses to determine the effect of presentation speed on combined test scores as well as verbatim and gist scores taken separately. The entire sample was also used to investigate differences between males and females. However, the nature of the instrument used to categorize this sample set into one of two cognitive styles dictated the creation of





83


a subset of subjects (n= 129) from the original sample set of subjects (n-7204) for the second portion of the study, as demonstrated in Figure 1. In the identification of subjects as to being either impulsive or reflective, a resulted in a certain portion of participants


Impulsive Slow -Inaccurate
N=65 N=37


Fast-Accurate Reflective
N=38 N=64

Figure 1
Subjects Included in the MFFT-20 Cognitive Style Test


did not qualify as being in either category. As per the instructions provided with the test instrument, these outliers (i.e., the fast-accurates and slow-inaccurates) were excluded from this portion of analysis to determine the effects of cognitive style on immediate memory. Subjects were categorized by cognitive style as determined by a computerized program specifically designed for this study. The program scored the subjects on each test question according to latency to first choice and the total number of errors. The overall sample and the subset were both randomly placed into one of three groups and shown the video at one of the three different presentation speeds.

Instrumentation

To categorize the subjects by cognitive style, a 20-itemn version (MFFT-20) of the Multiple Familiar Figures Test (MFFT) developed by Cairns and Cammock (1984) at the University of Northern Ireland was administered. The original MFFT is an instrument developed by Jerome Kagan (1965; 1966) that was subsequently evaluated for validity and reliability and adapted over the years by several individuals (Arizmnendi, Paulsen, &





84


Domino, 198 1; Block et a]., 1974; Watkins, Lee, and Erlich, 1978) to detennine impulsive-reflective tendencies. The original forinat of the MFFT-20 was developed as a paper version in which participants made their choice of selecting the matching figure from a set of six distracters by pointing to their choice. The investigator was responsible for keeping track manually of the number of choices made and timing latency to first response using a stopwatch. For the current study, the paper copies of the figures and alternative choices that were to be matched were scanned into a computer and imported into a program written in Macromedia 8.5 that was specifically developed for this study. The program presented the pictures and their alternatives on one screen and allowed subjects to click on their selected picture to indicate their response. The computer program automatically kept track of the total number of choices made by each participant and the amount of time to first choice for each of the item sets.

As with the paper versions of the MFFT, subjects in this study were presented

with 12 sample pictures of familiar items and are then asked to identify which one of six alternatives is identical to the sample. If an error was made the subjects were subsequently asked by the computer to retry until a correct response was found. Subjects were also automatically timed as to how quickly or slowly they made their initial choices (latency) and how many total errors they made. The dividing line between impulsive and reflective was determined by calculating a median split score for both latency and total number of errors. The scores were placed into quadrants made up of two intersecting axes. Those who made very quick but inaccurate decisions ended up in a quadrant labeled 'impulsive'. Those who were more deliberate (i.e., showed an increased latency to first response) and made fewer errors than the calculated median were to be determined





85

'reflective'. Subjects found to be fast-accurate (i.e., faster and more accurate than the calculated medians) or slow-inaccurate are placed in two other cells. As per the instructions provided, subjects who landed in either one of these latter two categories were excluded from the study.

According to the literature, (Berry, 1991; Green, 1985) the impulsive-reflective test, in addition to leveling-sharpening and dependence-independence, has been one of the most commonly used and more accurate means to test for cognitive style and to show how individuals perceive and process visual patterns. Because the treatment in the current study was to show accuracy of visual processing as well as processing speed, it was determined that impulsive-reflectivity would be the most closely related of the three cognitive style classifications to use.

Another goal of this study was to determine if there might be an interaction

between a subject's reaction to rapidly-presented content and his or her cognitive styles. Rapid visual presentation has been found in several studies to increase cortical arousal. Further, impulsivity has been reported by some as a stable mediator of the rate of change in arousal states. Anderson and Revelle (1994) have demonstrated that impulsive are susceptible to attention lapses that are, in turn, "mediated by impulsivity-related phase differences in dismal arousal patterns" (p. 334). There is no one 'correct' style. There could well be a proper time and a place for being impulsive or being reflective. It was determined that a test for the interaction between cognitive style and how well subjects receive and process rapid visual presentations could result in some very interesting findings that could be quite useful in helping to determine the proper use of rapid visual presentations in the educational environment.





86


The MFFT has been subject to several attempts to refute it as a valid

categorization test (Salkind & Wright, 1977; Watkins et al., 1978). Further, lkegulu and Lkegulu (1999) found that the notion of a generalized visual processing rate may be questionable, based on the fact that there have been few repeated measurement studies to test the generalizability of the dimension. Other research has indicated that impulsivereflective designation might be best depicted on a continuous (i.e., from low to high), rather than a bi-polar scale, as reported in the impulsive-reflective array (Salkind & Wright). On the other hand, Salkind and Wright demonstrated in other studies that continuous scaling seems to contradict a basic definitional premise of a cognitive style (that is style by its very nature is bi-polar). This apparent anomaly appears to some to create a potential lack of power for the impulsive-reflective scale to be useful in accurately classifying a cognitive style. Ault, Mitchell, and Hartmann, (1967) contributed a loss of power due to Kagan's possible over-reliance on latency rather than number of errors to determine reflective versus impulsivity. The findings of Ault et al. seem to contradict Kagan's original hypothesis, that stated categorization is the result of both speed and error-rate considered together (Kagan, 1965).

In spite of these and other attempts to dispute it, Kagan's MFFT instrument has been supported in several research studies that more than reinforced its validity (Arizmendi et al., 198 1; Green, May, 1985), making a strong enough case to justify its use it in the current study. To further strengthen power of the MFFT as a categorization technique, a 20-item variation of the original instrument was chosen for the current study because it had been validated by its authors in five separate reliability tests, and who





87


strongly assert that their instrument overcame most of the objections that arose in the previous studies.

Cairns and Cammock (1984) developed the 20-item MFFT (henceforth referred to as the MFFT-20) and presented five case studies that asserted an increased reliability and accuracy in that subjects were more accurately categorized into one of the four quadrants (impulsive, reflective, fast-accurate, or slow-inaccurate). Their instrument used 20 sets of pictures (instead of the 12 in Kagan's original test) that were reduced down from an original list of 32 items that was, in turn, concatenated and prioritized by several earlier studies. The authors then ran a test on 90 subjects chosen at random. The 32 items (2 samples and 30 test items) were then paired down to a final list of 20, and chosen on the basis of their being the most commonly missed. The authors then performed four separate reliability tests of over 300 total additional subjects that developed sets of norms, and established strong correlations between order position (i.e., the order in which the picture sets are presented), error rates, and interactions between age and sex. They then reviewed the selected items for highest item-total error correlation. Subjects were classified as reflective or impulsive using the double median split criterion. The median error and latency scores were total 34 errors and 11. 7 seconds respectively. The coefficient alpha for their 30-item test was .98 for latency and .81 for errors, both at (p< .01). The authors then developed their 20-item instrument, and re-tested it in three additional reliability studies. The first study aimed at obtaining reliability coefficients using corrected correlation between split-halves given two weeks apart. This study also yielded productmoment correlations for errors and latency of .80 and .83 respectively. Using SpearmanBrown, the authors then determined that the complete 20-item. test would have





88


reliabilities of .89 for errors and .91 for latency. A third study was then conducted to investigate test-retest reliability of the MFFT-20. The results were favorably compared to the most reliable form (form F) of original MFFT for error and latency when administered to a similar age group (Egeland & Weinberg, 1976). Based on the information discovered in their studies, the authors determined that the MFFT-20 was a superior instrument with regards to reliability and validity to earlier tests.

A computerized instrument was developed by the current investigator to test

subjects' memories immediately following the presentation of a video. The video was a copy of a kinestasis film depicting a snapshot history of the United States first produced by Charles Braverman (1969). The current holder of the copyright to this video has provided permission to the current investigator to utilize a copy for use in this study (R. Wright, personal communication, March 21, 2001).

Test Descriptions

Initial Developme

Two sets of question banks for verbatim memory were developed from which the final multiple choice test items were derived. The first bank consisted of 15 sets of four multiple choice questions from which participants would be asked to first pick out the one picture in each group that actually appeared in the video. The second bank, also consisting of 15 questions, was made up of similar groupings of still images, but the participants would be asked to select the one picture in each group of four that did not appear on the video. Care was taken to insure that all pictures were presented in identical size, shape and fidelity. To develop a final set of questions concerning gist memory, several questions were written that asked viewers about the overall content and major





89


themes presented in the video. Some of these were pictorial questions and some were text-based. These three question banks (consisting of a total of 45 questions) were then presented to three social studies teachers and one English teacher who evaluated them for appropriate content, duplication of pictures, and to check for potential fidelity problems that might distinguish correct responses from distracters. The overall set of 45 questions was pared down to 25, based on the feedback from the teachers. The questions were then presented to a several student test viewers who provided additional feedback. As a result of this phase, the program was changed to randomly place the responses on the screen so as to randomize the placement of correct answers on the screen and to eliminate the opportunity for participants sitting next to another to share answers. In addition, the color schemes for the screen background and text were changed to reflect those combinations deemed more pleasing to the younger age group of participants for whom this test was designed.

Pilot Test

A pilot test was conducted with a group of students from a different high school in the area consisting of subjects of identical age and similar demographic make-up as the final test group. The pilot group consisted of 97 subjects. They were given the test in a group setting that was arranged in such a way that no one could see any on else's screens. So as to supplement the viewers' understanding of the instructions, an audio instructional track of a female reader was added to the computerized program so that the participants could both see and hear the directions on how to take the test. Headphones were used to reduce background noises and distractions. To determine reliability, two alpha reliability scale tests were run. The 15 items dealing with verbatim recall and the 10 items asking





90


gist (i.e., comprehension) questions were analyzed both together and separately. Adjustments were made to overcome any potential reliability shortcomings prior to administering the actual test to the selected subjects. A reliability co-efficient on the combined question set yielded an alpha level of .9028. The co-relational alpha tests to determine reliability of the test instrument to assess verbatim and gist memory individually resulted in co-efficient ratios of .7548 and .8662 respectively. These results indicate that the reliability of the test instrument fell within reasonable ranges of acceptability for its intended use in this study. Test Administration

To test for cognitive style, a computerized version of the MFFT-20 test was developed for use in the current study. The same pictures from the original MFFT-20 were scanned into digital form and placed on an 800 x 600 computer screen format using Director 8.5, an interactive software program developed by Macromedia. Multiple copies of an executable form of the program were made so that each subject was able to view his or her own individual screen in a lab containing 25 Dell Pentium IV and 5 Apple G-4 Computers. Controls were put in place so that no subjects were able to see another's screens. Subjects also wore headphones in order to hear the directions that were also displayed on the screens. As with the pilot study, the headphones also add the convenience of providing a more focused environment by eliminating surrounding stray noises and disruptions. The program presented two sets of sample items and 20 sets of actual pictures in the exact order as the paper version of the MFFT-20.

The computer program for both the MFFT-20 and the verbatim and gist questions automatically kept track of the responses by the participants and outputted that





91


information to text files. These files were then imported into Microsoft Excel and concatenated to consolidate the results for each test instrument. These results were then exported into an SPSS version 10.0 data file for analysis. Stimuli

Subjects were shown a video entitled American Time Capsule, a kinsestatic visual portrayal of a chronological history of the United States which is presented very rapidly through a montage of approximately 1,300 still photographs, portraits, and paintings. The original video was imported into a Macrornedia Director 8.5 program written specifically for this study in which it could be played back at one of three different speeds (fast, medium, and slow), based on a specific password that was randomly provided to each subject. A spreadsheet was developed to track which password was assigned to each subject and to ensure an equal distribution of assignments over the sample base, for each gender, and for each cognitive group. At the 'fast' rate, subjects saw the video at its original speed (averaging 300 milliseconds per picture) that lasted approximately four minutes. The 'medium' speed presented the video in about six minutes. The slowest speed presented the 1,300 pictures in about eight minutes, or one every 500-600 milliseconds. The subjects were first shown a sample portion of the video presented at the same rate of speed for the purposes of gaining their attention, accustoming them to the presentation style and speed, and providing an overview of the subject matter. The subjects were able to begin viewing the entire video when they felt they were ready by clicking a button on the screen. In order to facilitate correct interpretation of the directions supplied, the subjects were provided the same set of instructions in a small




Full Text
31
encoded into memory. Intraub reported that the process of stabilizing integration errors
can take place when pictures are presented at a rate of about three to four pictures per
second. This is roughly equivalent to the presentation rate of the pictures presented to the
fast-paced group of the current study.
Encoding
Perception might have been conceived of as a totally individual operation was it
not for the fact that there exists some type of standardized analytical coding thesis on
which to base a more stable view of the process. In reality, humans process information
by breaking perceptual units down into sub-units called symbols and icons (Neuman &
Prinz, 1990). Corcoran (1981) proposed the idea that symbols and icons comprise the
very basic subset of cognition. He referred to them as the minimal units of perceptual
analysis or that product of semiotic analysis which is the smallest interchangeable
element that still retains the basic properties of the whole symbol system but cannot be
further subdivided without losing those properties (p. 118). He identified several forms
of icons and symbols, including (but not limited to) words, pictures, and auditory
patterns. He also referred to rhythmic patterns and control as symbols of perception that
can be translated to enhance learning but he did not limit his conceptualization to
audition. He also claimed that visual rhythmicity is a form of symbolic representation that
can be learned as a cueing system for attention and recall (Corcoran).
Humans process symbols and icons on more than one plane, depending on the
structure of the input. Words, pictures, sounds, etc. are perceived and treated in similar,
but distinct ways. Paivio (1986) proposed three different procedural methods:
representational, or the direct activation of verbal (i.e., text-based) or non-verbal (i.e.,


115
differences in gist memory was masked when the verbatim and gist memory scores were
considered together reveals that researchers may need to identify new paradigms that take
into consideration the goals and intended outcomes of the instructional activities they are
investigating. Not all classroom experiences need to have rote memory as a sole learning
outcome. Stimulating gist memories, like teaching reading comprehension, may have its
own place in overall instructional schemes.
To a degree, the results of the cognitive style instrument used in this study to
categorize subjects indicate that some things about learning styles have changed since the
original instrument was analyzed and developed. When one compares the results of
MFFT-20 cognitive style in the current study to the norms provided by Cairns and
Cammock (1984), not only has the median total number of errors decreased (from 28-30
in the Cairns and Commock studies to eight in the current study), but also so has the
median latency to first response (from 18 to 9.12). These reductions seem to indicate that
latencies to first response for visual activities are growing shorter, but the quicker
responses do not always translate to higher error ratios. Students appear to be developing
a propensity for remembering things from rapid visual presentations.
Another change that took place is the shrinking of the differences in visual
cognition between males and females. With Cairns and Cammock, female responses were
considered to lie so dramatically outside of the norms that they were systematically
eliminated from their studies. In the current study, any differences between males and
females that did exist were found to be non significant. While females still may be found
to be more reflective than their male counter-parts, these differences may be growing
smaller.


9
determining the affective or emotional response of message receivers" (Seidman, 1981, p.
49). Lang et al. (1999) and Lang et al. (2000) looked into presentation speed as its own
construct, comparing/combining it to/with arousing content. They discussed the effect of
adding interesting and arousing content as having a positive effect on cortical arousal,
and therefore, recall and recognition. Lang et al. (2000) also alluded to future research
that should continue to probe the shape of the relationship between presentation rate and
recall and test even faster rates of edits to determine whether there is a point at which
memory begins to decline. They suggested that producers who want their messages to be
remembered should create arousing messages that are presented slowly or at medium
speed, or calm messages that are presented at medium or rapid speed. They concluded
that producers should not create messages that are either calm and slow or arousing and
fast.
The effect of integrating increased message presentation speed directly into
instructional messages in an educational environment whose content might be considered
by some students as less than arousing will be examined in the current study. It should
also be noted that Lang et al. (2000) considered cuts to be fast if they changed at a top
rate of eleven to twelve per thirty-second segment (i.e., one every 2-3 seconds). The
current study looks at a top rate of one cut every one-third to one-half of a second, or
roughly ten times as fast. The slow messages will run roughly equivalent to Lang's top
rate.
It should also be noted that the term cuts. as it was used in connection with
commercial television research, differs from what how it was used in the current study. In
those studies, a cut might simply mean a wipe or a swipe, or a change from one camera


56
determined, for example, to hinder learning performance. Conversely, Boyden and Gilpin
(1978) found latency and error rates to be independent of measures of distractibility. The
relationship between impulsivity and academic achievement is not necessarily tied to
aptitude or intelligence but to ones ability to attenuate and/or process specific types of
inputs (Kogan, 1971; Leino, 1981; Messer, 1970). Cooper (1982) suggested that
differences in processing speeds may well be accurate indicators of ones ability to
process mostly visual information. Ridberg et al. (1970) found that cognitive style is
predictive of performance in a variety of measurement tools, including those for reading
recognition, secondary learning, and reasoning. Hedberg and McNamara (1985) found
that when using visual information the tendency to reflection or impulsivity might be an
important predictor of performance, particularly in relation to time and error under
conditions of response uncertainly and time pressure. Merrinboer (1990) was able to use
the classification as a predictor of academic performance and use it to pre-arrange
feedback strategies to increase effective computer usage, especially in younger students.
It appears as though impulsivity is a characteristic that mitigates with age. Wright
(1979) found that impulsivity lessens over time as one grows into adulthood. Okun et al.
(1979) found that adults have significantly higher latencies than middle-aged and
younger adults. In other words, as one grows older an individual tends to become more
logical in his or her outlook, incorporating familiarization and planning into perceptual
processes. However, an increased ability to exert a reflective response to stimuli can also
work in reverse. OBrien (1968) discussed effects of becoming stimulus bound,
especially when exposed to substandard information. This phenomenon, which occurs
mostly in adults, causes individuals to so over-structure their internal response that they


47
Sheingolds (1973) work further complicates the ability to unwind these
contrasting views of resource allocation by showing that the age and sex of the viewers
tend to confound things. Lang et al. (2000) later tried to unravel these variables by
eliminating content as one of them. Using a measuring device that calculated skin
conductance and heart rate, she gauged secondary task reaction times during several
different studies and came to the conclusion that the limited capacity model most
accurately reflects what occurs (at least in a casual viewing environment) with the
following two hypotheses:
1. Arousing content is likely to elicit feelings of arousal and measurable
sympathetic nervous system activation. As pacing increases, viewers
encode more and, therefore, recognize more of specific content.
2. Arousal results in the automatic allocation of processing resources to
encoding and storage.
Langs conclusions suggest that media producers who want their messages to be
remembered should create arousing messages that are presented slowly or at medium
speed, or calm messages that are presented at medium or fast speed. These producers
should not create messages that are at the same time calm and slow or arousing and fast.
While Langs work has been a watershed for casual television viewing, questions remain
as to how much of their work relates to student viewers and in an educational
environment where content might simply be considered boring. In spite of these
reservations, the limited capacity model was used to develop a portion of the hypotheses
for the current study.
The Hueristics of Pictures
To the extent that processing of viewed information is controlled by highly
varying individual past experiences, extracting information also varies greatly. While


66
information increases with the increase speed, short-term memory limitations are soon
reached and information is lost (Lang et al., 2000). The fact that broadcast medium
presents information linearly and does not permit a child to return and review content
puts the onus on the producer to utilize commonly recognized techniques such as
maintain scenes for longer periods, to limit the objectives of the program, slow its speed,
and/or to use slow motion. Kozma points out that shows like Sesame Street accommodate
these limitations by constraining the objectives of each episode and representing
information in different formats. However, simply slowing the speed down may not be
the total solution either. Cronbach and Snow (1977) point out that viewers who are
familiar with the content might actually need a faster speed to maintain their attention.
Salcedo (1985) looked at such features as focal length and found that close-ups
were regarded as more intimate emotionally by viewers. Other features elicited similar
varied impressions (Barnett, 2000; O'Brien, 1968): time on screen (the longer characters
appeared on screen, the better impression they left on viewers), types of transitions (fades
are considered to be more positive or emotional than cuts), and slow motion among them.
In addition to its presentation speed, the video used in the current study utilizes several of
the above features to increase comprehensibility and emotional involvement of its
viewers.
Symbol Systems
Symbol systems differ along several dimensions, which Salomon (1979; 1994)
believed are to be primarily those of notationality, repleteness (density) and resemblance.
A notational symbol system is one in which there is a strong, clear, and consistent
correlation between a set of symbols and a set of objects or concepts, such as a written


ANOVA (verbatim and gist) x (fast, medium, and slow). The design for this analysis is
shown in Table 2.
94
Table 1
2x3 Two-way ANOVA (204 Subjects)
PRESENTATION SPEED
GENDER
Fast
Medium
Slow
Male
(N= 107)
34
38
35
Female
(N = 97)
41
31
25
Table 2
Multivariate Analysis of Variance (204 Subjects)
PRESENTATION SPEED
QUESTION TYPE
Fast
Medium
Slow
VERBATIM vs GIST
75
69
60
Last, the procedures for administering the MFFT-20 specified that a certain
number of the subjects, by definition, would be systematically excluded from analysis.
The administration of the cognitive style test yielded a subset of the sample of 129
subjects who were categorized as either impulsive (n=64) or reflective (n=66) and to
Table 3
2x3 Two-way ANOVA (129 Subjects)
PRESENTATION SPEED
STYLE
Fast
Medium
Slow
Impulsive
(N = 64)
22
25
17
Reflective
(N = 66)
24
21
20


117
fast-accurates between the Caims and Cammock studies and the current
study, there appears to be a general lowering of the latencies and errors and
a general increase in the number of fast-accurates as a percentage of the
total sample. Further investigations could be developed to determine what,
if any, the increased usage of rapidly presented digital media that was
found in the review of the literature may have on cognitive style.
2. Other trends between the Caims and Cammock study and the current one
appear to have developed. For one, females were excluded from the Caims
and Cammock study due a determination that the differences between
males and females were significant enough at that time that including them
would confound the results. In the current study, the differences between
males and females were not significant. An investigation could be made to
determine whether the differences in cognitive styles between males and
females have lessened with the latters increased usage of digital media.
3. One could investigate the optimal duration of rapidly presented frames
before verbatim and gist memories deteriorate to the point where
presentation speed has a significant affect.
4. One could look into the effect of other variations of presentation speeds on
memory. The current study only looked into rapid presentation. There have
been several studies that have looked with limited success into the effects
of slow motion on long-term memory (Corcoran, 1981; Kozma, 1986;
O'Brien, 1968; Olson, 1974; Pearl, 1982). Some of these studies are quite
old and modem digital creation techniques make it much easier to design


65
concrete, visible sources relating their experiences and concerns does exert an influence
on issue perception (Zillman & Brosius, p. 110). They go on further by stating that the
incidental use of image exemplars that add non-redundant, specific information to the
text of a news report does influence issue perception. Specifically, the incidental nature
of a pictorial supplement to a story goes unrecognized and, as a result, is integrated with
the narrative information in fostering perceptions and judgment. Over time, if presented
enough times, the photographs are actually remembered longer and the retained image of
the incident was of the photograph, not of the text. Furthermore, if the image that
supplements a news story presents a negative or positive aspect, that image can actually
influence ones value judgments. This is especially so if the presentation speed of that
image is varied (Barnett, 2000; Corcoran, 1981; O'Brien, 1968; Pearl, 1982).
However, presentation speed has not always been consistently viewed as a
positive media effect. Pearl (1982), for example, talked of television being a magnet on
viewers that "shapes their behavior" (p. 1). In an attempt to investigate the ties to violence
and television viewing she found herself in conflict with her own research. First she
claimed that "the rapid form of presentation characterizing American television in which
novelty piles upon novelty in short sequences may well be counterproductive for
organized and effective learning sequences" (Pearl, p.20). But she then goes on to state
that this may be more of a problem with younger children. Kozma (1986) appears to
support this view. He points out that a fast tempo does not necessarily sustain visual
attention, and that an increased tempo is not likely to provide enough time for the viewer,
especially one inexperienced with television formats, to extract that which is needed
bring it into working memory. In addition, others have indicated that as the amount of


88
reliabilities of .89 for errors and .91 for latency. A third study was then conducted to
investigate test-retest reliability of the MFFT-20. The results were favorably compared to
the most reliable form (form F) of original MFFT for error and latency when
administered to a similar age group (Egeland & Weinberg, 1976). Based on the
information discovered in their studies, the authors determined that the MFFT-20 was a
superior instrument with regards to reliability and validity to earlier tests.
A computerized instrument was developed by the current investigator to test
subjects memories immediately following the presentation of a video. The video was a
copy of a kinestasis film depicting a snapshot history of the United States first produced
by Charles Braverman (1969). The current holder of the copyright to this video has
provided permission to the current investigator to utilize a copy for use in this study (R.
Wright, personal communication, March 21, 2001).
Test Descriptions
Initial Development
Two sets of question banks for verbatim memory were developed from which the
final multiple choice test items were derived. The first bank consisted of 15 sets of four
multiple choice questions from which participants would be asked to first pick out the
one picture in each group that actually appeared in the video. The second bank, also
consisting of 15 questions, was made up of similar groupings of still images, but the
participants would be asked to select the one picture in each group of four that did not
appear on the video. Care was taken to insure that all pictures were presented in identical
size, shape and fidelity. To develop a final set of questions concerning gist memory,
several questions were written that asked viewers about the overall content and major


53
of images displayed for 200 to 300 and milliseconds, which equates to approximately 1/3
of a second, and those displayed for more than a second. Optimal retention occurred
between 200-300 milliseconds and one to two seconds. Conversely, in other studies, it
was found that images displayed for more than one to two seconds actually caused a
decrease in cognition due to loss of attention (Thorson & Lang, 1992). It might be that
duration becomes a variable of lesser import as the size of the learning set increases and
the number of categories decreases (Keller, p.82). In short, this research appears to
indicate that longer displays do not necessarily signify a relative increase in recall, and
can actually cause degeneration. In the current study, images appear on screen for as little
as 1/3 of a second to a maximum of one second, with the majority appearing for less than
1/2 of a second, well within the range indicated by Potter and Levy (Keller) and
Goodglass.
How Learner Attributes Affect What is Learned
Cognitive style
The concept of cognitive style has endured for more than fifty years. It came out
of the New Look movement in perception that was bom during a symposium sponsored
by the American Psychologist Association held in New York in 1949 (Witkin &
Goodenough, 1981). Participants in the New Look movement were a loose confederation
of psychologists who became critical of the dominant approaches to perception then in
vogue. Their main criticism was that most current approaches tended to ignore the person
doing the perceiving. Out of that symposium came a flood of new studies aimed at
looking at the personality traits of individuals (i.e., individual differences) during the


58
conditions (Bridgeman, 1980). For sure, the impulsivity label was considered powerful
by some (Kagan, 1965; Kagan, 1966) and appealing by others (Hedberg & McNamara,
1985). However, some researchers (Ayabe, 1973; Bridgeman, 1980; Campbell & Davis,
1982) found that some weakness exists in the outcome of using classification as it relates
to discriminant validity and generalizability. In other words, there appears that some
evidence of loss of power, due to a corresponding loss of independence of traits in the
subjects tested in certain studies. The criterion used to justify discriminate validity in the
identification of a classification method or methods, which measure a discrete trait to the
exclusion of others (Campbell & Fiske, 1959). An additional problem in using this
classification is that it appears to be more heavily weighed down by the errors subjects
made than latency itself (Ayabe; Block et al., 1974). Further, there is an indication that
many studies failed to yield consistent findings in an ecological (i.e., classroom) setting
due to inconsistencies with the exact definition of impulsivity (Campbell & Davis).
Lastly, previous studies systematically excluded females. Some researchers went
so far as to suggest that gender differences were so extreme that they would skew the
results (Caimes & Cammock, 1984). The current study looked at gender differences to
see if the same rationale is still justified.
Factors From Film and Television That May Affect Learning
Preconceived Mental Demands
Salomons (1984) with mental effort appears to confirm the idea that learning that
is or is not realized by a medium can be influenced by ones previous experiences with
that medium by coloring student preconceptions about the mental demands being placed
on them by that medium. In his studies, students reported that they exerted more effort


113
When considering gist memory, the means table indicates that the scores
improved as the presentation speed got faster. The analysis of variance reveals that these
differences were significant at the .05 level. Because there were three presentation groups
being compared, a follow-up test was run to pin-point where the significant differences
existed. The Bonferroni test shows that the significant difference between test scores
exists between the fasted and the slowest presentation speeds, while the differences
between the other two groups (fast-medium and medium-slow) were not significant. The
difference only mattered when one compares the two extremes of the presentation speeds.
Cognitive Style
By definition, reflective individuals were expected to score better on tests than
impulsives. The means and standard deviation table used confirms that this assumption
held true in this study as well. Reflective subjects did score significantly better in the
overall test scores in which the verbatim and gist questions were combined. However, in
a separate analysis in which the verbatim and gist scores were viewed separately, the
differences for verbatim memory were found to be not significant at the (p>.05) level. On
the other hand, there was a significant difference in gist memory when considering
cognitive style.
The mean number of overall correct responses for impulsive subjects was 11.92
out of a possible 25 (approximately 48%). For reflective subjects, the overall mean was
13.29 correct responses out of 25 questions (approximately 53%). The descriptive
statistics are more revealing when verbatim and gist scores are separated. When looking
at the verbatim scores alone, impulsive subjects scored an average of 6.77 out of 15
possible verbatim responses (approximately 45%). Reflective subjects correctly answered


97
The next chapter will report the results of these tests. The statistics will be
presented in both narrative and table form, leading to the subsequent chapters where the
results will be analyzed and discussed with regards to the kinds of inferences that may be
made.


128
Cronbach, L. J. & Snow, R. E. (1977). Aptitudes and instructional methods: A handbook
of research on interactions. New York: Irvington.
Crowder, R. G. (1976). Principles of learning and memory. Hillsdale, NJ: Lawrence
Erlbaum Associates.
Dale, E. (1969). Audiovisual methods in teaching. New York: Holt, Rinehart & Winston.
Davis, E. T., Scott, K., Pair, J., Hodges, L.F., & Oliverio, J. (1999). Can audio enhance
visual perception and performance in a virtual environment? Paper presented at
the Human Factors and Ergonomics Society-43rd Annual Meeting, Los Angeles,
CA.
Dick, A. O. (1973). Utilization of visually presented information: Final Report.
Washington, DC: National Center for Educational Research and Development.
(ERIC Document Reproduction Service No. ED 078 365).
Dixon, N. F. (1981). Pre-conscious processing. New York: John Wiley & Sons.
Downs, E. (1989). The effects of cueing strategy, level of information, and motion
condition on children's interpretation of implied motion in pictures. (Doctoral
Dissertation, University of Florida, 1989). Dissertation Abstracts International,
51 (03), 0792A.
Durso, F. T. & Johnson, M. K. (1979). Facilitation in naming and categorizing repeated
pictures and words. Journal of Experimental Psychology: Human Learning and
Memory, 5(5), 449-459.
d'Ydewalle, G., & Vanderbeeken, M. (1990). Perceptual and cognitive processing of
editing rules in film. In R. Groner, d'Ydewalle, G., Parham, R. (Ed.), From Eye to
Mind: Information Acquisition in Perception, Search, and Reading (pp. 129-139).
Amsterdam: Elsevier Science Publishers, B.V.
Ederlyi, M. H. (1985). Psychoanalysis: Freud's cognitive psychology. New York: W. H.
Freeman.
Edgar, G. K. (1997). Visual accommodation and virtual images: Do attentional factors
mediate interacting effects of perceived distance, mental workload, and stimulus
presentation modality? Human Factors, 59(3), 374-382.
Egeland, B., & Weinberg, R. A. (1976). The matching familiar figures test: A look at its
psychometric credibility. Child Development, 47, 483-491.
Ellis, W. D. (1938). A source book of gestalt psychology. New York: Harcourt, Brace &
World.


139
Worth, S., & Adair, J. (1997). Through Navajo eyes: An exploration in film
communication and anthropology (2nd Edition). Albuquerque: University of New
Mexico Press.
Wright, J. C. (1979). Changes in conceptual tempo over time: Developmental versus
logical familiarization. Paper presented at the Bicentennial Meeting of the Society
for Research in Child Development. March 15-18, San Francisco.
Yerkes, R. M., & Dodson, J. D. (1908). The relationship of strength of stimulus to
rapidity of habit formation. Journal of Neurological Psychology, 18, 459-482.
Zillman, D. (1991). Television viewing and physiological arousal. In J. Z. Bryant, D.
(Ed.), Responding to the screen: Reception and reaction processes (pp. 103-133).
Hillsdale, NJ: Lawrence Erlbaum Associates.
Zillman, D. & Brosius, H. (2000). Exemplification in communication. Mahwah, NJ:
Lawrence Erlbaum Associates.


63
a scriptwriter shows a man getting into his car and driving out of the driveway. Then in
the next scene he is on the highway on the outskirts of a big city. Finally, he pulls up into
his parking space at work. Enough information has been presented to allow viewers to
imagine that time has passed and the man has driven all the way to work and are able to
fill in the information in between. Bryant and Rockwell (1991) found that viewers well-
versed in television and/or film techniques can fill in gaps of omitted information, even
after missing as much as five or more minutes of a television show simply because they
are familiar with the repetitive formats used by screenwriters. Abelman (1995) showed
that viewers can learn how to build an internal concept map for the most often-used
devices in films and teleplays (including time leaps, fall backs, dream sequences, scenes
showing characters recalling past events, etc.) even without comparable real-world
experiences. This appears to further reinforce the theory that television viewing is a
learned activity and highlights the inter-relatedness of childrens linguistic, cognitive, and
perceptual skills for accurate comprehension of televisions most basic narrative device:
temporal sequencing. Corcoran (1981) suggested that intelligence may be defined as a
skill in a particular medium and that symbolic codes that serve communication purposes
can be internalized by a receiver to serve in a cognitive capacity. In other words,
filmed/televised techniques such as zooming, slow motion, or rapid montage may
actually be thought of as tools of thought. Viewers appear to learn how to use these tools
naturally without being trained, because these types of story-building and post-production
editing devices are used redundantly in one show after another (Abelman; Amheim,
1974; Corcoran; Jankowski & Fuchs, 1995; Zillman & Brosius).


108
identified the source of the significant variations. Of the two types of memory tested, the
gist comparison proved to be significant, whereas the differences between reflective
subjects on the verbatim portion of the memory test were found to be random variations
at the .05 level.
The role of this chapter was to report on the specific statistics as they related to
the current study. The results of the tests show that many of the variances that were
uncovered were not significant at the .05 level. However, two of the comparison tests (the
one contrasting gist memories and presentation speed and the one that evaluated
cognitive style and presentation speed) revealed significant differences. The impact of
these variations and considerations as to how they relate to the current and potential
future studies will be discussed in the next chapter.


55
(1965) with cognitive tempo (i.e., rate of cognition) in younger aged children in which
their speed and attention to detail was indexed. Children deemed to be impulsive tend to
react very quickly and make quick decisions (i.e., they select the first answer that occurs
even though it may be wrong), while reflectives tend to take more time to consider
various options but are also generally more accurate with their interpretations. Contrary
to many stereotypes about bright children thinking quickly, Kagan (1966) found that
neither tendencies for fast or slow decision times were significantly related to verbal
ability or innate intelligence. According to Block, Block, and Harrington, (1974) some
have criticized Kagans findings, noting that although response times were positively
correlated with performance, overall, the correlations were often quite low (ranging from
close to zero to around .45). However, according to others (Ayabe, 1973; Bridgeman,
1980) theirs and others research appeared to ignore the converse of the measurement
scale that compares slow-accurate to fast-accurate, which states that accuracy and not
speed counts the most in complex problem-solving situations. Bridgeman points out that,
even with timed tests, the negative correlation between speed and accuracy can also be a
plus, suggesting that there are many children who are fast and accurate (p .212). The
problem, then, appears to be taking an extremely bi-polar classification and make
subjective determinations in overall psychological and/or personality evaluations that
may require less polarization and more sub-categorization. These classifications (or close
derivations thereof) have been used quite successfully and deemed to be valid
classifications in specific educational evaluations. Campbell and Davis (1982) found that
the reflection-impulsivity style construct emerges as an ecologically valid and
parsimonious descriptor of a component of student behavior (p. 8) where it is


87
strongly assert that their instrument overcame most of the objections that arose in the
previous studies.
Caims and Cammock (1984) developed the 20-item MFFT (henceforth referred to
as the MFFT-20) and presented five case studies that asserted an increased reliability and
accuracy in that subjects were more accurately categorized into one of the four quadrants
(impulsive, reflective, fast-accurate, or slow-inaccurate). Their instrument used 20 sets of
pictures (instead of the 12 in Kagans original test) that were reduced down from an
original list of 32 items that was, in turn, concatenated and prioritized by several earlier
studies. The authors then ran a test on 90 subjects chosen at random. The 32 items (2
samples and 30 test items) were then paired down to a final list of 20, and chosen on the
basis of their being the most commonly missed. The authors then performed four separate
reliability tests of over 300 total additional subjects that developed sets of norms, and
established strong correlations between order position (i.e., the order in which the picture
sets are presented), error rates, and interactions between age and sex. They then reviewed
the selected items for highest item-total error correlation. Subjects were classified as
reflective or impulsive using the double median split criterion. The median error and
latency scores were total 34 errors and 11.7 seconds respectively. The coefficient alpha
for their 30-item test was .98 for latency and .81 for errors, both at (p< .01). The authors
then developed their 20-item instrument, and re-tested it in three additional reliability
studies. The first study aimed at obtaining reliability coefficients using corrected
correlation between split-halves given two weeks apart. This study also yielded product-
moment correlations for errors and latency of .80 and .83 respectively. Using Spearman-
Brown, the authors then determined that the complete 20-item test would have


64
Pearl (1982) discussed the relationship between form and content and admitted
that her research showed that it is the form (that is, the way it uses verbal and linguistic
codes), not the content of television that is unique. However, she also cautioned that form
and content cannot always be distinguished -no more than grammar and meaning in
any verbal language can (p. 24). Like grammar, some forms are unique to the medium
and apply syntactical meaning only in the context of that medium. For example, slow
motion is not real and its meaning must be learned. But once learned, these formats are
generally used by people in their own thinking. Pearl outlined a taxonomy for this coding
structure of formal features. For example, entropy or form complexity is a combination of
the number of different scenes in a show, the number of characters, and the number of
times the scene and characters appear. Dynamism is the rate of change in scenes and
characters. In this view, content variation is merely determining which features appear
together, their tempo or rate, and/or their variability within scenes.
Unfortunately, the link back to attribution or lack of perceived mental demand has
been caused by television producers who tend to use these same technical features over
and over again as a means of convenience in an entertainment industry that tends to value
volume over quality (Jankowski & Fuchs, 1995). However, redundancy of format and
content can also play a role in causing viewers to make value judgments about what they
see. Zillman and Brosius (2000) cite studies showing that an increased number of
occurrences sometimes influences viewers perception of the news story. For example, if
the number of occurrences of a particular story in a newscast is increased, viewers tend to
be influenced by the story in that the event was probably more significant than it actually
was. (i.e., it tends to slant the news). There is strong evidence that the sheer number of


110
Findings
All hypotheses are stated in the operational null form and were tested at the .05
level.
Hypothesis 1. There are no significant differences in overall test scores (both
verbatim and gist scores are pooled together), for the overall sample set based on
presentation speed of the video.
The null hypothesis was not rejected.
Hypothesis 2. There are no significant differences in overall test scores between
male and female subjects in the overall sample set.
The null hypothesis was not rejected.
Hypothesis 3. There is no significant interaction between presentation speed and
gender for overall test scores.
The null hypothesis was not rejected.
Hypothesis 4. There are no significant differences in verbatim test scores, for the
overall sample set, based on presentation speed of the video.
This hypothesis was not rejected.
Hypothesis 5. There are no significant differences in gist test scores for the overall
sample set, based on presentation speed of the video.
The null hypothesis was rejected. Further, a Bonferroni post-hoc test indicates that those
watching the video at the fastest speed scored significantly better on the gist portion of
the test than did those watching the video at the slow speed.


71
commercial television. Market and psychological research (Reeves & Nass, 1996;
Stephens, 1996) has demonstrated its effectiveness on memory in commercials and
marketing campaigns. Further, researchers (Reeves et al., 1986; Reeves & Nass) appear
to concur with Olsons assertions in light of a rapidly changing technological arena.
Rapid changes are also in process for the way in which researchers view media
assessment. Several things gave rise to an increased interest in looking at the
psychological and sociological aspects of television viewing during the late 1980s. An
evaluation of the then-current research by Reeves et al. found that there was an increased
understanding about attention. First, was the fact that recognition was not always present,
and second, and most important, that continued attention was not always a necessary
prerequisite for learning and understanding. Until that time, most evaluation of attention
was limited to measuring direct eye contact. However, Reeves et al. discovered that due
to increased changes in visual displays, with their rapid scene changes, quick movements,
and zooming, that viewers appeared to have developed a sense of being able to fdter
stimuli and, in essence multi-process. At the same time, much attention was paid to the
potential harm this newer, faster paced television was having on its younger viewers
(increased proclivity for violence, poor attitudes and moral judgment, etc.). The question
Reeves et al. began to look into was why these same effects could not be implemented
into instructional programs in order to accomplish increased educational goals. The
answer may or may not be obvious and requires that one take a deeper look at the aspects
of the new characteristics of message delivery brought on by newer technological
developments. In other words, attention is only one of the variables. Reeves et al. also
found that earlier studies to determine which program attributes caused children to attend


demands on their viewers. Yet these viewers have been found to be able to perceive and
remember content from these messages.
The purpose of the study was to investigate whether youthful viewers are able to
perceive video images that are presented very rapidly and then immediately remember
verbatim information and assimilate the contextual gist of the overall passages. The study
also attempted to determine whether or not there are differences in the way males and
females process these images. In addition, subjects were categorized as being either
impulsive or reflective to determine whether cognitive style has an effect on pictorial
memory.
The study revealed that memory is not significantly affected by changes in
presentation speed when test results were viewed as a composite of verbatim and gist
scores added together. However, when the two scores were separated, gist memories
were positively affected by increases in presentation speed. The study also revealed that
cognitive style had a significant impact on pictorial memory.
IX


20
and other sound (Corcoran, 1981). Much of the literature presented in Chapter 2 refers to
screen media. For purposes of this study screen media and video media are considered to
be identical.
Verbatim/rote memory refers to subjects being able to identify whether specific
objects within a scene, or overall scenes occurred in the video that has just been seen
(Brainerd & Gordon, 1994; Reyna & Kieman, 1994). In some studies, verbatim memory
also has been referred to as object recognition (Tse, Vegh, Marchionini, & Shneiderman,
1999). For purposes of this study, verbatim memory is assessed positively and falsely.
Subjects were asked to select an image that was presented in the video display (positive
identification) from a group of pictures immediately after presentation of the test video.
Subjects were also are asked to decide whether any picture or series of pictures in a group
actually appeared in the video (false identification).
Summary
Meyrowitz (1985) posited the idea that new media, through its ability to merge
many formerly distinct knowledge situations, appear to be breaking down the
boundaries among various disciplines, opening new dialogues, and fostering the
development of cross-disciplinary areas of study (p. 327). He also speculated that these
new kinds of electronic media may be "introducing our children to a different way of
thinking that involves the integration of multiple variables and overlapping lines of
simultaneous actions" (p. 326). This assertion proposes, among other things, that
electronic video media have already greatly reduced the influence that time and location
used to have on what people know. It is not, therefore, unreasonable to agree with
Stephens (1996) proposal that a transition may be occurring in the way the youth of


39
intentional versus incidental learning that shows that recall improves when the subjects
are warned of an ultimate memory test, where recognition does not change significantly
and may even decrease. According to Estes (1994), recognition, might be better described
as two distinct processes: absolute and recognition in context. The former relates to
whether a subject recognizes a particular person, or he or she remembers ever
encountering that person before. The current study attempts to elaborate on Estes work,
comparing the ability to remember verbatim facts with subjects ability to remember
contextual (i.e., gist) sequences. Estes further suggests that recognition does not in
general, provide a direct window to memory (p. 230). Rather, it may be said that
recognition provides the best available basis for estimating memory storage, as long as
interpretations of data are guided by appropriate models (p.230). It is the problems
associated with these model interpretations where much of the on-going controversy
about the recognition-recall debate appears to reside.
Verbatim versus Gist Memory
The apparently unresolved confusion with regards to the use of the terms
recognition and recall led to the need to find another means to assess immediate memory
in the current study. A further review of the literature led to a discovery of relatively new
research on two complementary memory processes that have proven useful to the current
study. The foundation of these newer ideas about describing memory in terms of a dual
relationship in a bi-polar representational system lies in Brainerd and Reynas (1990)
fuzzy-trace theory. Fuzzy trace is consistent with many of the ideas about childhood
memories but runs counter to Piagets developmental/structuralist approach that uses the
mind as a computer comparison (Miller & Bjorkland, 1998). Brainerd and Reynas


CHAPTER 4
ANALYSIS OF THE DATA
Introduction
This study had four overall purposes. The first was to determine whether
differences in presentation speed of a video montage would have an effect on a memory.
Second, this study attempted to determine whether the presentation speed affect males
and females differently. A two-way analysis of variance was used in these two portions
of the study. Third, the study attempted to determine whether there would be differences
between immediate verbatim and gist memory for the content of the video caused by
changes in presentation speed. The design of this portion of the study was a multivariate
analysis of variance. Last, an investigation was made to determine whether the immediate
memories for subjects classified as either impulsive or reflective would be influenced by
changes in presentation speed. The design of this portion of the study was a 2 x 3
factorial design ANOVA.
Results
The data were analyzed with analyses of variance and a multivariate ANOVA.
The relevant statistics are presented in ANOVA summary tables. In addition, descriptive
statistics are included for means and standard deviations.
Hypothesis 1. There are no significant differences in overall test scores (in which
verbatim and gist scores are pooled together), for the overall sample set based on
presentation speed of the video.
98


52
awareness of larger spans of information, and processing rates can be accelerated without
loss of comprehension.
Zillman (1991) showed that the tempo of a message can be influenced by more
than its structural or production process. Message content, for example, may have a
carry-over effect. He showed that the interspersion of fast-paced humor is often mediated
by cortical arousal. Semi-attentive children in the audience may be made alert by the
inserted humorous tidbits, and this alertness may have carried over into exposure to the
immediately subsequent educational material (Zillman, p. 126). These content studies
were not limited to humor. Other studies looked at attentional reflexes caused by
fireworks (Reeves & Nass, 1996; Zillman, 1991), and negative or violent content (Pearl,
1982; Reeves et al., 1986; Reeves & Nass, 1996). Hill and Lang (1993) used a
combination of image complexity and the number of structural or formal features in a
message to manipulate effects of pacing. Other studies looked into the interaction
between pacing and content. In his studies into the levels of arousal found in fast cuts in
music videos, Hitchon et al. (1994) found that fast-paced video montage did not
contribute negatively to the complexity or ambiguity issues as long as the story line was
not confusing or ambiguous. Theoretically speaking, if increasing the speed of message
delivery does not necessarily cause attentional overload, one might be able to make a
stronger case by looking at a corollary question (i.e., does the slowing of the pace that an
image is presented necessarily increase its retention?). In video productions, presentation
speed has a direct relationship to the number of frames an image appears, with 30 frames
per second being the standard (Stephens, 1996). Potter and Levy (Keller, 1976) and
others (Goodglass, 1971) found a significant correlation between retention and duration


TABLE OF CONTENTS
page
ACKNOWLDEGMENTS
LIST OF TABLES vii
ABSTRACT viii
CHAPTERS
1 INTRODUCTION 1
Introduction 1
Statement of the Problem 2
Need for the Study 4
Background 11
Purpose 12
Research Questions 15
Delimitations of the Study 16
Limitations of the Study 16
Definition of Terms 17
Summary 20
2 REVIEW OF THE LITERATURE 22
Perception and Knowledge 22
Neurology of Vision 24
Visual Perception Theories 25
Fast Seeing 28
Encoding 31
Relationship of Memory and Learning 35
Textual versus Pictorial Memory 36
Recognition and Recall 37
Verbatim versus Gist Memory 39
The Effect of Arousal on Memory 42
The Heuristics of Pictures 47
Pacing versus Presentation Speed 50
How Learner Attributes Affect What is Learned 53
IV


135
Pezdek, K. (1977). The cross-modality semantic integration of sentence and picture
memory. Experimental Psychology: Human Learning and Memory, 3, 515-524.
Plass, J. L., Chun, D. M, & Mayer, R. E. (1998). Supporting visual and verbal learning
preferences in a second-language multimedia learning environment. Journal of
Educational Psychology, 90(1), 25-36.
Posner, M. I. (1982). Cummulative development of attentional theory. American
Psychologist, 37, 168-179.
Postman, N. (1986). Amusing ourselves to death: Public discourse in the age of show
business. London: Heineman.
Ramachandran, V. S. (1994). 2-D or not 2-D: That is the question. In H. R.I. Gregory, J.
(Ed.), The Artful Eye (pp. 249-267). Oxford, England: Oxford University Press.
Reeves, B., Thorson, E. & Schleuder, J. (1986). Attention to television: Psychological
theories and chronometric measures. In J. Bryant, & D. Zillman (Ed.),
Perspectives on media effects (pp. 251-279). Hillsdale, NJ: Lawrence Erlbaum
Associates.
Reeves, B., & Geiger, S. (1994). Designing experiments that assess psychological
responses to media messages. In A. Lang (Ed.), Measuring Psychological
Responses to Media (pp. 243). Hillsdale, NJ: Lawrence Erlbaum Associates.
Reeves, B., & Nass, C. (1996). The media equation: How people treat computers,
television, and new media like real people and places. New York: CSLI
Publications.
Rettenbach, R. (1999). Do deaf people see better? Texture augmentation and visual
search compensate in adult but not in juvenile subjects. Journal of Cognitive
Neuroscience, 11(5), 560-580.
Reyna, V. F., & Kieman, B. (1994). Development of gist versus verbatim memory in
sentence recognition: Effects of lexical familiarity, semantic content, encoding
instructions, and retention interval. Developmental Psychology, 20(2), 178-191.
Ridberg, E. H., Parke, R. D., & Hetherington, E. M. (1970). Modification of impulsive
and reflective cognitive styles through observation of film-mediated models.
Journal of Experimental Psychology, 5(3), 369-377.
Salcedo, A. M. (1985). A rhetorical and structural analysis of instructional television .
Anaheim, CA: Association for Educational Communications and Technology.
(ERIC Document Reproduction Service No. 256 332).


133
Long, G. M. (1982). Interaction of arousal and task difficulty upon perceptual restriction.
Perceptual and Motor Skills, 32(1), 259-264.
Luckett, J. D. (1996). Music preparation and the good language learner. (Doctoral
Dissertation, University of Central Florida, 1996). Dissertation Abstracts
International, 57 (10), 4296A.
Marr, D. (1982). Vision. New York: Freeman Press.
Mayer, R. E., Bove, W Bryman, A., Mars, R., & Tapangco, L. (1996). When less is
more: Meaningful learning from visual and verbal summaries of science textbook
lessons. Journal of Educational Psychology, <3<3(1), 64-73.
McCollum, J. F., & Bryant, J. (1999). Pacing in children's television programming. Mass
Communication & Soc/e(y(March), 2-40.
McLuhan, M. (1964). Understanding media: The extensions of man. New York.
Merrinboer, J. J. G. (1990). Instructional strategies for teaching computer programming:
Interactions with the cognitive style reflection-impulsivity. Journal of Research
on Computing in Education, 23(1), 45-53.
Messer, S. B. (1970). Reflection-impulsivity: Stability and school failure. Journal of
Educational Psychology, 61, 467-490.
Messick, S. (1970). The criterion problem in the evaluation of instruction: Assessing
possible, not just intended outcomes. In D. E. W. M. C. Wittrock (Ed.), The
evaluation of instruction: Issues and problems New York: Holt Rinehart.
Meyrowitz, J. (1985). No sense ofplace: The impact of electronic media on social
behavior. New York: Oxford University Press.
Miller, P. H & Bjorkland, D. F. (1998). Contemplating fuzzy-trace theory: The gist of it.
Journal of Experimental Child Psychology, 71, 184-193.
Moreno, R., & Mayer, R. E. (2000). Coherence effect in multimedia learning: The case
for minimizing irrelevant sounds in the design of multimedia instructional
messages. Journal of Educational Psychology, 92 (l), 117-125.
Murch, G. M. (1973). Visual and Auditory Perception. Indianapolis, IN: Bobbs-Merrill
Company.
Murdock, B. B. (1982). Recognition memory. In C. R. Puff (Ed.), Handbook of Research
Methods in Human Memory and Cognition, (pp. 2-27). New York: Academic
Press.


67
language, musical notes, or mathematical symbols. Repleteness or density refers to the
relative richness or number of dimensions that the information conveys, like a sketch as
compared to a full-color painting, a textual passage as compared to a picture (even more
so if that picture is moving). Salomon distinguished among symbol systems with respect
to the degree that they resemble the object or concept being represented. Symbol systems
tend to vary in terms of their depictive and/or descriptive reference, from a one-to-one to
a totally abstract representation. Salomon also distinguished between psychological and
real resemblance. Depicting an object with a high degree of realism may or may not be
required for understanding, depending on an individuals pre-conceived ideas as to that
something should look like. As all cognition and learning are based on internal symbolic
representations that are central to all systems of mediated communication and thinking,
actual resemblance (i.e., fidelity) may or may not be required in individual cases.
Gardner, Howard, and Perkins (1974) agreed with this thinking and proposed that
concepts like fidelity should be viewed in its broadest context. That is, the symbolic
process should be compared to how one imagines or conceives something, not
necessarily how true to form it is (i.e., colors, lines etc). In other words, a medium might
be said to be faithful to the real world if it presents information in a way that is true to the
way one thinks (i.e., in a streaming sense).
Salomon (1979) explains that it is not the medium itself that makes the difference
in message processing. Rather, he held that film and television are generally not symbol
systems themselves, but rather are a place where multiple symbol systems are used.
Because they manifest no one system, television and film adopt the symbol system of the
content they depict. For example, television and film may use photography, print, speech,


82
score (referred to henceforth as the total score or total combined score) in order that
comparisons may be made for speed, gender, and style.
Verbatim memory was tested in two ways. First, subjects were asked in a series of
questions to pick out from groups of four the one picture that appeared in the video. They
were then asked to identity the picture in each of a series of four still images that did not
appear on the video. In order to test whether the subjects were able to comprehend (i.e.,
get the gist of) contextual meaning conveyed by the video, subjects were asked to place
pictures in chronological sequence, to identify by way of multiple choice questions some
of the major techniques used in the video to project moods and attitudes, and to answer
some questions about the circumstances (i.e., storyline) presented in the video.
Population
The subjects for this study came from a population of ninth grade students in a
high school in North Central Florida. Ninth graders were selected due to their closeness
in age to the norms developed for the instrument selected to categorize cognitive style so
that some historical comparisons may be made. That instrument provided normative data,
which was very useful in helping to categorize and compare the current group of students
as to their impulsive-reflective tendencies and current trends.
Sample
The overall sample set consisted of 204 subjects. The entire sample was used in
the analyses to determine the effect of presentation speed on combined test scores as well
as verbatim and gist scores taken separately. The entire sample was also used to
investigate differences between males and females. However, the nature of the instrument
used to categorize this sample set into one of two cognitive styles dictated the creation of


76
who referred to a concept of permitting producers to record movement in the most
complex combinations and [sic] to place points wherever they wanted (p.102). The
Russians began to produce short films that placed short cuts of still photography and
pasted them together to form a surreal photomontage in which "the meaning of a shot is
dependant upon the shots that surround it" (Stephens, p. 102). It is this view of montage
as a collection of highly related thoughts expressed visually that provided the catalyst for
its use in the current study as a way to provide gist memories in videos without the need
for textual supplements.
Montage, in combination with fast cuts, has become a whole new form of
intellectual discourse that Stephens (1996) called complex seeing (p. 178), a technique
that was not available to early television producers. Much of early television lingered
much longer over scenes, allowing the narrative to play a significant role in telling the
story. According to Stephens, new media technologies are changing the paradigm. Now
that the cuts and montage have increased the speed of the instances we see, we are no
longer dwelling on the individual instance but, rather, are being persuaded by the
onslaught and the continuity of the entire series. Not only are we affected by each
individual image but their composite total (p. 180). In essence, the ability of an
individual to become accustomed to the fast-cut montage may be a newer, higher level
form of persistence of vision -the gestalt-like concept that forms the basis of being able
to watch films. The mind closes the gaps in between pictures in a stop-action series of
movements to form one continuous motion. The current study attempts to see if montage
has the potential to follow a similar track in an educational setting.


93
the retrieval of some internal visual representations. Mayer et al. (1996) indicated that
words may not serve as proper feedback format to visual motion cues. For these reasons,
verbal questions were kept to a minimum in the current study.
In spite of the initial intent to eliminate text-based questions, it was eventually
determined that the gist portion would require some of these types. The gist questions
were intended to simulate reading comprehension-like assessments where subjects are not
asked to memorize every word they read from a book. Rather, they are asked about their
overall impressions and understandings about its content. Therefore, it was determined
that it was not important that the subjects remember every picture in the gist portion of
this the test used in this study, but, like the reading comprehension test, they would be
asked to recall contextual information from of the video. So as to further ameliorate the
potential confound imposed by the additional reading requirements imposed by the text-
based questions, all instructions and questions for the textual portion of the test were
presented in both visual and auditory form.
Research Design
Three separate analyses of variance were utilized in this study. First, it was
determined that to accurately obtain overall test results for the effects of presentation
speed and to detect possible genders differences, the entire sample set had to be analyzed.
To analyze the effect of gender and presentation speed on the overall sample, a two-way
2x3 factorial design ANOVA (2 gender categories and three presentation speeds) was
administered to the entire sample of 204 subjects, as shown in Table 1. Next, it was
determined that testing for two separate score results required a separate multivariate


32
pictorial) representations, referential, the activation of the verbal system by non-verbal
system or vice-versa, and associative, the activation of representations within the same
verbal or non-verbal system. These differences in method of processing of stimuli
contrast in many ways. While researchers like Carr (1982) believe that a common
semantic system exists that provides similar representations for words and pictures,
others have suggested that differences in processing exist or that one form of symbol
takes precedence over the others. For example, the structural organization of a perceived
sequence can play an analogous part in its perceptual segmentation (Corcoran, 1981).
Corcoran (1981) also proposed that verbal and non-verbal forms of information are
processed at different rates. Still others (Archer, 1965; Gummerman & Gray, 1972; Huba
& Velletino, 1980; Rettenbach, 1999) posited the idea that increasing the speed in
perceptual encoding might be a learned activity that varies based on age and/or sex.
Perhaps the most significant debate with regards to the role perception has on
memory has taken place between those who believe in the superiority of verbal coding
over non-verbal, or vice versa. In their studies comparing the two, Smith and Haviland
(1972) introduced two complimentary assumptions. The influence hypothesis argued that
the perceptual superiority of words is due to a subjects opportunity to use the
redundancy in words for making perceptual inferences. The unitization hypothesis
proposed that words are more perceptible because there are fewer units to process in a
word than a non-word. Other researchers concluded that visual coding takes precedence.
Pezdek (1977) performed an experiment in which the sentence the bird is perched atop of
the tree was shown repeatedly at the same time as a picture of an eagle atop of a tree.
After time, the participants responded to a follow-up questionnaire that the sentence


25
the neo-cortex where the signal can be analyzed and then sent to the amygdala and
emotional response added after cognition (Seward-Barry). This dual processing theory
provides an insight as to why visual perception takes on such a personal meaning. It also
introduces the idea that will be further developed later in this chapter that there can be
two very different emotional reactions to perception: one helpful to cognition, and one
that can be detrimental.
Visual Perception Theories
For some theorists, seeing is more than a simple biological function. According to
Seward-Barry (1997), visual perception is actually taking the visual image, combining it
with other data from the other senses, and synthesizing it with previous experiences.
Perceptual psychologist J. J. Gibson (1979) made a distinction between the visual field,
the image that appears on the retina, and a mental creation of what comprises our world,
giving further credence to the fact that to an individual, reality is very interpretive and
subjective. To constructivist theorists, visual reality is simply a map-like image, the end
product of a process that begins with light refraction in the environment and ends in the
intricate and complex dynamics of the mind (Seward-Barry, p. 15). Viewed this way,
perception is what intercepts an outside stimulus for a person to create an individual
reality. When video producers present their images in unique ways (such as utilizing
rapidly presented montage) they simulate this mapping process and create opportunities
for viewers to construct their own views and interpretations.
Gestalt theory, as outlined by Ellis (1938) and later by Wertheimer (1959), shares
some common beliefs with constructivists. However, this theory proposes a more
consistent view of perception among individuals through the concept of idea grouping


69
Evaluating the Educational Impact of New Media
The technologies which are helping video and computing converge, and with
which it is now possible to more easily create visual productions full of rapidly presented
montages, are often referred to as new media. In spite of its inference, the term new
media is not new. Researchers have been alluding to television in this way for almost
fifty years. In writing his introduction to the Seventy-Third Yearbook of the National
Society for the Study of Education, David Olson (1974) noted that new
technologies/media are not the panacea to all possible educational reform needs that
researchers were looking for:
It would be of much greater promise to discover the areas in which
media diverge and hence serve different purposes. Perhaps, the
function of the new media is not primarily that of providing more
effective means for conveying the kinds of information evolved in the
last five hundred years of a book or literate culture but, rather, it is that
of using new media as a means of exploring and representing our
experience in ways that parallel those involved in that literate culture.
(p. 8)
Olsons ideas have been backed up by other researchers. Gross (1974) suggested
that new media does not reduce the vital importance of competence in the basic modes of
intelligence and communication. Chu and Schramm (1968) found in earlier studies that
screen size, using animation, changing the aspect ratio or size of the screen made few, if
any, significant difference in learning, even though these changes all were seen to be
preferred by students in follow-up questionnaires. While the current study also takes
advantage of many of the features of new media, it is not new media by itself that can
take credit for any positive (or negative) outcomes with regards to memory. The basic
premises for how to evaluate the use of media in educational settings remain. Olson
reaffirmed the need to base research of new media using a symbol systems approach. He


74
turn, these goals influence the proportion of resources allocated to the various mental
sub-processes (such as storage and retrieval). The structural and content features of the
message elicit orienting behavior (OR) and an automatic allocation of resources. When
there are insufficient resources available to carry out all the sub-processes, some aspects
of processing are performed less well, indicating that too many inputs are overloading the
system. Fast-cuts can be this overloading stimulus, unless the content is delivered in a
structured and continuous way (i.e., it builds its story through a series of related schema).
Lang et al. (1999) further noted that it is the medium itself, by calling upon
certain processing requirements (i.e., its symbol system) that can elicit an Orienting
Response. The Limited Capacity Theory suggests that stimuli that elicit arousal result in
the automatic allocation of resources to encoding and to storage. Thus, an increase in
presentation speed by itself can increase a viewers arousal levels and result in the need
to allocate additional resources to encoding and storage. Lang and Basil (1998) and Lang
et al. (1999) and Lang et al. (2000) found that content alone can also elicit arousal.
Arousal will create the cognitive process to reach its top limit much more quickly if
content is also arousing. On the other hand, as the presentation speed increased for
messages containing only calm content, subjects were able to take on additional cognitive
tasks. Langs studies provided evidence that one should blend the types of content with
the speed of the pace it is presented. Lang et al. (2000) found that memory capacity
problems can be ameliorated if the content is structured so that the amount of new
information provided is limited, irrespective of the presentation speed of the video
presentation. Further, they found that, if the pieces of information are related to some
larger overall schema, the presentation speed also did not matter as much. These concepts


10
angle to another. The content may not change at all, only the viewers perspective. In the
current study, a cut always signifies a change in content. One picture or image was
replaced with an entirely new one.
Lang et al. (2000) indicated that under very broad interpretation, the speed of
message presentation of motion and edits/cuts can be considered a form of cueing.
Downs (1989) also referred to presentation speed in a similar way, referring to it as a
messages domain attribute (p. 3). Salomon (1979) acknowledged that media attributes
are those that are within the mediated stimulus, possibly shared to some extent with
other forms of [sic] media and make the presented information more comprehensible or
better memorized by learners of particular characteristics (pp. 5-6). Under Salomons
definition, symbols include most objects, marks, events, models, or pictures (p. 29). It
is assumed in this study that the rhythmic patterns afforded by fast-cuts are an event.
Where the current study varies from previous research is that it takes the interrogation of
symbols and attributes to another level. The current study aims to show that rapid
presentation speed (also referred to as fast-cuts) may be considered an invaluable
communicative attribute of media and is, therefore, capable of being studied separately to
discover its contribution to learning. This study aims to determine to what extent that
rapid presentation speed of video images either aids or interferes with recognition and
recall (specifically, gist and verbatim memories), taking into consideration the changes
that appear to be taking place in an ever-increasing media-centric society. Even if fast,
moderate, and slowly presented visual stimuli are all shown to be equivalent in their
ability to stimulate recognition and recall, educational research might benefit from
discovering whether visual stimuli presented at an increased delivery pace can provide


LD
1780
20 CjL
X\^
UNIVERSITY OF FLORIDA
3 1262 08557 2450


90
gist (i.e., comprehension) questions were analyzed both together and separately.
Adjustments were made to overcome any potential reliability shortcomings prior to
administering the actual test to the selected subjects. A reliability co-efficient on the
combined question set yielded an alpha level of .9028. The co-relational alpha tests to
determine reliability of the test instrument to assess verbatim and gist memory
individually resulted in co-efficient ratios of .7548 and .8662 respectively. These results
indicate that the reliability of the test instrument fell within reasonable ranges of
acceptability for its intended use in this study.
Test Administration
To test for cognitive style, a computerized version of the MFFT-20 test was
developed for use in the current study. The same pictures from the original MFFT-20
were scanned into digital form and placed on an 800 x 600 computer screen format using
Director 8.5, an interactive software program developed by Macromedia. Multiple copies
of an executable form of the program were made so that each subject was able to view his
or her own individual screen in a lab containing 25 Dell Pentium IV and 5 Apple G-4
Computers. Controls were put in place so that no subjects were able to see anothers
screens. Subjects also wore headphones in order to hear the indirections that were also
displayed on the screens. As with the pilot study, the headphones also add the
convenience of providing a more focused environment by eliminating surrounding stray
noises and disruptions. The program presented two sets of sample items and 20 sets of
actual pictures in the exact order as the paper version of the MFFT-20.
The computer program for both the MFFT-20 and the verbatim and gist questions
automatically kept track of the responses by the participants and outputted that


27
guide the senses to transform individual visual inputs into composite visual units made up
of real and imagined stimuli, forcing one to make inferences about them (Murch, 1973).
These views of perception lend credence to a belief that a standardized or
continuous view of the perceptual process can be formulated and forms the basis to the
theoretical approaches used in this study as they relate to an individuals ability to
remember and aggregate more complex meaning from video montage. Studies by gestalt
psychologists Richard Gregory and Semir Zeki into concepts, such as phi phenomenon
(processing movement as individual sequences separated by brief instances) and apparent
movement (the impression of movement from two stationary stimuli), reveal that there
might be two distinct systems of visual perception in the brain (Seward-Barry, 1997).
They also show how specialized areas of the visual cortex work together to create a
unified perception. One is responsible for perceiving movement between brief flashes and
the other takes care of perceiving movement between long flashes. These two concepts
provide further credence to the idea that the visual whole is made up of several
independent parts which, taken together, can furnish additional meaning, as afforded by
fast-cut montage found in new media technologies.
The visual process, as efficient as it appears to be, is not perfect (Seward-Barry,
1997). Much of gestalt theory is based on the law of pragnanz which stipulates that
psychological organization will always be good (i.e., simple, regular, symmetrical) as
the prevailing conditions will allow (Koffka, 1935/1963, p. 110). It is also the early
twentieth century counterpart to Aristotles concept of common sense that posited that
efficiency is achieved through simplicity, regularity, and symmetry (Seward-Barry,
1997, p. 47). These same theoretical concepts (proximity, similarity, simplicity, and,


85
reflective. Subjects found to be fast-accurate (i.e., faster and more accurate than the
calculated medians) or slow-inaccurate are placed in two other cells. As per the
instructions provided, subjects who landed in either one of these latter two categories
were excluded from the study.
According to the literature, (Berry, 1991; Green, 1985) the impulsive-reflective
test, in addition to leveling-sharpening and dependence-independence, has been one of
the most commonly used and more accurate means to test for cognitive style and to show
how individuals perceive and process visual patterns. Because the treatment in the current
study was to show accuracy of visual processing as well as processing speed, it was
determined that impulsive-reflectivity would be the most closely related of the three
cognitive style classifications to use.
Another goal of this study was to determine if there might be an interaction
between a subjects reaction to rapidly-presented content and his or her cognitive styles.
Rapid visual presentation has been found in several studies to increase cortical arousal.
Further, impulsivity has been reported by some as a stable mediator of the rate of change
in arousal states. Anderson and Revelle (1994) have demonstrated that impulsives are
susceptible to attentional lapses that are, in turn, mediated by impulsivity-related phase
differences in diurnal arousal patterns (p. 334). There is no one correct style. There
could well be a proper time and a place for being impulsive or being reflective. It was
determined that a test for the interaction between cognitive style and how well subjects
receive and process rapid visual presentations could result in some very interesting
findings that could be quite useful in helping to determine the proper use of rapid visual
presentations in the educational environment.


137
Singer, J. L. (1980). The power and limitations of television: A cognitive-affective
analysis. In P. L. Tannenbaum (Ed.), The entertainment functions of television .
Hillsdale, NJ: Lawrence Erlbaum Associates.
Smith, E. E., & Haviland, S. E. (1972). Why words are perceived more accurately than
non-words: Inference versus unitization. Journal of Experimental Psychology, 92,
59-64.
Sperling, G. (1963). A model for visual memory. Human Factors, 5, 5-31.
Stephens, M. (1996). The rise of the image the fall of the word. New York: Oxford
University Press.
Sterrit, G. M., Camp, B., & Lipman, B. S. (1996). Effects of early auditory deprivation
upon auditory and visual information processing. Perceptual and Motor Skills, 23,
123-130.
Thompson, A. D., Simonson, M. R., & Hargrave, C. P. (1996). Educational technology:
A review of the research. Washington, DC: Association for Educational
Communications and Technology.
Thorson, E., & Lang, A. (1992). The effects of television videographics and lecture
familiarity on adult cardiac orienting responses and memory. Communication
Research, 19(3), 346-349.
Tovee, M. J. (1998). The speed of thought: How fast can you see? In M. J. Tovee (Ed.),
The speed of thought: Information processing in the cerebral cortex (pp. 143-
152). New York: Springer-Verlag.
Tse, T., Vegh, S., Marchionini, S., & Shneiderman, B. (1999). An exploratory study of
video browsing user interface designs and research methodologies: Effectiveness
in information seeking tasks. Paper presented at the 62nd ASIS Annual Meeting,
Washington, DC.
Tulving, E. (1972). Episodic and semantic memory. In W. E. Tulving & Donaldson
(Ed.), Organization of Memory (pp. 381-403). New York: Academic Press.
Tyner, K. (1998). Literacy in a digital world: Teaching and learning in the age of
information. Mahwah, NJ: Lawrence Erlbaum Associates.
Verhagen, P. W. (1992). Preferred length of video segments in interactive video
programs. Proceedings of Selected Research and Development Presentations at
the Convention of the Association for Educational Communications and
Technology. (ERIC Document Reproduction Service No. ED 348 036).


51
also indications (Corcoran, 1981) that rhythmic control can be learned and translated as a
coding structure in its own right to enhance learning and can be made to act as a cueing
system. Zillman (1991) found that rapidly presented visual programs foster superior
attention, and potentially superior learning. While he admitted that much of the successes
of faster speeds may be attributable to arousal, he also quoted studies that showed that the
creation of transient alertness, even by primitive means such as the frequent instigation
of the orienting reflex, tends to facilitate information acquisition in audiences for which
high levels of attentiveness cannot be (otherwise) expected (p. 126). Apparently,
alternating the speed of an instructional message tends to increasingly arouse the cortex,
which, in turn, as many researchers have suggested, tends to make a viewer more alert
(Lang et al., 1999; Pearl, 1982; Reeves et al., 1986; Zillman, 1991). Nelson (1990)
quoted studies that seem to indicate that rhythm serves both organizational and
expressive functions that can also be shown to significantly help to organize thinking and
cognition. While investigating the effects of training using tachistoscopes Peck (1987)
found that by increasing the speed of a lesson slightly, it may be possible to increase
attention and, by doing so, also increase learning. Peck also found that the effects of
altering the speed are more pronounced for slow than for fast readers, something later
confirmed by Walters (1983). In fact, a review of the literature into the use of
tachistoscopes in remedial and rapid reading training programs (Dick, 1973; Long, 1982;
Schale, 1971; Sheingold, 1973; Woodley, 1984) appears to confirm the positive effect a
more rapid presentation speed can have on increasing attention and cognition, especially
when the content includes redundant clues and it represents conceptual information.
These rapid reading training techniques have been shown to also increase visual


118
and evaluate these types of studies than it was previously.
5. Another dimension of learning is long-term memory and forgetting. The
current study only dealt with short-term memory, an essential but
insufficient condition for learning. A study could be developed to
investigate the effects of repetition and rehearsal, and whether presentation
speed has any interaction effect. Future studies could look into using short,
rapidly presented videos as an orientation/leaming set, and expand the
subject base to make the results more generalizable to a larger population.
6. It is possible that the impulsive-reflective comparisons for gist memory
indicated a significant result was due to a relatively low number of test
questions devoted to that portion of the test (10 questions devoted to gist
versus 15 for verbatim). For the current study, a determination had to be
made to balance the need for statistical power with that for limiting the
total number of questions to a reasonable number. While the reliability
analysis performed in the pilot study on the gist questions proved
successful, it is recognized that the low number of questions may have had
a bearing on the results. A future study could be designed to study only the
effect of gist memories where more questions could be asked.
7. Joseph Hill (1981), prior to his untimely death, had begun to look into the
effects of contextualism on recognition and recall. These studies could be
expanded and investigated to determine how contextual memory correlates
with presentation speed.
8. One of the premises of the current study was that the ability of todays


I am grateful to all the many teachers and students who helped with the review of
the test instruments, pilot test, and test administration, in particular Rachael Maland,
James Keith, and Marshall Broitman. I am particularly grateful to Tom Wilson and Alan
Smolowe, who were there with me every step of the way, with a constant watchfulness,
keen insights, special interest, and concern for what I was doing.
To my family, I thank my son, Rob, and daughter, Erin, and their spouses who
cheered me on, often wondering in amazement why I would attempt to go down such a
path at this point in my life.
Lastly, I extend a very special thanks to my wife, Sandra, to whom I dedicate this
work. I am extremely grateful for her endurance, sacrifice, and constant encouragement
through this odyssey that started more than six years ago. While I often have wondered
where the time has gone, I never could imagine spending it with anyone else.


61
voices, long zooms, and long periods of inactivity were found to be less comprehensible
and less interesting.
Other studies used foreign (Salomon, 1994) or indigenous (Worth & Adair, 1997)
subjects unfamiliar with certain features and/or function common to film and television
production in the United States confirmed these findings. Salomon found that Israeli
children unfamiliar with the format or content were able to recognize and recall
information from various episodes of Sesame Street just as well as American children
who were well exposed to the shows. In teaching film-making young Navajo Indians who
had never seen movies before, Worth and Adair were able to demonstrate in qualitative
and longitudinal studies that certain film communication contains universal that do not
appear to vary, even in light of cultural biases. The Navajo films showed, as a group,
similar functional patterns in their visualizations to those of other cultures. Worth and
Adair found that how we interact and become social through our manipulations of a
variety of symbols and symbol forms (such as those associated with film) enables people
to communicate more fully and fruitfully with each other. For example, while observing
general movement patterns in a film, a viewer can also know exactly what the producer is
trying to communicate. Worth and Adair (p. 8) wrote, you can actually see what is being
done -how it moves. If you write an entire book about it, then it is still. You give it to
someone and he reads it and he does not get the same picture in his mind. What I want to
see is how something can move in front of my eyes." Worth and Adair were suggesting
that a movie is a movie, regardless of the language and culture in which the maker lives.
There are subtle differences in the use of specific techniques, content, and so forth, but
overall, most features and patterns were the same (i.e., the linear nature of episodic


86
The MFFT has been subject to several attempts to refute it as a valid
categorization test (Salkind & Wright, 1977; Watkins et al., 1978). Further, Ikegulu and
Ikegulu (1999) found that the notion of a generalized visual processing rate may be
questionable, based on the fact that there have been few repeated measurement studies to
test the generalizability of the dimension. Other research has indicated that impulsive-
reflective designation might be best depicted on a continuous (i.e., from low to high),
rather than a bi-polar scale, as reported in the impulsive-reflective array (Salkind &
Wright). On the other hand, Salkind and Wright demonstrated in other studies that
continuous scaling seems to contradict a basic definitional premise of a cognitive style
(that is style by its very nature is bi-polar). This apparent anomaly appears to some to
create a potential lack of power for the impulsive-reflective scale to be useful in
accurately classifying a cognitive style. Ault, Mitchell, and Hartmann, (1967) contributed
a loss of power due to Kagans possible over-reliance on latency rather than number of
errors to determine reflective versus impulsivity. The findings of Ault et al. seem to
contradict Kagans original hypothesis, that stated categorization is the result of both
speed and error-rate considered together (Kagan, 1965).
In spite of these and other attempts to dispute it, Kagans MFFT instrument has
been supported in several research studies that more than reinforced its validity
(Anzmendi et al., 1981; Green, May, 1985), making a strong enough case to justify its
use it in the current study. To further strengthen power of the MFFT as a categorization
technique, a 20-item variation of the original instrument was chosen for the current study
because it had been validated by its authors in five separate reliability tests, and who


100
Table 6
Source Table of Analysis of Variance for Gender
Source
SS (Type III)
df
MS
F
Sig.
Between Groups
GENDER
15.604
1
15.604
1.566
.212
Error
2013.101
202
9.966
Total
2028.706
203
Table 7
Score Means and Standard Deviation for Gender
GENDER
Mean
N
SD
Male
13.03
107
3.34
Female
12.47
97
2.94
Total
12.76
204
3.16
Table 7. The mean scores and standard deviations are shown by gender (i.e.,
males and females). The table shows a difference in scores and standard deviations
between males and females, with males obtaining a higher average score. However, as
the differences are not significant, they are considered statistically as random variations.
Hypothesis 3. There is no significant interaction between presentation speed and
gender for overall test scores.
Table 8 shows the interaction effect for between subjects of SPEED GENDER
resulted in an F ratio of .435. This F ratio is not significant at (p<.05). Therefore, this null
hypothesis was not rejected.
Table 9 shows the means and standard deviations comparing presentation speed
with gender. Males had higher memory scores than females but, these differences are
shown to be not significant. As such, they are considered random fluctuations.


46
how interesting the subject is, how relevant the information is, or simply whether the they
want to remember it at all. Furthermore, Lang proposed that the medium itself can control
some aspect of the automatic allocation of processing resources by also eliciting it own
orienting responses (ORs) from the viewer. The prime example Lang used in her research
was an increase in presentation speed. Lang found that these ORs are automatic,
reflexive, and attentional responses to changes in the environment or to stimuli, and
people have learned that they signal important information. She also found that this
automatic allocation of resources is a relatively short-tem response, occurring over
seconds. She concluded limited, elicited arousal results in the automatic allocation of
resources to both encoding and storage.
Kahneman (1973) proposed a variable capacity model that predicts that arousal
elicited by a message will increase one's overall pool of available cognitive resources.
While he failed to specify some type of upper limit, he did suggest that secondary
reaction time might slow down as presentation speed increases, especially if content is
also arousing. Conversely, as speed increases for messages with calm content, rapidity
alone might be sufficient to increase the size of the cognitive pool allocated to respond to
the message. Basil (1994) proposed a fixed capacity model that predicts that higher
resources may be committed to the message but the overall size of the cognitive pool
remains constant. Overall, arousing messages might cause slower reaction times than
calm messages if the presentation speed also increases because resources are
automatically allocated to the encoding sub-processes as the result of orienting response.
Conversely, if more resources are committed to creating a faster response, resources
available for encoding will be reduced.


83
a subset of subjects (n=129) from the original sample set of subjects (n=204) for the
second portion of the study, as demonstrated in Figure 1. In the identification of subjects
as to being either impulsive or reflective, a resulted in a certain portion of participants
Impulsive
Slow -Inaccurate
N = 65
N = 37
Fast-Accurate
Reflective
N = 38
$
it
Z
Figure 1
Subjects Included in the MFFT-20 Cognitive Style Test
did not qualify as being in either category. As per the instructions provided with the test
instrument, these outliers (i.e., the fast-accurates and slow-inaccurates) were excluded
from this portion of analysis to determine the effects of cognitive style on immediate
memory. Subjects were categorized by cognitive style as determined by a computerized
program specifically designed for this study. The program scored the subjects on each
test question according to latency to first choice and the total number of errors. The
overall sample and the subset were both randomly placed into one of three groups and
shown the video at one of the three different presentation speeds.
Instrumentation
To categorize the subjects by cognitive style, a 20-item version (MFFT-20) of the
Multiple Familiar Figures Test (MFFT) developed by Cairns and Cammock (1984) at the
University of Northern Ireland was administered. The original MFFT is an instrument
developed by Jerome Kagan (1965; 1966) that was subsequently evaluated for validity
and reliability and adapted over the years by several individuals (Arizmendi, Paulsen, &


I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy
¡O/. Yidm*
Lee J. Mlally, Chair /
Associate Professor of Teaching and
Learning
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy
Jef
Associqre Professor of Teaching and
Learning
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy
M. David Miller,
Professor of Educational Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor (/fPhilosophy
David Ostroff
Professor of Journalism and
Communications
This dissertation was submitted to the Graduate Faculty of the College of
Education and to the Graduate School and was accepted as partial fulfillment of the
requirements for the degree of Doctor of Philosophy.
May, 2002
Dean, Graduate School


84
Domino, 1981; Block et al., 1974; Watkins, Lee, and Erlich, 1978) to determine
impulsive-reflective tendencies. The original format of the MFFT-20 was developed as a
paper version in which participants made their choice of selecting the matching figure
from a set of six distracters by pointing to their choice. The investigator was responsible
for keeping track manually of the number of choices made and timing latency to first
response using a stopwatch. For the current study, the paper copies of the figures and
alternative choices that were to be matched were scanned into a computer and imported
into a program written in Macromedia 8.5 that was specifically developed for this study.
The program presented the pictures and their alternatives on one screen and allowed
subjects to click on their selected picture to indicate their response. The computer
program automatically kept track of the total number of choices made by each participant
and the amount of time to first choice for each of the item sets.
As with the paper versions of the MFFT, subjects in this study were presented
with 12 sample pictures of familiar items and are then asked to identify which one of six
alternatives is identical to the sample. If an error was made the subjects were
subsequently asked by the computer to retry until a correct response was found. Subjects
were also automatically timed as to how quickly or slowly they made their initial choices
(latency) and how many total errors they made. The dividing line between impulsive and
reflective was determined by calculating a median split score for both latency and total
number of errors. The scores were placed into quadrants made up of two intersecting
axes. Those who made very quick but inaccurate decisions ended up in a quadrant labeled
impulsive. Those who were more deliberate (i.e., showed an increased latency to first
response) and made fewer errors than the calculated median were to be determined


121
reinforce existing theories about learners and knowledge gained from viewing pictures.
The results of this study simply show that todays youths have changed their propensity
for and skills in mediated pictorial cognition. Technological improvements have both
significantly increased the occurrence of rapidly presented montage passages found in
television programs, in movies and movie trailers, and in commercials in particular. They
have also made it possible to more effectively test the effect of the rapid presentation
speeds and cognitive style that was not practicable previously.
The analyses of variance in this study revealed both some significant and non
significant results. They showed that ones cognitive style does mediate memories for
pictorial representations, but, overall, presentation speed does not. In addition, there are
no significant differences between the way males and females remember them.
The lack of significance in certain portions of this study may be significant. First,
the fact that there were no more than random differences between slow medium and fast
presentation speeds is noteworthy in itself, as these results may not be what one might
have expected in light of previous research into the interactions between presentation
speeds and object memory. Further, once the non-significant, combined test scores were
separated into their verbatim and gist portions and were subsequently analyzed, the gist
score variations were found to be significantly affected by presentation speed. Co
mingling the scores for memory together into one combined result had the effect of
masking a significant interaction that was occurring -one that might have otherwise gone
unnoticed. These results further reinforce long-standing, generally accepted instructional
principles that promote audience analysis and the matching of competency assessment
with very specific the instructional goals.


50
length of time images are displayed) to stratify presented information into manageable
chunks to aid in the recognition and recall of its viewers.
Pacing versus Presentation Speed
When referring to the speed of presentation, non-education researchers often use
the terms, rhythmicity, tempo and pacing, interchangeably. However, their definitions do
not always equate. Educational multimedia developers often refer to pacing as the rate
that a learner proceeds from one place to the next, as in self-paced instruction (Canelos,
1986; Kozma, 1986). In this connotation, pacing is more associated with interactive
video, where interaction also implies some degree of learner control of the rate and
sequence of the events that take place. On the other hand, in broadcast or instructional
video, where the events are presented to viewers at a pre-defined rate, pacing may be
better defined as the rate of information presentation (Kozma, p. 14). With this latter
definition, pacing is more associated with rhythmicity or tempo. It is within the latter
context that the term pacing is referred to in the current study. In recognition of the fact
that, in educational circles, the term pacing has long been associated with Canelos
connotation, the term presentation speed is being used exclusively in the current study,
even though pacing was the operative term used most often in the literature, especially in
studies conducted in non-educational settings.
There is one point in the research where both casual viewing and educational
researchers appear to agree: that changing the speed in, and/or adding musicality or
rhythmicity to a pictorial presentation adds to memory for its content (Flowers, 1995;
Hitchon et al., 1994; Luckett, 1996; Patel, Peretz, Tramo, & Labreque, 1998; Shaffer,
Greenspan, Tuchman, Cassily, Jacokes, & Stemmer, 2000; Wagely, 1978). There are


132
Krendl, K. A., & Watkins, B. (1983). Understanding television: An exploratory inquiry
into the reconstruction of narrative content. Educational Communication and
Technology Journal, 31, 201-212.
Ksobiech, K. J. (1976). The importance of perceived task and type of presentation in
student response to instructional television. A V Communication Review, 24, 401 -
412.
Lang, A. (Ed.). (1994). Measuring psychological responses to television. Hillsdale, N. J.:
Lawrence Erlbaum Associates.
Lang, A. (1996). Negative video as structure: Emotion, attention, capacity, and memory.
Journal of Broadcasting and Electronic Media, 40(4), 460-477.
Lang, A., & Basil, M. (1998). Attention, resource allocation, and communication
research: What do secondary task reaction times measure anyway? In M. Roloff
(Ed.), Mass Communication Yearbook (Vol. 21, pp. 443-474). Beverly Hills, CA:
Sage.
Lang, A., Bolls, P., Potter, R., & Kawahara, K. (1999, Fall). The effects of production
pacing and arousing content on the information processing of television messages.
Journal of Broadcasting & Electronic Media, 43(4), 451-468.
Lang, A., Zhou, S., Schwartz, N. Bolis, P., & Potter, R. (2000). The effects of edits on
arousal, attention, and memory for television messages: When and edit is an edit,
can an edit be too much. Journal of Broadcasting & Electronic Media, 44(1), 94-
109.
Langer, E. J., Blank, A., & Chanowitz, B. (1978). The mindlessness of ostensibly
thoughtful action: The role of "placebic" information in interpersonal interaction.
Journal of Personality and Social Psychology, 36, 635-642.
Langer, E. J., & Imber, L. G. (1979). When practice makes imperfect: Debilitating effects
of over-learning. Journal of Personality and Social Psychology, 37, 2014-2024.
LeDoux, J. (1991). Emotion and the limbic system concept. Concepts in Neuroscience,
2(2), 169-199.
Leino, J. (1981). Psychometric test theory and cognitive processes: A theoretical scrutiny
and empirical research. Research Bulletin No. 57, Helsinki University, Institute of
Education, Finland. (ERIC Document Reproduction Service. No. ED 223 672).
Loftus, G. R., & Kallman, H. J. (1979). Encoding and use of detail information in picture
recognition. Journal of Experimental Psychology: Human Learning and Memory,
5(3), 197-211.


49
In conveying any mediated information, it has long been deemed important to
separate the flow into manageable chunks and to isolate and focus on some events at the
expense of attention to occurrences in between (Kozina, 1986). According to Zillman and
Brosius (2000), a mediated narrative (even an instructional message) broken down into
exemplifying chunks may jump from event to event, irrespective of the events proximity
in location or time, so long as it shares enough similarity that makes it sufficient to
provide reliable information about other events in the group (or the group itself). A
picture or series of pictures, then, may become a form of an exemplar to be used to aid
recall and recognition as long as they formulate a schema for their viewers. While it is
has been shown that learners remember more when they attach personal meaning to an
idea or concept (Brown & Kulick, 1977; Tulving, 1972), research into casual television
viewing has also demonstrated that viewer recognition and recall for content of news
shows (Zillman & Brosius), music videos (Hitchon et al., 1994), and/or advertisements
(Hill & Lang, 1993) can be modified, either through a form of exemplification of the
information by means of lashing loosely related content or structure in associated
pictorial representations. All three of these studies, taken together, show that by using a
combination of stratifying content into meaningful and related chunks and employing
certain cueing strategies can overcome some of the memory-related limitations imposed
by presenting unfamiliar content. Mayer et al. (1996) suggest that attention and memory
for pictures (especially moving pictures) can be controlled by the use of formal
exemplifying features such as zooms, presentation speed, sound effects, and music.
Likewise, the format of the instrument used in the current study employs many of these
visual exemplification techniques (such as similar coloration, zooming, and differing the


2
delivery approach. According to Seward-Barry (1977), sleeping has become the only
activity that occupies childrens time more than watching television or playing video
games.
For these reasons, a change may be warranted in the way in which educators view
the role video media play in initiating learning opportunities and conveying more
complex or abstract thought. This thinking appears to be incompatible with earlier
communications theorists like Edgar Dale (1969), who felt that television finished
somewhere in the middle of a twelve-point influence scale in its ability to convey
contextual ideas. New digital media that employ visual imagery supplemented by quick
motion, sound, and computer editing (Stephens, 1996, p. 69) may be ready to take on the
same predominance as a communication medium in only fifty years that it took
Guttenbergs great invention five hundred years to accomplish. Corcoran (1981) defined
intelligence as a skill in a particular medium and suggested that the symbolic codes used
in that medium that serve communication purposes and are internalized by a receiver
become an authentic tool of thought. Research studies in which technology (i.e., media)
was successfully used as cognitive mind tools appear to back up this thinking (Hokanson,
2000; Jonassen, 1996).
Statement of the Problem
Researchers and educators looking into the intrinsic instructional value of video
media have presented conflicting views on the role visual perception plays in attention,
motivation, and recall. There have been several studies that have looked into the potential
affect mediated coding systems have on cognition (Davis, Scott, Pair, Hodges, &
Oliverio, 1999; Nugent, 1982; Seidman, 1981; Walma van der Molen & Van der Voort,


35
work found that the signal is sufficient to cause an emotional response, it wasnt strong
enough to reach conscious thought, and that the emotional responses are autonomous and
primal, similar to those that humans, have needed for survival throughout their
evolutionary process (Dixon, 1981). This inherent automatic preparation of their response
to stimuli possibly explains why humans might likely be unconsciously biased towards
accepting or rejecting ideas, and occasionally acting on them. It is also where researchers
believe the most significant contributions to research into subliminal message processing
might lie (Seward-Barry, 1997).
Relationship of Memory and Learning
Studies involving learning and memory date back more than a century (Crowder,
1976). Beginning with the Ebbinhaus preliminary studies into brain associations during
the late nineteenth century, the psychology of learning principles evolved by the mid
twentieth century into Kohlers framework of analyses that addressed three processes
associated with learning and memory: stage analysis, coding analysis, and task analysis.
While learning may actually encompass many more of these analyses processes (such as
habituation, acquisition strategies, operant training, etc.) all of which are significant
when taken in their entirety they are beyond the scope of the current study. This study
deals with simple recognition and comprehension assessed in stage analysis, an essential
pre-condition to learning, and certain aspects of coding analysis, a subset of the memory
process that has been found to be an adequate premise on which to base assumptions as to
whether learning may be taking place (Estes, 1994). Estes related the ability of
individuals to complete the process of recalling information from memory to their ability
to classify or categorize that information in a meaningful way. Under this definition, the


38
limitation of assessing learning based on recognition alone includes the fact that it is not a
useful measurement when the responder doesn't have the object in front of him or her for
identification. Perhaps the most significant contribution to the learning process that
recognition plays relates to what Singer (1980) called the differential affect. In short,
differential affect is an original startle that occurs when new and novel information is
first presented. The brain processes this information on the right side of the brain in the
same way it processes information to be later used in recognition activities. Singer noted
that the startle effect takes some getting used to, but once the initial startle is over, the
brain is better able to process subsequent information it receives. Theories surrounding
the startle effect suggest that increasing recognition from stimuli used in the quick-cut
imagery in todays new media is both a trainable act and is sufficient to gain viewers
attention so that they may recognize it again during future stimuli engagements. As such,
it helps to formulate a theoretical basis for certain hypotheses in the current study as well
as suggestions for potential future research found in Chapter 5.
Another assertion about the differences between recognition and recall is that they
are basically the same with the exception that the former may be a somewhat easier
process because it involves a lower capacity threshold requirement (Crowder, 1976).
Crowder found that this idea about recognition and recall has been discredited by some
due to its tie to a concept that recognition and recall are basically the same processes.
Crowder also reports that researchers have more recently looked upon the two as linked
but separate operations. Recall involves the generation of an additional process that can
be tested separately. For example, a variable might improve recall while, at the same
time, damage recognition. Additionally, there is evidence in experiments involving


THE EFFECTS OF COGNITIVE STYLE AND GENDER ON
VERBATIM AND GIST MEMORY FOR
RAPIDLY-PRESENTED MONTAGE VIDEO
i *
l i. V i *
By fj.
ROBERT F. KENNY
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSPOPHY
UNIVERSITY OF FLORIDA
2002


LIST OF TABLES
Table Eage
I.2x3 Two-Way ANOVA (204) Subjects 94
2. Multivariate analysis of variance (204) Subjects 94
3. 2 x 3 two-way ANOVA (129) Subjects 94
4. Source table of analysis of score variance by presentation speed 99
5. Score means and standard deviation for presentation speed 99
6. Source table of analysis of variance by presentation speed and gender 100
7. Score means and standard deviation of scores for gender 100
8. Source table of analysis of variance for speed versus gender 101
9. Score means and standard deviation for speed versus gender 101
10. Source table of analysis of variance for verbatim score for speed 101
II. Source table of analysis of variance for gist score for speed 101
12. Multiple comparisons between speed and gist scores 103
13. Score means and standard deviation for gist versus verbatim test items 103
14. Source table of analysis of variance by presentation speed and cognitive style 103
15. Means and standard deviation for overall scores for style 104
16. Source table of analysis of variance for verbatim and gist and cognitive style 105
17. Means and standard deviation for gist and verbatim scores for style 106
vii


89
themes presented in the video. Some of these were pictorial questions and some were
text-based. These three question banks (consisting of a total of 45 questions) were then
presented to three social studies teachers and one English teacher who evaluated them for
appropriate content, duplication of pictures, and to check for potential fidelity problems
that might distinguish correct responses from distracters. The overall set of 45 questions
was pared down to 25, based on the feedback from the teachers. The questions were then
presented to a several student test viewers who provided additional feedback. As a result
of this phase, the program was changed to randomly place the responses on the screen so
as to randomize the placement of correct answers on the screen and to eliminate the
opportunity for participants sitting next to another to share answers. In addition, the color
schemes for the screen background and text were changed to reflect those combinations
deemed more pleasing to the younger age group of participants for whom this test was
designed.
Pilot Test
A pilot test was conducted with a group of students from a different high school
in the area consisting of subjects of identical age and similar demographic make-up as the
final test group. The pilot group consisted of 97 subjects. They were given the test in a
group setting that was arranged in such a way that no one could see any on elses screens.
So as to supplement the viewers understanding of the instructions, an audio instructional
track of a female reader was added to the computerized program so that the participants
could both see and hear the directions on how to take the test. Headphones were used to
reduce background noises and distractions. To determine reliability, two alpha reliability
scale tests were run. The 15 items dealing with verbatim recall and the 10 items asking


7
for yielding any new intellectual construct of its own. Previous studies into the value of
using television as a medium for intellectual change have demonstrated mostly negative
progress in that the results showed television did not actually interfere with learning, or
that it was not less effective than other forms of media (Thompson, Simonson, &
Hargrave, 1996; Wetzel et al. 1994). The current study attempts to clarify where rapidly
presented montage might fit into the overall instructional scheme.
Previous attempts at investigating rapid video and montage presentation were
limited by some incorrect assumptions as well as limitations on technology. For example,
Intraubs (1999) experiments that demonstrated weaknesses in an individuals ability to
understand and remember briefly-glimpsed images dealt with pictures that were not or
only very loosely related. The proper use of montage implies that the pictures included
are at least conceptually related. Intraub indicated that subjects might be able to hold
more than one picture at a time in a conceptual buffer, so long as the series was not too
long (p. 57). Keller (1976) tried to fix this shortcoming in his dissertation but ran into
some serious technical flaws in his overall design. Intraub even admitted that his
experiments might have been more successful if they included pictures that were related
to one another. His intuition may have been correct. It appears that humans may have the
ability to construct meaning from these types of presentations through the use of
interpretive coding (i.e., the process by which meanings are put together from specific
parts of visual communications). In describing their research in teaching Native
Americans how to use film to communicate meaning, Worth and Adair (1997) noted that
the process of coding has been neglected in the study of most of the fine arts, including
film. Their comments alluded to the fact that the form of a medium might be what carries


79
are presented medium or fast. Producers should not create messages that are, at the same
time, calm and presented slowly or arousing and presented too fast. Carrying this thought
into an educational setting, Verhagen (1992) found that increasing the presentation speed
of video images can be fully interpreted and remembered if learners start their viewing
task with a realistic expectancy of its demand characteristics and are motivated to tune
their mental effort accordingly. Further, research in tachistoscope training (Schale, 1971)
showed that rapid reading training increases visual awareness of larger spans of print, and
that the processing rates can be accelerated without significant loss of comprehension. As
a minimum, even if comprehension is the same, or even if it suffers only slightly, fast cut
montage may not be a negative influence.
Media choices may be as much about cost and efficiency as about cognition and
learning (Cobb, 1997). If it is accepted that one type of efficiency is cognitive efficiency,
it follows that media choices can profit from an understanding of cognitive processes
much in the same way as the latter can be positively leveraged by the type of media that
is ultimately selected.
In this chapter, a theoretical basis was laid out for the methodologies, procedures
and analyses used in the current study. An analogy may be made to describe the way that
video montage derives its ability to convey meaning through the use of closely-related
pictures and rapid presentation. One should view montage similarly to the concept of
chunking in speed reading of text-based passages. In montage, the individual pictures
take the place of words and sentences. Several related pictures become the paragraphs.
Presentation speeds may vary, just as one might vary reading rates, causing variations in
the ability of individuals to remember specifics and derive contextual meaning from the


70
resolved that previous research showed the lack of a strong theoretical base, and that new
media would not reach their potential until some scheme is shown through empirical
evidence that they can re-specify information, how that this re-structuring would be
influenced by the media that presents it, and what the psychological consequences of
relying on that re-structuring might be. Media are similar in the knowledge they
communicate, but they differ as to the skills needed and developed by using them.
Evidence appears to show that educational media cannot be chosen simply because of
their ability to convey certain kinds of content. Rather, they must be chosen based on
their ability to develop the intellectual processing skills that they help to develop.
Olson (1974) outlined several criteria for analyzing the theoretical potential of
various media for use in educational settings which became a basis some of the goals of
the study and several of the recommendations in Chapter 5. To evaluate the instructional
impact of media one must:
Identify the specific symbol system it uses.
Identify the specific basic skills that are required and how much of literacy
is required for use of this medium.
Identify the intellectual consequences of exposure to that medium (i.e., is
the ability to imagine an action in slow motion dependent upon having
seen slow-motion film?).
Identify the scholastic goals for which the medium is most appropriate.
Determine how knowledge and skill (intelligence) in the medium can be
evaluated.
New Media and Memory
New media make it very much easier to create new an interesting pictorial
storylines using rapidly presented montages, as evidenced by the exponential use in


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
THE EFFECTS OF COGNITIVE STYLE AND GENDER
ON VERBATIM AND GIST MEMORY FOR
RAPIDLY-PRESENTED MONTAGE VIDEO
By
Robert F. Kenny
May, 2002
Chair: Lee J. Mullally, Ph.D.
Major Department: School of Teaching and Learning
Younger students live in a media-centric world. Researchers have shown that
todays youth spend more time watching television and movies than most any other
leisure-time activity. Moreover, the presentation speed of passages on commercial
television has increased significantly in the past fifty years. Researchers have shown that
viewers automatically leam to cope with symbolic pictorial presentation methods through
repeated exposure to television and its visual patterns. Increased exposure to rapid
sequence and presentation speed brought on by rapidly-presented montage found in
television programs aimed at youth raises the question as to whether these individuals
remember more from this type of presentation techniques than can their adult
counterparts. Recent advancements in technology have changed the way viewers look at
and interpret video media. Rapidly presented video media place additional attentional


12
Purpose of the Study
This study was designed to determine if there are differences in the way visual
montage messages that vary only in their presentation speed (i.e., fast, medium, or slow)
are perceived and immediately remembered in an educational setting. Recent reports
found in the literature, (Brainerd & Reyna, 1990; Brainerd & Gordon, 1994; Reyna &
Kieman, 1994) have suggested that researchers have been able to successfully parse
memory into verbatim (i.e., precise/literal) memory for specific details and gist memory
(i.e., contextual or contextual remembrances similar to that which is assessed in reading
comprehension tests). This study looks at immediate memory as a whole and separately,
using these same categorizations. The educational setting was determined to be several
ninth grade classrooms that were categorized by gender and also further delineated by the
subjects individual cognitive/leaming style. The overall sample set for this study came
from that population group. For purposes of this study the identification of cognitive style
was limited to the reflective-impulsive scale, as originally developed by Jerome Kagan
(1965; 1966), and later refined and re-catalogued by Cairns and Cammock (1984).
Identical videos that varied only in their speed of message delivery were presented to
three randomly assigned groups of students. The procedural requirements for
implementing the cognitive style test instrument indicated that the analysis was to be
performed on a smaller subset of the sample base to see if the variance in delivery speed
also affects gist and verbatim memory, and if there might be any interaction with their
prospective cognitive styles.
Previously, presentation rate had been found to affect immediate memory in
examinations of viewers in a commercial viewing environment that compared several


18
scenes. The objective of assessing gist memory in this study was to determine whether
the subjects comprehend the basic contextual themes presented on the video and to
compare it to subjects ability to immediately remember specific (i.e., verbatim) details
from the video presentation.
Kinestasis is a term coined by Charles Braverman (1969) that is derived from
combining the Greek words for moving and static. The term refers to a film or video
process that presents still and/or moving images at speeds that create the illusion that they
are moving in a continuous flow. The treatment video used in this study is a video copy
of a kinestasis film developed by Braverman, and whose copyrights are owned by
Pyramid Media.
The term media includes films, television, and compressed motion pictures found
in digital multimedia. Previous researchers into presentation speed (Wetzel et al., 1994)
and compressed motion pictures (Arnold, 1996) delved into one or more of these types of
media and have referred to them as virtually interchangeable in meaning. This study also
refers to the term new media. This term connotes an inference about the convergence (the
merging of television, computers, and the telephone) that is taking place. The current
study uses screen media and video media interchangeably when referring to its
capabilities to present a pictorial story. It uses new media to mean traditional media that
are being technologically advanced due to convergence.
Minimal unit of analysis is that product of semiotic analysis which is the smallest
interchangeable element that still retains the basic properties of the whole symbol system
but cannot be further subdivided without losing those properties (Corcoran, 1981). Hill
(1981) referred to this concept as a symbol, or the basic unitary element of intellectual


75
were supported by earlier research (Amheim, 1974; dYdewalle & Vanderbeeken, 1990)
into recalling visual sequences where individual frames of information were best
remembered when they related to an overall schema of events. An important factor in
memory enhancement in all these experiments was the cohesiveness of the contextual
content of the messages being presented.
Montage
Mitchell Stephens (1996) elaborated on the potential for an increase in memory
capacity for related schemes of events in his chronicles of the development of montage in
cinema and television. He delineated the beginnings of the use of montage from its use in
the early films of Sergei Einstein (1949), to Chuck Bravermans (1969) use of kinestasis
editing, and its later extensive use in MTV music videos. Stephens demonstrated how
rapid presentation rates have not only enhanced montage videos. He also provided
several examples how producers have been able to replace textual messages as a primary
form of communications. His work appears to support the same type of technological
determinism (p. 21) found in McLuhans (1964) the medium is the message principle:
that the means we use to express our thoughts also change our thoughts. Stephens
suggests that, conceptually, most of thinking with regards to mankinds current system of
logic is based on the literacy of the written word. He even predicts that new media, with
its ability to empower individuals to regularly use vast communicative strategies like
montage, will create a new, unique form of abstract logical thought, similar to what Alan
Kay (1999) had in mind.
In order to provide full understanding of the concept of montage, Stephens (1996)
traces the etymology of the term. Montage was first introduced by Russian filmmakers,


130
Hayes-Roth, B., & Hayes-Roth, F. (1977). Concept learning and the recognition and
classification of exemplars. Journal of verbal learning and verbal behavior, 18
(5), 321-338.
Hedberg, J. G., & McNamara, S. E. (1985). Matching feedback and cognitive style in
visual CA1 tasks. Paper presented at the Annual Meeting of the American
Educational Research Association, Chicago, IL, March 31 April 4, 1985. (ERIC
Document Reproduction Service No. ED 260 105).
Hill, J. E. (1981). The educational sciences: A conceptual framework. West Bloomfield,
MI: Hill Educational Sciences Research Foundation.
Hill, S., & Lang, A. (1993). The effects of redundancy, pacing, and visual complexity on
memory and recognition of in-stadium advertisements. Paper presented at the
International Communication Association, Miami, FL.
Hill, W. F. (1985). Learning: A survey of psychological interpretations (4th ed.). New
York: Harper & Row.
Hitchon, J., Druckler, P., & Thorson, E. (1994). Effects of ambiguity and complexity on
consumer response to music video commercials. Journal of Broadcasting and
Electronic Media, 38, 294-306.
Hokanson, B. (2000). Accelerated thought: Electronic cognition, digital image creation
and analysis as a means to examine learning and cognition. Dissertation
Abstracts International, 61 (03), 953A. (UMI No. 9963005).
Huba, M. E., & Vellutino, F. R. (1980, August). The development of visual encoding and
retention skills. Journal of Experimental Child Psychology, 30, 88-97.
Ide, T. R. (1974). The potentials and limitations of television as an educational medium.
In D. R. Olson (Ed.), Media and Symbols: The forms of expression,
communication, and education (pp. 330-356). Chicago: University of Chicago
Press. The Seventy-Third Yearbook of the National Society for the Study of
Education.
Ikegulu, P. R., & Ikegulu, T. N. (1999). The effectiveness of window presentation
strategy and cognitive style of field dependence status on learning from mediated
instructions. (ERIC Document Reproduction Service ED 428 758).
Intraub, H. (1999). Understanding and remembering briefly glimpsed pictures:
Implications for visual scanning and memory. In V. Coltheart (Ed.), Fleeting
memories: Cognition of brief visual stimuli (pp. 47-94). Cambridge, MA: The
MIT Press.


57
were unable to subsequently correctly perceive a very clear and undistorted image.
Although adults tend to be more reflective in response to stimuli, these same reflective
tendencies could work against them when comparing the correctness of their responses to
younger individuals, especially if the younger person participates more often in visually
oriented past times. There appears to be a crossover of the effectiveness of ones
increased developmental reflective abilities and ones ability to effectively develop a
corresponding facility to realize perceptual closure from visual data that is either less than
highest in fidelity and/or is displayed at a very rapid rate.
Anderson and Revelle (1994) looked at the effect changes in daily arousal rhythm
patterns play in causing similar alterations in impulsivity-reflectivity. In this context,
arousal is meant to portray the processes that mediate non-specific alertness, or liveliness.
Impulsives tend to demonstrate high alertness and sense of arousal, a key element in
learning preparedness. Highly impulsive children have also been linked to certain
attention deficit syndromes, as portrayed by their lacking in the ability to sustain longer
periods of attention (Anderson & Revelle). Anderson and Revelles research indicates
that the impulsivity rates vary (i.e., are more prominent) by the time of day, with
impulsivity in those individuals with impulsive tendencies being more pronounced in the
morning hours.
Using impulsivity-reflection to categorize individual differences has not been
without a certain amount of controversy. While the impulsive-reflective scale can be very
beneficial in predicting certain academic outcomes and has significance in timed tests,
there are shortcomings in making a completely accurate independent identification with
regards to whether a student fits this classification in all contexts and under which


16
Delimitations of the Study
This study was conducted with certain limitations. In previous normative data
validation and reliability studies for the cognitive style testing instrument being used in
this study, subjects were limited to ninth grade students from a specific geographical
area. Only those students who agreed to participate and whose parents signed a consent
form are included in the study. The study only looked into the amount of information that
could be immediately remembered from the video passages. While immediate memory
has been deemed an important prerequisite to learning, this study makes no
representation with regards to whether the material is subsequently learned (i.e., retained
for longer periods of time, or able to be utilized in subsequent tasks) by the subjects. The
fact that learning may or not actually take place implies that additional pedagogical and
cognitive techniques are subsequently employed, and is beyond the scope of this study.
Limitations of the Study
This study contains limitations and assumptions that may affect the
generalizability of the results. The study is only generalizable to the population from
which the sample is taken. In this case, subjects came from one high school of a mixed
ethnographic and demographic make-up in North Central Florida. While it is understood
that students who have previous knowledge of the topic (i.e., American history) might be
more likely to recognize and recall the pictorial representations in the videos, it is also
assumed that this knowledge base was most likely equally distributed throughout the
sample set because subjects were selected at random. Also, as the students were
randomly assigned to the three presentation groups, it was determined that any previous
history knowledge would not materially affect the results.


91
information to text files. These files were then imported into Microsoft Excel and
concatenated to consolidate the results for each test instrument. These results were then
exported into an SPSS version 10.0 data file for analysis.
Stimuli
Subjects were shown a video entitled American Time Capsule, a kinsestatic visual
portrayal of a chronological history of the United States which is presented very rapidly
through a montage of approximately 1,300 still photographs, portraits, and paintings. The
original video was imported into a Macromedia Director 8.5 program written specifically
for this study in which it could be played back at one of three different speeds (fast,
medium, and slow), based on a specific password that was randomly provided to each
subject. A spreadsheet was developed to track which password was assigned to each
subject and to ensure an equal distribution of assignments over the sample base, for each
gender, and for each cognitive group. At the fast rate, subjects saw the video at its
original speed (averaging 300 milliseconds per picture) that lasted approximately four
minutes. The medium speed presented the video in about six minutes. The slowest
speed presented the 1,300 pictures in about eight minutes, or one every 500-600
milliseconds. The subjects were first shown a sample portion of the video presented at the
same rate of speed for the purposes of gaining their attention, accustoming them to the
presentation style and speed, and providing an overview of the subject matter. The
subjects were able to begin viewing the entire video when they felt they were ready by
clicking a button on the screen. In order to facilitate correct interpretation of the
directions supplied, the subjects were provided the same set of instructions in a small


112
slower groups did not score better might have been because they became disinterested
(the videos lasted six and eight minutes versus four for the fastest group) as the duration
of the video grew longer -a phenomenon observed by the principle investigator while
conducting this study. The poorer scores may have been due to a lack of attention that
could have counter-balanced any positive gain in memory that the subjects might have
enjoyed by being able to look at the pictures for longer periods of time.
Gender
The total mean score for males (13.03) was larger than that for females (12.47).
However, the analysis of variance did not indicate that this difference was significant at
the .05 level, indicating that any differences in scores that did exist between males and
females were random variations. Therefore, it was determined that gender, as a variable,
was not a factor and need not be considered separately in the remainder of this study.
Verbatim versus Gist Memory
The mean scores for both verbatim (7.11 out of 15 possible correct responses) and
gist (5.65 out of 10) both represent an accuracy level of about 50% of the total number of
questions for each type. For verbatim memory questions, subjects who watched the video
at the slowest rate performed better than did those who watched the video at the medium
or fastest rates. The scores appeared as if they would improve as the presentation rate
slowed down, but the differences between the fast and medium speeds were virtually
identical (6.93 and 6.97 respectively). Further, these differences were not significant at
the .05 level at any speed, therefore were considered to be random fluctuations. In
summary, verbatim memory was not affected by presentation speed.


62
sequences). While the study with the Navajo tribe dealt with making films, Worth and
Adair found that much of their findings would carry over to video. In later experiments
several years later, their results with video followed a parallel course. For example, they
found an aversion on the part of their subjects to the use of zooming and close-ups while
teaching them to use the medium. These same proclivities carried over, even with
advanced technologies and easier access found on the video cameras.
Salomon (1979; 1994) suggested that a proper review of media effects of
television involves the examination of the subtle interactions among components of the
medium, individual characteristics of its viewers, and the proposed outcomes for which
the medium is being used. Zillman and Brosius (2000) make a further differentiation. To
them, many of the effects attained by mass media are due to the redundant, stereotypical,
and cyclical attributes by which viewers learn by extracting personally relevant material
(i.e., episodic memory per Tulving (1972)). According to Zillman and Brosius, rather
than creating a sense of mindlessness, this redundancy is actually what creates the power
of the medium. They pointed out that all media (especially mass media) create learning
through the use of specialized media effects (i.e., exemplars) that may be best realized
when used over and over again. For example, most successful mass media producers
understand that, in conveying information, it is important to cut the flow into manageable
chunks and to isolate and focus on some events at the expense of attention to occurrences
in between. In other words, a story line may jump from event to event irrespective of the
events' proximity in location or time. It is an effective strategy so long as the succeeding
events share enough similarity and are sufficient to imply that every individual event is
capable of representing the group of events at large (Zillman & Brosius). In other words,


15
which permission has been granted by the copyright holder to modify for use in this study
(R. Wright, personal communication, March 21, 2001).
Finally, Kellers (1976) subjects were college level communications students who
were segregated by grade point average. He identified the need to find a more rigorous
randomization variable than GPA, noting that there was a significant interaction effect
among the subjects, who accounted for almost 24% of variance in picture memory
recognition, while the treatment effect only accounted for 6%. The current study
attempted to look at an additional classification method (i.e., cognitive style) that had
been identified and validated in research subsequent to Kellers study (Green, 1985;
Kagan, 1965; Okun, Callistus, & Knoblock, 1979; Ridberg, Parke, & Hetherington, 1970;
Salkind & Wright, 1977; Witkin & Goodenough, 1981).
Research Questions
Following are research questions raised in this study:
1. Is there a difference between the amount remembered in terms of verbatim
(specific details) and gist (contextual content) memory when changes in
presentation speed occur?
2. Are there gender differences in ones ability to remember information presented
in rapid video presentations?
3. Are verbatim and gist memory affected by a learners cognitive learning style?
These specific questions are intended to lead to an over-riding question as to whether if
the amount of information retained through fast-cuts/montage is found to be equivalent to
more moderately paced instruction, is a faster delivery more educationally efficient, as
determined by the fact that the same amount of material can be successfully presented in
a shorter time interval?


127
Campbell, D. S., & Davis, R. B. (1982). On the validity of reflection-impulsivity as a
construct in classroom research. Paper presented at the Annual Meeting of the
American Educational Research Association, New York, March, 1982, (ERIC
Document Reproduction Service No. ED 222 502).
Canelos, J. (1986, January). External pacing as an instructional strategy for the design of
micro-computer based instructional program to improve performance on higher
level instructional objectives. Paper presented at the Annual convention of the
Association for Educational Communications and Technology, Las Vegas, N.V.
Carr, T. H. (1982). Words, pictures, and priming: On semantic activation, conscious
identification, and the automaticity of information processing. Journal of
Experimental Psychology Human Perception and Performance, 8, 757-777.
Chu, G. C. & Schramm, W. (1968). Learning from television: What the research says.
Washington, D. C.: National Association of Educational Broadcasters, Stanford,
CA. Institute for Communication Research.
Clark, R. E. (1983). Reconsidering research on learning from media. Review of
Educational Research, 53, 445-459.
Cobb, T. (1997). Cognitive Efficiency: Toward a revised theory of media. Educational
Technology Research and Development, 45(4), 21-35.
Coltheart, V. (1999). Introduction. In V. Coltheart (Ed.), Fleeting Memories: Cognition
of brief visual stimuli (pp. 1-12). Cambridge, MA: The MIT Press.
Comstock, G. A., Chafee, S., Katzman, N., McCombs, & M., & Roberts, D. (1978).
Television and human behavior. New York: Columbia University Press.
Cooper, L. A. (1982). Strategies for visual comparison and representation: Individual
differences. In R. J. Sternberg (Ed.), Advances in the Psychology of Human
Performance Hillsdale, NJ: Lawrence Erlbaum Associates.
Cooper, R. (2000). Visual dominance and the control of action, [Online]. Available:
http://www.psyc.bbk.ac.uk/staff/rc/publications/cogsci98/visual_dominance/
[2001, March 11],
Corcoran, F. (1981). Processing information from screen media: A psycholinguistic
approach. Educational Communications and Technology Journal, 29, 117-128.
Cowan, N. (1998). Children's memories according to fuzzy-trace theory: An endorsement
of the theories purpose and some suggestions to improve its application. Journal
of Experimental Child Psychology, 71(2), 144-54.


103
as a percentage, those subjects who watched the video at the two faster speeds also
tended to get more gist questions correct than those watching at the slow speed. For the
verbatim questions, there were no significant differences based on changes in speed.
Table 12
Multiple Comparisons Between Speed and Gist Scores
SPEED
SPEED
Mean
Difference
Std.
Error
Sig.
95%
Confidence
Interval
Lower
Bound
Upper Bound
Fast
Medium
.25
.287
1.000
-.45
.94
Slow
.96*
.298
.004
.24
1.68
Medium
Fast
-.25
.287
1.000
-.94
.45
Slow
.72
.303
.057
-.01
1.45
Slow
Fast
-.96*
.298
.004
-1.68
-.24
Medium
-.72
.303
.057
-1.45
.01
* Significant at (p<.05)
Table 13
Score Means and Standard Deviation for Gist versus Verbatim Test Items
SPEED
Mean
Std. Deviation
N
VERBATIM
Fast
6.93
2.47
75
(15 questions)
Medium
6.97
2.26
69
Slow
7.48
2.33
60
Total
7.11
2.36
204
GIST
Fast
6.01
1.79
75
(10 questions)
Medium
5.77
1.64
69
Slow
5.05
1.72
60
Total
5.65
1.76
204
Hypothesis 6. There are no significant differences in overall test scores for those
subjects from the overall sample set who are determined to possess impulsive or
reflective tendencies.
As seen in Tables 14 and 15, the sample size is smaller (n=129 versus n=204) due
to the procedures involved in determining impulsive and reflective tendencies. A portion


CHAPTER 1
INTRODUCTION
Media educators and theorists for years have been analyzing Marshall McLuhans
famous quip, the medium is the message (Meyrowitz, 1985). In some regard, McLuhans
statement may be considered a pre-emptive retort to later critics of educational media like
Richard Clark (1983) who claimed that media are mere vehicles that deliver instruction
but do not influence achievement any more that the truck that delivers groceries causes
changes in nutrition (p. 445). Communications theorists like Walter Ong (1982) not only
agreed with McLuhan, but also extended the meaning of his message to imply that the
types of media people use define the way they think. Ongs notions bring to mind
possible questions as to whether todays media-centric youth perceive differently than
previous generations, with implications as to the kinds of mediated instructional
strategies that might be successful in motivating them to learn as well as providing
perceptual stimuli for recognition and recall.
The predominant types of production techniques used in todays electronic media
are changing. Encouraged by the successes of early pioneers of rapidly presented music
video montages on networks like MTV, VH1, and Nickelodeon, and helped by rapid
advances in technology, todays television producers regularly communicate very
complex messages using fast-cuts and video montage (Stephens, 1996). Significant
increases in viewership of these programs may be providing fertile opportunities for
todays youth to practice their ability to receive and assimilate this fast-paced message


CHAPTER 2
REVIEW OF THE RELATED LITERATURE
Perception and Knowledge
When humans learn from the world around them, they receive stimuli
(information) via one or more of their five senses: audition (hearing), gustation (taste),
olfaction (smell), kinesis (touch), and vision (sight) (Murch, 1973). Processing received
stimuli can be automatic or be the result of some literal attempt to react to it. For
example, when one smells food, it may make him or her feel hungry. This may happen
with or without that person actually realizing what is happening.
Besides being the catalyst for automatic processing of bodily functions,
perception has also been considered by educational researchers as a fundamental building
block for knowledge and learning. Murch (1973) posited that humans must be able to
perceive something before they are able to learn about it. This information may be
firsthand, or via an artificial artifact (also known as a medium). Gibson (1969) defined
perception as a process of receiving stimuli and consciously reacting to them by
assigning some meaning. Accordingly, humans tend to react to perceptions individually,
based on previous personal experiences, making this experience very subjective. Winn
(1982) described the process of learning as a taxonomy of three procedural steps. After a
person first receives information to be stored, and later when he or she attempts to
retrieve it from memory, a long process of transformation, abstractions, and elaborations
takes place. The first step is perception, generally considered an automatic operation, in
22


105
that impulsive subjects had significantly lower correct scores than reflective subjects,
regardless of presentation speed.
Hypothesis 7. There are no significant differences in verbatim test scores for those
subjects who are determined to possess impulsive or reflective tendencies.
In order to further investigate the differences in scores obtained for impulsive or
reflective styles, a one-way analysis variance was developed (Table 16). An analysis of
Table 16
Source Table of Analysis of Score Variance for Verbatim and Gist and Cognitive Style
SS (Type III)
df
MS
F
Sig.
GIST
Between
12.089
1
12.089
4.410
.038*
Groups
Within Groups
348.144
127
2.741
Total
360.233
128
VERBATIM
Between
14.267
1
14.267
2.925
.090
Groups
Within Groups
619.423
127
4.877
Total
633.690
128
* Significant at (p<.05)
was performed and obtained an F (1, 127)of 2.925 for verbatim scores. The F ratio for
between subjects was not significant at (p<.05), therefore, the null hypothesis was not
rejected.
Hypothesis 8. There are no significant differences in gist test scores for those
subjects who are determined to possess impulsive or reflective tendencies.
The same analysis of variance used for Hypothesis 7 was used to show both
verbatim and gist scores (Table 16). The analysis obtained an F (1, 127) of 4.410 for
verbatim scores. The F ratio for between subjects was significant at (p<.05), therefore,
the null hypothesis was rejected.
A means and standard deviation table (Table 17) was developed to further


126
Block, J., Block, J. H., & Harrington, D. M. (1974). Some misgivings about the Matching
Familiar Figures Test as a measure of reflection-impulsivity. Journal of
Developmental Psychology, 10, 611-632.
Boyden, J. G., & Gilpin, A. R. (1978). Matching familiar figures test and Stroop test
performance in adults. Perceptual and Motor Skills, 46, 854.
Brainerd, C. J., & Reyna, V. F. (1990). Grist of the gist: Fuzzy-trace theory as the new
intuitionism. Developmental Review, 10, 3-47.
Brainerd, C. J., & Gordon, L. L. (1994). Development of verbatim and gist memory for
numbers. Developmental Psychology, 30(2), 163-177.
Braverman, C. (1969). The world of kinestasis. Media & Methods, 6(3), 61-62.
Bridgeman, B. (1980). Generality of a "Fast" or "Slow" test-taking style across a variety
of cognitive tasks. Journal of Educational Measurement, 17, 211-217.
Brown, R. & Kulick, J. (1977). Flashbulb memories. Cognition, 5, 73-99.
Brumby, M. N. (1982). Consistent differences in cognitive styles shown for qualitative
biological problem-solving. British Journal of Educational Psychology, 52, 244-
257.
Bruning, R. H., Shraw, G. J., & Ronning, R. R. (1999). Cognitive psychology and
instruction. Upper Saddle River, NJ: Prentice-Hall.
Bryant, J., & Rockwell, S. C. (1991). Evolving cognitive models in mass communication
reception processes. In J. Bryant, Zillman, D. (Ed.), Responding to the screen:
Reception and reaction processes. Hillsdale, N.J.: Lawrence Erlbaum Associates.
Bums, B. (1992). Percepts, concepts, and categories: The representation and processing
of information. Amsterdam: Elsevier Science Publishers.
Caims, J., & Cammock, T. (1984). The 20-Item Matching Familiar Figures Test. (ERIC
Document Reproduction Service: No. 015681-4).
Calvert, S. L. & Scott, C. (1989). Sound effects for children's temporal integration of
fast-paced television content. Journal of Broadcasting and Electronic Media,
33(233-246).
Campbell, D., & Fiske, D. (1959). Convergent and discriminate validation by the multi
trait/multi-method matrix. Psychological Bulletin, 56, 81-105.


44
attention is subject to habituation that explains, for example, how people become
accustomed to living in cities and with noises and after a time becomes accustomed to
them to the point of eventually being able to block them out without loss of focus on
other tasks at hand. OR Theory might explain how youths of today might have become
accustomed to fast-paced media so that they can able to learn from it more efficiently
than their elders who are generally less exposed to these new media techniques.
These new ideas about arousal also encompass the concept that attention (and
subsequent arousal) in humans involves mechanisms that are used differently depending
on the task to be performed. Lang et al. (1999) differentiated between cortical and
emotional arousal in which the former can cause the body to bring more attention to a
situation rather than less. However, even cortical arousal may have its limits. In
reviewing Zillmans limited capacity model, Lang proffered four related principles:
1. A viewer will allocate an overall level of processing resources to the
complete viewing task based on goals, interests, etc.
2. A viewer's goals influence the proportion of resources allocated to the
various sub-processes (such as storage & retrieval).
3. The structural and content features of the message elicit orienting behavior
and the automatic allocation of resources to encoding.
4 The content and structural attributes of a message can also elicit arousal,
which results in the automatic allocation of resources to encoding and to
storage.
Zillman (1991) admitted that when there are insufficient resources available (i.e.,
cognitive overload) to carry out all the sub-processes, some aspects of cognitive
processing will be performed less well. This idea of finding optimum levels of arousal to
enhance motivation and performance was further developed by Winifred Hill (1985) who
developed a series of laws and theories that included several about stimulating students


59
towards reading than with television, but they rated the latter as more realistic and more
efficient. Greater mental effort requirements attributed to reading seemed to result in
greater perceived demand (PDC) towards that medium. Conversely, as these same
students perceived less mental effort was required towards television, they exerted a
correspondingly less effort and paid less attention. The amount of effort exerted was also
in direct proportion to preconceived notions about ones perceived self-efficacy about
either medium. The more comfortable they were with the medium, they less effort was
exerted. While Salomons studies showed some overall weakness due to some
discrepancies between self-reports and actual performance, his studies were backed up by
several others. For example, Ksobiech (1976) reported that pre-conceived task demands
affected the proportion of time students requested to see the visual portion of a video
presentation or only the audio portion, based on whether they were told that they were
going to be examined on the material, or were only asked to observe it for evaluation.
Students sought the video or audio source that maximized the purpose for which they
were to receive the material. Krendl and Watkins (1983) found that students viewing
television for entertainment value differed in the number of items they were able to recall
from it from those who were asked to view a presentation to learn from it.
Accordingly, some researchers (Langer, Blank, & Chanowitz, 1978; Langer &
Imber, 1979; Schank & Abelson, 1977) hold the view that most encounters with
television cause viewers to generally approach it with a sort of mindlessness that can
undermine its role in an educational setting. It was very important for the success of the
current study that the subjects do not consider the activity of viewing the video and
subsequently answering the questions that followed as an entertainment activity,


24
mind being able to discriminate parts to the whole and make other spatial determinations
that are described further later in this chapter. But first, to understand more fully the
Gestaltist view of visual intelligence and how it relates to this study, it might be best to
consider some of the literature concerning the neurological process involved with vision.
Neurology of Vision
According to Murch (1973), biologists define vision as a process where the eye
sees the world through a biological process of photic radiation, an energy source that
stimulates the eye. The process begins as reflected light that bounces off objects in the
environment. This optic array is focused by the cornea and lens onto the visual field on
ones retina that lines the back of the eye (Seward-Barry, 1997). These images are then
transmitted to the brain through various routes, one being the cortex/limbic system. This
system provides stimulus to the brain, including the amygdala: the sub-cortical region
within the temporal lobe (LeDoux, 1991). According to Seward-Barry, researchers have
identified four parallel systems involved in the different attributes of vision: one each for
motion and color, and two for form. Goldstein (1989) points out that perception is based
not on direct contact with the environment, but on the brains contact with electrical
signals the represent the environment. We can think of these electrical signals as forming
a code that signals various properties of the environment to the brain (p. 50). In other
words, perception as a biological function may be described as being based on a coding
system that takes in the visual signal and is interpreted by the brain. LeDoux (1991)
viewed this interpretation process as a dual operation. He proposed that sensory
perceptual processing takes two routes. The first travels along the amygdala and readies
the body to react, even before it recognizes the need to do so. The second route is through


APPENDIX B
SCREEN SHOT FROM TEST INSTRUMENT SHOWING SAMPLE VERBATIM
QUESTION
123


99
The analysis resulted in an F (2, 201) of .317 (see Table 4). An F ratio of .317 is
not significant at (p <.05), therefore, this null hypothesis was not rejected.
Table 4
Source Table of Analysis of Score Variance by Presentation Speed
Source
SS (Type III)
df
MS
F
sis-
Between Groups
SPEED
6.377
2
3.189
.317
.729
Error
2022.329
201
10.061
Total
2028.706
203
The means and standard deviations for presentation speed are presented in Table
5. In Table 5, Fast represents the results for those who watched the video presented at
its original speed (averaging 300-400 milliseconds per picture), Medium at 1/3 of the
original (approximately 500-700 milliseconds), and Slow (at approximately 1/2 the
original speed (each picture was presented at approximately 1 second each).
Table 5
Score Means and Standard Deviation for Presentation Speed
SPEED
Mean
SD
N
Fast
12.99
3.18
75
Medium
12.70
3.05
69
Slow
12.57
3.30
60
Total
12.76
3.16
204
Flypothesis 2. There are no significant differences in overall test scores between
male and female subjects in the overall sample set.
The analysis resulted in a between subjects effect F (1, 202) of 1.542 (see Table
6). Because an F ratio of 1.566 for this main effect is not significant at (p <.05), this null
hypothesis could not be rejected. The means and standard deviations are presented in


CHAPTER 3
METHODOLOGY
Introduction
In this study the researcher attempted to elicit information about how differing the
presentation speed might affect immediate memory for the pictorial representations found
on the video montage. For this portion of the study, scores for verbatim and gist
memories were pooled together. The researcher then investigated whether the same
pooled scores differed between males and females taking the test. Next, the scores for
verbatim and gist memories were segregated and reviewed in light of the differing
presentation speeds. It is recognized that gist memory is a less exact indication of
immediate memory about individual pictures than is verbatim memory. However, it was
anticipated that a review of gist memory might provide additional overall insights to
memory in that it requires viewers to draw certain conclusions about the contextual
meanings much like a typical reading comprehension test of text-based materials. Lastly,
the researcher investigated the effect cognitive style might have on the verbatim scores
and gist scores taken together, as well as for each of them individually.
A computerized instrument was specifically prepared for this study that displayed
the video at one of three different presentation speeds and then immediately tested
subjects to determine what they remembered from the video. The program was encoded
to score verbatim and gist memory scores separately and also combined into a single
81


8
its meaning. In an earlier attempt to extend this notion to television, Pearl (1982)
discussed the relationship between form and content and admitted that it is the form (that
is, the way it uses verbal and linguistic codes) not the content of television that makes
it a unique communicative medium. However, she also cautioned that form and content
cannot always be distinguished no more than grammar and meaning in any verbal
language can (p. 24). She went on to say that some forms are unique to a particular
medium and apply syntactical meaning only in the context of that medium. For example,
slow motion is not real and its meaning must be learned. Other researchers have shown
that these unique formats, once learned, generally become used by people in their own
thinking (Barnett, 2000). That is perhaps why, for example, when one applies slow
motion to a video message, it generally carries some contextual or emotional connotation
each time it is used. The current study tries to apply the same logic to fast cuts and video
montage to see if this presentation format can be interpreted in such a way as to portray
some implied meanings of its own.
Previous research into using increased or compressed presentation speed and
movement in multimedia has been the subject of controversy in the literature.
Some studies of commercial broadcast video have shown that the relatively rapid
presentation speed of programs may have a deleterious effect on recall (Alwitt, Anderson,
Lorch, & Levin, 1980; McCollum & Bryant, 1999). Other research has shown that the
instructional value of video is aided by a systematic presentation speed of information as
a presentation strategy (Comstock, Chafee, Katzman, McCombs, & Roberts, 1978). Still
others have shown that presentation speed and rhythmicity in leisure-time media can
actually heighten enjoyment, enhance motivation, and can play an important part in


77
Summary
In researching the current trends of the so-called media effects, one thing has
become evident. It is the newer structures and the form (of which rapid presentation is a
part) of the montage video segments that continue to play an ever-increasing role in how
messages are delivered. This view is not a new one. McLuhan (1964) and Salomon
(1979) predicted media effects long before technology caught up with televisions
promise.
Flowers (1995) cited research as early as the 1970s that contended that educational
environments might be too visuo-centric (p. 570), indicating that looking into visual
processing may have been overdone in many previous studies and that its role in learning
might have been exaggerated. In some respects, Flowers might be correct. Singer (1980)
admitted that many previous assertions regarding textual versus visual image processing
had yet to be supported directly by research evidence, and suggested the need for further
experiments on length of video sequences and its effect on retention of content (or at least
intelligent comprehension of content). Intraubs (1999) experiments appear to answer that
need. She found that humans possess a remarkable capability (p. 67) to understand
scenes that are presented at a pace far more rapidly than normal scanning, so long as the
scene changes overlap with at least somewhat related content.
Over the past decade, media researchers have started to look at the impact new
media have on cognition in a casual viewing environment (Basil, 1994; Hawkins et al.,
1997; Hill & Lang, 1993; Hitchon et al., 1994; Walma van der Molen & Van der Voort,
2000; Watt & Krull, 1977; Zillman, 1991). Some are also beginning to conduct more
extensive research on the effects on memory of editing techniques and the content found


129
Estes, W. K. (1994). Classification and cognition. New York: Oxford University Press.
Flannagan, P. (1998). Aurally and visually guided search in a virtual environment.
Human Factors, 40 (3), 461-469.
Flowers, J. H. (1995). Musical versus visual graphs: Cross-modal equivalence in
perception of time-series data. Human Factors, 57(3), 553-570.
Gardner, H., Howard, V., & Perkins, D. (1974). Symbol systems: A philosophical and
educational investigation. In D. R. Olson (Ed.), Media and Symbols: The forms of
expression, communication, and education (pp. 27-55). Chicago: University of
Chicago Press. The Seventy-Third Yearbook of the National Society for the Study
of Education.
Gentry, C. G. (1998). Introduction to instructional development. Belmont, CA:
Wadsworth Publishing.
Gibson, E. J. (1969). Principles of perceptual learning and development. New York:
Appleton-Century-Crofts.
Gibson, J. J. (1979). The ecological approach to visual perception. Boston: Houghton
Mifflin.
Goldstein, E. B. (1989). Sensation and perception, (3rd Edition). Pacific Grove, CA:
Wadsworth.
Goodglass, H. (1971). Stimulus duration and visual processing time. Perceptual and
Motor Skills, 33, 179-182.
Green, K. E. (1985). Cognitive style: A review of the literature Chicago, IL: Johnson
O'Connor Research Foundation, Human Engineering Lab. (ERIC Document
Reproduction Service No. ED 289 902).
Gross, L. (1974). Modes of communication and the acquisition of symbolic competence.
In D. R. Olson (Ed.), Media and symbols: The forms, expression, communication,
and education (pp. 56-80). Chicago: University of Chicago Press. The Seventy-
Third Yearbook of the National Society for the Study of Education.
Gummerman, K., & Gray, C. R. (1972). Age, iconic storage, and visual information
processing. Journal of Experimental Psychology, 13, 165-170.
Haber, R. N. (1970, May). How we remember what we see. Scientific American, 221,
104-112.
Hawkins, R. P., Pingree, S., Bruce, L., & Tapper, J. (1997). Strategy and style in attention
to television. Journal of Broadcasting & Electronic Media, 41, 245-264.


21
today think. This appears to be at odds with the linear thinking processes associated with
print media. The current study, through the questions posed in the current chapter and a
review of the literature found in the next, attempts to find out how significant these
changes in perception of new media are.


29
initial pattem/object stimulus. Through studies on rhesus monkeys, researchers have been
able to point out which cells within the brain are responsible for certain functions such as
pattern recognition. Once identified, these brain cell studies were translated to human
subjects, who were tested to determine how fast the brain is able to react, recognize, or
otherwise discriminate between faces of individuals. Researchers found that after only a
20-millisecond delay, test subjects could recognize and discriminate face identity,
indicating that the minimum amount of time that neurons need to be active in order to
mediate recognition and discrimination is in the 20-30 millisecond range (Tovee). The
next stage to processing this information is the brain reacting to these stimuli and
initiating reactionary motor commands. Independent follow-up research, based on
clinical tests performed on epileptic patients in preparation for brain surgery, indicates
that this process takes somewhere in the 150-200 millisecond range (Allison, et al.,
1994). These researchers confirmed that the whole process of pattern recognition and
reaction takes in the neighborhood of 400-500 milliseconds, indicating that, not only can
the visual system process information rapidly, it can also rapidly update and modify
responses, based on previous experiences (Tovee). For example, people can rapidly and
accurately reconstruct meaningful objects out of fragmentary or ambiguous evidence
(Ramachandran, 1994). These findings appear to denigrate traditional assumptions about
perception that awareness must mediate between stimulus and response. In fact,
researchers now know that many responses are automatic and emotional, and may bypass
cognitive processing all together (Seward-Barry, 1997). The fact that initial responses can
precede cognitive processing has caused a renewed interest in subliminal research, which
deals with subconscious message registration. However, results from research in this area


30
have been inconclusive due to the inability to date of researchers to quantitatively
delineate and measure perceptual responses (Seward-Barry).
Most previous research into limits on perception and recognition of rapidly
presented stimuli dealt with verbal information (Coltheart, 1999). However, Intraub
(1999) worked extensively with scene recognition and memory. Citing experiments of
Potter, she suggests that abstract representations of scenes are used in the interpretation of
pictures and scenes. Using a technique called conceptual masking, she was able to show
that the attentional demands of new and meaningful pictures interfere with ones ability
to continue processing their predecessors. Perhaps, sharing the gestaltist view of closure,
Coltheart further demonstrates this phenomenon, referring to it as boundary extension
(p. 6). Even in rapidly presented stimuli, subjects indicated that they tended to remember
information that was not shown but was likely to have been present outside the pictures
boundaries. Further, Intraub produced experiments that showed that photographs, when
shown in rapid succession, (at a rate that mimicked ones ability to scan them) were
unable to be recognized by their viewers, even a few moments later. These studies seem
to report limitations on memory, rather than perception, shown with Intraubs studies of a
series of rapidly presented pictures that were separated by blank screens of varying
lengths of time. It appears that the longer time lags between meaningful pictures
permitted viewers to properly process, categorize, and consolidate their meaning. Potter
argued (Intraub) that a picture is identified within about 100 milliseconds. Until the item
is identified, it is vulnerable to visual masking by a new visual event (p. 52). Once
identification is complete, however, the pictorial representation is maintained in short
term memory for a few hundred milliseconds, long enough for it to be consolidated and


Recommendations 116
Summary 120
APPENDICES 122
Appendix A Screen Shot of MFFT-20 122
Appendix B Screen Shot of Sample Test Question 123
REFERENCES 124
BIOGRAPHICAL SKETCH 140
vi


23
which features are integrated into complete visual displays. Assimilation is the next level
of complexity and involves the recognizing and integrating of information into schemata
for later recall. Analogy is similar to assimilation but is an even more complex process
that involves the temporary assimilation of new information into abstract schemata that
embody other concepts that are peripherally related to it. Explained in this way, Winn
alluded to perception as a necessary but insufficient step in the whole learning process.
Bruning, Shraw, and Ronning, (1999) posited that the body holds perceptions in a
series of registers called memory so that perceptual analysis can occur before that
information is lost. When all is working well, perception allows humans to detect the
incoming stimuli and allocate attention to them. Through pattern recognition, individuals
may recognize a stimulus because of their previous experiences. However, on occasion,
pattern recognition may be limited, thereby preventing any learning. Marr (1982)
proposed a structural approach to the relationship between perception and intellectual
endeavors. He imagined that knowledge is stored in the brain as a set of statements,
templates, prototypes, or models of distinctive features of a particular object or class of
objects that has been perceived. Bruning et al. (1999) appeared to agree, stating that using
specific, structured descriptors as an approach to recall and recognize patterns may be
very helpful for teaching because it suggests that perception can be a guided activity.
Teachers can instruct students the proper knowledge needed for an accurate structural
description (p. 27). Although the visual field may be partially obstructed, the mind is
still able to make determinations about what it sees, using clues and mental mappings
(visual patterns) from previous experiences (Seward-Barry, 1997). Gestalt psychologists
refer to this process as visual intelligence and base their theories on this emphasis of the


41
counterbalance the more traditional ideas about encoding, and the verbal versus
nonverbal comparisons, and more importantly for this study, in analyses of the ability of
subjects to recognize and utilize information perceived in visual presentations (Haber,
1970; Loftus, 1979).
According to fuzzy-trace theory (Brainerd & Reyna, 1990), both gist and
verbatim memories are based on similar inputs, but are stored separately. Verbatim
memory may be associated with higher ability students, but some researchers have shown
that reasoning power can often be disassociated with strict recall of specific details.
Others have indicated that gist determination is associated with reasoning powers,
making it higher-level activity events (Reyna & Kieman, 1994). It follows then that gist
memory can be assessed independently without requiring that the subjects also score well
at verbatim memory and vice versa. Reyna and Kieman did point out that the only
exception to the verbatim-gist independence rule is the fact that success with verbatim
recall has been shown to be higher with linear text-based presentations as opposed to
pictures.
Although it is a relatively new theory, fuzzy trace serves to explain an important
concept of recognition and recall from non-verbal displays as used in the present study.
As Brainerd and Gordon (1994) pointed out, so little is known about the developmental
and functional relationships between verbatim and gist memories that it is necessary for
researchers to proceed step by step, investigating these relationships within delimited
classes of inputs in the hope that general conclusions will ultimately emerge (p. 163).
This gives rise to questions that investigate the effect cognitive style and presentation
speed might have on these two forms of memory determination. It is the intent of the


138
Voss, J. F., Tyler, S. W & Bisanz, G. L. (1982). Prose comprehension and memory. In
C. R. Puff (Ed.), Handbook of research methods in human memory and cognition
(pp. 349-392). New York: Academic Press.
Wagely, M. (1978). The effect of music on affective and cognitive development of sound-
symbol recognition among preschool children. Unpublished Doctoral
Dissertation, Texas Woman's University, Denton.
Walma van der Molen, J. H., & Van der Voort, T. H. A. (2000). The impact of television,
print, and audio on children's recall of the news: A study of three alternative
explanations of the dual coding hypothesis. Human Communication Research, 26,
3-26.
Walters, D. L. (1983). The relationship between personal tempo in primary-aged
children and their ability to synchronize movement with music. Dissertation
Abstracts International, 44 (02), 423A.
Watkins, J. M., Lee, H. B., & Erlich, O. (1978). The generalizability of the matching
familiar figures test. Paper presented at the Annual Meeting of the American
Educational Research Association (ERIC Document Reproduction Service No.
ED 175 882), Toronto, CA.
Watt, J. H., & Krull, R. (1977). An examination of three models of television viewing
and aggression. Human Communication Research, 3, 99-112.
Wertheimer, M. (1959). Productive Thinking (Enlarged Ed.). New York: Harper &
Rowe.
Wetzel, C. D., Radtke, P. H., & Stem, H. W. (1994). Instructional effectiveness of video
media. Hillsdale, NJ: Lawrence Erlbaum Associates.
Winn, W. (1982). Visualization in learning and instruction: A cognitive approach.
Educational Communications and Technology Journal, 30, 3-25.
Witkin, H. A., & Goodenough, D. R. (1981). Cognitive styles: Essence and origins. Field
dependence and field independence. New York: International Universities Press.
Wittrock, M. (1979). The cognitive movement in instruction. Educational Researcher, 8,
5-11.
Woodley, J. W. (1984). The role of non-visual information in visual perception: An
integration of research perspectives Austin, TX: (ERIC Document Reproduction
Service No. ED 283 134).


BIOGRAPHICAL SKETCH
Mr. Kenny holds a bachelors degree in Education from Niagara University in
New York and a masters degree in instructional technology/media: educational
technology from the University of Central Florida. Prior to entering into the doctoral
program at the University of Florida, he spent more than 20 years in the computer
services industry and in private business. He has spent the last six years teaching media
studies and technology on the secondary level in Orlando, Florida.
Mr. Kenny is an active member in the Florida Association of Media Educators
(FAME), and the Association of Educational Communications and Technology (AECT)
and has presented several papers at conferences sponsored by both of these organizations.
Mr. Kenny also published a textbook in 2001, entitled Teaching Television
Production in the Digital Age, a book dedicated to showing new teachers how to establish
television and media studies programs in their schools.
Mr. Kenny has been married to his wife, Sandra, for thirty years. He has two
married children.
140


68
dance, music, etc. and can be replete, resemble reality, and approximate notions of
notational ability, borrowing their power from other media. This same power to employ a
wide range of symbol systems may also define film and televisions potential
shortcomings, reflecting a fallacy in the belief that they always deliver to their potential
(Wetzel et al., 1994). For example, if a video screen is simply placed in front of the class
that displays nothing more than a talking head lecture covering an irrelevant content,
there will be little hope for realizing a valuable instructional delivery. All media carry
with them a systematized methodology for symbolic processing, but that varies in how
they are used, in what context, and by which individuals.
Symbol systems theory plays a significant role in formulating some of the
hypotheses for the current study. First, the pictures found in the depiction of historical
events are iconic in nature and, as such may be considered primary symbols. Second,
although the speed of delivery is considered rapid or fast (at about 200-300 millisecond
per image), the speed within this context is varied, which manipulates the amount of
mental effort required to comprehend the conceptual context. As the images are presented
in combination in a montage format, they represent reality to viewers in a way similar to
the way they think (i.e., in a streaming sense). Lastly, as outlined by Salomon (1979),
fidelity not withstanding, the images in the videos used in the current study are indeed
symbols in that they represent the sole source of reality about the historical events to the
viewers who were not live witnesses to those events. In addition, as per the
conceptualizations of Snow et al. (1965), the amount of information the subjects are able
to extract from the videos may well depend on the amount of previous experience they
have with this medium (television).


43
(1982) described four basic ideas about attention. Each is listed below with parenthetical
comments regarding relevance to the current study:
1. Mental operations related to attention take time to perform and the amount
of time required will correspond to fixed qualities of the stimuli (such as
presentation speed) and the discretionary strategies of individuals (such as
cognitive style).
2. Mental events occurring closely in time are processed successively.
(Montage video attempts to replicate these closely related mental events.)
3. Internal events can be studied by observing the amount of facilitation or
inhibition they cause and this process is hierarchical. (Immediate memory
feeds longer term recollection.)
4. Attentional processing favors stimulus change. An orienting reflex biases
people toward fresh or novel sources of stimulation. (Presentation speed
can act like an orienting reflex.)
The works of Reeves et al. (1986), Reeves and Geiger (1994), and Reeves and
Nass (1996) have centered on the individuality of the attentional reflex. They clearly
defined attention as "a psychological cognitive process that varies within individuals over
time" (Reeves et al., p. 254). These more modem theories look at attention as more of a
filtering device that is not exactly an all or nothing gateway. Further, there might be
different types of attention requiring different types of responses. For example, vigilance
tasks require people to wait for an event and then respond quickly. They decrease over
time and require individuals to attend to only one at a time. These are primitive orienting
cues that stem from primal man. Attentional preparation tasks give people cues or primes
before stimulus occurs. Perceptual intrusion tasks are those so arranged that people
cannot avoid attending to them. Finally, attention switching tasks are those in which
subjects must switch their attention from one task to another. Reeves et al. elaborated
further. They posited an Orienting Response Theory (OR) that suggests, in part, that


80
passages. What remains is trace memory for remembering the gist of the storyline, even if
the specific words or phrases (i.e., verbatim memory for the specific pictures) cannot be
recalled exactly.
The next chapter outlines the specific methodologies used in this study to
investigate some of the affects that variation of presentation speed, personal cognitive
style, and gender may have on memories for information presented in the passages. These
methods follow closely the ideas of looking at combined memories, and then breaking
these down into verbatim and gist recollections to see if there are any differences. After
analyzing the results, implications are reviewed and recommendations are made for this
initial look into memories for pictorial representations.


73
Olsons list of criteria for analyzing the theoretical potential of various media for use in
educational settings.
Effect of Fast Cuts on Memory
Reeves et al. (1986) re-defined the result of attentional changes in humans,
referring to them Orienting Responses (ORs). These types of responses help to explain,
for example, how humans become habituated to living in large cities with lots of loud
noises, or how one becomes so used to the common sounds found in his or her home (like
a cuckoo clock) and begins to tune them out. Reeves et al. identified several
characteristics of ORs that have implications for evaluating the effectiveness of fast cuts
in an educational setting. Reeves et al. also related these changes to cortical arousal, in
which an individual may have some type of control over and learn how to manipulate for
his or her own purposes. They noted that there is some evidence to suggest that those
who habituate slower perform better on audio and visual tasks than those who habituate
quickly. They also documented studies that showed that high cortical arousal in adults is
related to an increased ability to remember information both of which are premises of
the current study. These ideas about ORs also formed the basis of some of the
recommendations found in Chapter 5.
Presentation Speed versus Content
Lang et al. (1999) more recently discussed adding interesting and arousing
content as having a positive effect on cortical arousal and, therefore, recall and
recognition. She also noted that too much arousal can be harmful. In describing a limited
capacity model, she outlined a sequence where viewers allocate an overall level of mental
processing resources to the complete viewing task based on their goals, interests, etc. In


ACKNOWLEDGMENTS
Special appreciation is extended to Jeffry Hurt, David Miller, and David Ostroff,
members of my doctoral committee, for their support in this undertaking, and all their
advice and counsel as I moved through the process of completing my degree. I took
classes from all of them. With each course, 1 became more inspired and awed by their
insight, knowledge, and enthusiasm for their profession. They have inspired me to be just
like them.
I would especially like to thank Lee Mullally, my committee chair, for his
direction and support, and especially his determination not to let me take any shortcuts, or
not to give my best efforts in the many classes and seminars I took from him, the reviews
and edits for the textbook I wrote, and especially this study. He has my unending
appreciation for his extraordinary patience and ability to keep me focused.
I am grateful to Chuck Braverman, whose kinestasis film was used in this study. I
saw it almost 30 years ago and knew then it would somehow become a part of my life. I
am particularly grateful to Randolph Wright from Pyramid Media, copyright holder of
American Time Capsule, who graciously granted me permission to use it in this study. I
am also thankful to Mitchell Stephens, whose book, The Rise of the Image, the Fall of the
Word became the catalyst for this study. Other producers and researchers in new media
like Jeff Scher and Annie Lang have also inspired me with their insights into visual
perception, fast-cuts, and montage video. I am happy and lucky to have met them.
11


131
Jankowski, G., & Fuchs, D. (1995). Television today and tomorrow: It won't be what you
think. New York: Oxford University Press.
Jonassen, D. H. (1996). Computers in the classroom: Mindtools for critical thinking.
Englewood Cliffs, NJ: Prentice-Hall.
Joyce, B., Weil, M., & Calhoun, E. (2000). Models of teaching (6th ed.). Boston: Allyn
and Bacon.
Kagan, J. (1965). Impulsive and reflective children: Significance of cognitive tempo. In J.
D. Krumboltz (Ed.), Learning and the Educational Process. Chicago: Rand
McNally.
Kagan, J. (1966). Reflection-impulsivity: The generality and dynamics of conceptual
tempo. Journal of Abnormal Psychology, 71, 17-24.
Kahneman, D. (1973). Attention and effort. Englewood Cliffs, NJ: Prentice Hall, Inc.
Kay, A. (1999). User interface: A personal view. In B. Laurel (Ed.), The Art of Human-
Computer Interface Design (13th ed., pp. 191-207). Reading, MA: Addison-
Wesley.
Keller, J. M. (1983). Motivational design of instruction. In C. M. Reigeluth (Ed.),
Instructional Theories and Models: A Cherview of Their Current Status New
York: Lawrence Erlbaum Associates.
Keller, P. F. G. (1976). Pictorial memory processes under conditions of the kinestasis
fdm. (Doctoral Dissertation, Southern Illinois University, 1976). Dissertation
Abstracts International, 38 (05), 2498.
Kini, A. S. (1994). Effects of cognitive style and verbal and visual presentation modes on
concept learning in CBI. Paper presented at the Annual Meeting of the American
Educational Research Association. (ERIC Document Reproduction Service No.
ED 371 032), New Orleans.
Koffka, K. (1935). Principles of gestalt psychology. Reprint 1963. New York: Harcourt,
Brace & World.
Kogan, N. (1971). Educational implication of cognitive styles. In G. S. Lesser (Ed.),
Psychology and Educational Practice Glenview, IL: Scott Foresman.
Kozma, R. B. (1986, Spring). Implications of instructional psychology for the design of
educational television. Educational Communications and Technology Journal. 34,
11-19.


102
Because the F ratio for verbatim recall was not significant at (p<.05), this null
hypothesis could not be rejected. The means and standard deviation are presented in
Table 14.
Hypothesis 5. There are no significant differences in gist test scores for the overall
sample set, based on presentation speed of the video.
The analysis of gist scores resulted in an F (2,201) of 5.491 (see Table 11).
Because the F ratio for verbatim recall was significant at (p<.05), this null hypothesis was
rejected.
Table 11
Source Table of Analysis of Variance for Gist Scores for Speed
Source
SS (Type III)
df
MS
F
Sig.
Between Groups
32.462
2
16.231
5.491
.005*
Error
594.127
201
2.956
Total
626.588
203
* Significant at (p<.05)
Because the results were found to be significant, and because there were more
than two groups to be compared, a Bonferroni test was run to compare individual pairings
of gist scores to each of the individual presentation speeds. Table 12 shows that the pair
wise comparisons are significant for gist scores when comparing Slow and Fast
speeds, but not significant between Medium and Slow or Medium and Fast.
In order to further compare verbatim and gist scores, a means and standard
deviation table (Table 13) is presented. Table 13 shows a mean score for fast speeds of
6.01 out of 10 possible gist responses for those viewing the video at the fast speed and
5.05 (out of 10) mean score for those viewing it at the slow speed. Table 13 shows that,


45
interest through arousing content. Among her assumptions were that optimal arousal
involves three factors: novelty, complexity, and intensity. The level of arousal (i.e.,
attentional demands) is function of an inter-reaction among these three factors. Highly
intensive and novel stimuli are more arousing than those that are weaker or more
familiar. Moderate intensity levels of arousal are more reinforcing than either high or low
levels of either the newness or novelty of content. According to Lang et al. (1999), it is
important to find a consistency and/or balance between the level of external arousal
applied by the presentation vehicle and any arousal that might be inherent in the content.
If task or conceptual complexity outweighs its novelty, a more moderate stimulus would
be more reinforcing for cognition. If, on the other hand, intensity and complexity are
relatively equivalent in strength, a change in content (or novel form of content) is what
will regain a viewers attention. Hill summarized by stating that a highly novel stimulus
will [sic] be more reinforcing if it is mild and simple, whereas among very familiar
stimuli the more intense and complex ones [sic] tend to be preferred (p. 176).
Lang et al. (1999) and Lang et al. (2000) elaborated further on the correlation
between intensity and complexity of content, relating the former to presentation speed and
the number of structural features occurring in a message. Again, referring to the limited
capacity model she suggested that, as the pacing in a message increases, the number of
orienting responses (ORs) elicited by that message (and the amount of information
available to be encoded) will also increase. Because of these increased ORs, a viewer will
automatically allocate more resources to encoding fast paced messages. She also
proposed that viewers control some aspects of the resource allocation process by making
decisions such as whether to watch, how carefully to watch, how hard to try based on


125
Arizmendi, T., Paulsen, K., & Domino, G. (1981, Spring). The Matching Familiar
Figures Test: A primary, secondary, and tertiary evaluation. Journal of
Psychology, 812-818.
Amheim, R. (1974). Virtues and vices of the visual media. In D. R. Olson (Ed.), Media
and Symbols: The forms, expression, communication, and education (pp. 180-
210). Chicago: University of Chicago Press. The Seventy-Third Yearbook of the
National Society for the Study of Education.
Arnold, R. S. (1996). Instructional effectiveness of compressed motion video signals.
(Doctoral Dissertation, University of Central Florida, 1996). Dissertation
Abstracts International, 51 (03), 1103A.
Ault, R. L., Mitchell, C., & Hartmann, D. P. (1967). Some methodological problems in
reflective-impulsivity. Child Development, 47, 227-231.
Ayabe, H. I. (1973). Measuring reflection-impulsivity accurately. Paper presented at the
Annual Meeting of the American Educational Research Association, February 25
- March 1, New Orleans, LA, (ERIC Document Reproduction Service No. ED
079 333).
Bargh, J. A. (1988). Automatic information processing: Implications for communications
and effect. In L. Donohew, Syphder, H. & Higgins, E. T. (Ed.), Communication,
social cognition, and affect (pp. 9-32). Hillsdale, NJ: Lawrence Erlbaum
Associates.
Barnett, B. (2000). The impact of slow motion video on viewer evaluations of television
news stories. News Photographer, 55(7), 4-11.
Bartlett, F. C. (1932). Remembering. Cambridge: Cambridge University Press.
Basil, M. (1994). Multiple resource theory I: Application to television viewing.
Communication Research, 21, 177-207.
Berio, D. K. (1960). The process of communication: An introduction to theory and
practice. Toronto, CA: Holt, Rinehart & Winston.
Berry, L. H. (1991). Visual complexity and pictorial memory: A fifteen-year research
perspective. Paper presented at the Annual Convention of the Association for
Educational Communications and Technology. (ERIC Documentation
Reproduction Service No. 334 975).
Biederman, L, Rabinowitz, J. C., Glass, A. L., & Stacy, E. W. (1974). On the information
extracted from a glance at a scene. Journal of Experimental Psychology, 103,
597-600.


101
Table 8
Source Table of Analysis of Variance of Speed versus Gender
Source
SS (Type III)
df
MS
F
Sig.
Between Groups
SPEED
8.437
2
4.219
.419
.659
GENDER
15.540
1
15.540
1.542
.216
SPEED GENDER
8.766
2
4.383
.435
.648
Error
1995.428
198
10.078
Total
35268.000
204
Table 9
Score Means and Standard Deviation for Speed versus Gender
SPEED
GENDER
Mean
SD
N
Fast
Male
13.56
3.10
34
Female
12.51
3.20
41
Total
12.99
3.18
75
Medium
Male
12.97
3.46
38
Female
12.35
2.48
31
Total
12.70
3.05
69
Slow
Male
12.57
3.46
35
Female
12.56
3.12
25
Total
12.57
3.30
60
Total
Male
13.03
3.34
107
Female
12.47
2.94
97
Total
12.76
3.16
204
Hypothesis 4. There are no significant differences in verbatim test scores, for the
overall sample set, based on presentation speed of the video.
The analysis of verbatim scores resulted in an F (2, 201) of 1.082 (see Table 10).
Table 10
Source Table of Analysis of Variance for Verbatim Scores for Speed
SS (Type III)
df
MS
F Sig.
Between Groups
12.035
2
6.018
1.082 .341
Error
1117.592
201
5.560
Total
1129.627
203


28
especially, closure) explain how humans can often be visually fooled by slights of hand
or other artistic manipulations of visual stimuli, how they are able to see paintings
differently, or be taken in by virtual reality. The mind tends to close off partially
completed geometrical patterns, and/or create its own sense of reality based on what it
thinks it sees (Seward-Barry). These same four laws also explain how montage video
provides the continuity needed to be able to tell a story visually.
Fast SeeinR
Evidence points to the ability of humans to be able to process visual patterns very
rapidly (Coltheart, 1999; Tovee, 1998). Theory suggests that impulses processed in the
visual cortex are able to process almost 87% of perceived information within 400
milliseconds. According to Tovee, clinical researchers also found that this operation is
not linear. In fact, the majority of information (around 67%) is actually available within
the first 20-50 milliseconds of the initial spike train (p. 143). The investigation of what
is understood from rapidly presented sequences of visual stimuli began with the research
of Potter and Levy, in the late nineteen sixties and Forster in the early seventies
(Coltheart). Their studies suggest that most information encoded in an impulse is
available to the brain to act on at the very beginning moments of reception. Forster
developed a technique of rapid presentation of sequences termed RSVP (rapid serial
visual presentation). Potter and Levy provided a comprehensive review of RSVP
methodology, which proposed that humans possessed a natural short-term memory
system (Coltheart). These researchers confirmed their studies by placing constraints on
how long the visual system has to process stimulus in both forward and backward
masking functions in which the masking stimulus follows very quickly the onset of the


104
of the sample was systematically excluded due to these subjects being cast as either fast-
accurate or slow-inaccurate, which placed them outside the parameters set forth by the
administrative instructions that accompanied the MFFT-20. An analysis of variance was
performed and obtained an F (2,123) of 6.560 for the main effect for cognitive style. The
F ratio for between subjects was significant at (p<.05), therefore, the null hypothesis was
rejected. The means and standard deviations are presented in Table 15. This table shows
Table 14
Source Table of Analysis of Score Variance by Presentation Speed and Cognitive Style
Source
SS (Type III)
df
MS
F
Sig.
SPEED
2.968
2
1.484
.160
.852
STYLE
60.870
1
60.870
6.560
.012*
SPEED STYLE
7.380
2
3.690
.398
.673
Error
1141.340
123
9.279
Total
21733.000
129
* Significant at (p<.05)
Table 15
Means and Standard Deviation for Overall Scores for Style
SPEED
STYLE
Mean
Std. Deviation
N
Fast
Impulsive
11.50
2.09
22
Reflective
13.50
3.31
24
Total
12.54
2.94
46
Medium
Impulsive
12.36
3.86
25
Reflective
13.24
2.14
21
Total
12.76
3.19
46
Slow
Impulsive
11.82
3.30
17
Reflective
13.10
3.02
20
Total
12.51
3.18
37
Total
Impulsive
11.92
3.17
64
Reflective
13.29
2.85
65
Total
12.61
3.08
129


33
actually was that an eagle is perched atop of a tree. In this case, visual perception
changed the actual perceived wording of the sentence. His studies were later confirmed in
additional studies performed by Gwen Nugent (1982) who also found that visual images
are recalled more readily than verbal texts. Her studies have been confirmed several times
by those evaluating the so-called Stroop effect where subjects incorrectly name a text
message by the actual font color of the text, rather than the color spelled out by the text
message (Shalev & Algom, 2000). Plass, Chun, and Mayer (1998) were less conclusive.
They developed a generative theory stated that learners actively selected relevant verbal
and visual information, organized the information into coherent mental representations,
and then integrated with one another these newly constructed visual and verbal
representations. Mayer, Bove, Bryman, Mars, and Tapangco, (1996) reinforced this idea
in studies of science curricula and found that there existed a coherence between the two,
so long as they complimented each other and that each was specific to their purpose.
Another way to look at the coding requirements between verbal and non-verbal
representations is that they are simply different from one another, with neither taking
precedence. In his studies, Singer (1980) found that verbal coding is a sequential process,
whereas the everyday sights and sounds of everyday life is a parallel operation that
encompasses a range of events within the same time. While its strengths make verbal
encoding more efficient for memory, they also limit its functionality in non-textual (i.e.,
non-reading) experiences.
Another comparison in the research of relative strengths or weaknesses of the
various encoding methods occurs between vision and audition; with neither being the
unanimous choice of researchers. Nazarro and Nazarro (1970) tried to equate aural and


3
2000). While many theorists have looked into the use of video, most of them had little
good to say about its ability to bring anything new to the table with regards to using it as
a communicative medium (Berio, 1960; Calvert & Scott, 1989; Ide, 1974; Kozma, 1986).
This may have been due to the limitations imposed by the technology in use at the time.
Recent technological advances in commercial television production techniques allow
todays producers to readily integrate fast-cuts and montage and provide new tools to
more easily communicate complex thought using a pictorial narrative structure. A
correctly constructed rapid-cut montage passage has been found to add clarity because of
the phenomenon that the interpretive whole of a montage segment literally communicates
more than the sum of its parts (Hitchon, Druckler, & Thorson, 1994; Stephens, 1996). In
other words, it is the composite whole of all the visual images in a passage or segment
when considered all at once that gives it extended meaning. In addition, newer editing
techniques have evolved that emphasize the perceptual continuity of a rapidly presented
image-based narrative structure, rather than the classical point of view of editing that
stressed the importance of applying strict rules in order to obtain slow and smooth
transitions between successive shots (dYdewalle & Vanderbeeken, 1990).
One of the specific problems being addressed in the current study is to update the
thinking with regards to using video as an instructional medium in light of these new
technological developments. A review of the literature of the past twenty years has
already yielded some studies into the impact of rapid video editing in an educational
setting (Stephens, 1996; Wetzel, Radtke, & Stem, 1994). However, more recent advances
in technology that have superceded those studies and an increased usage of rapid-cuts and
montage production techniques in commercial television programming have created the


19
activity. In the assessment of reading comprehension, a word is often associated with the
unit of analysis (Murdock, 1982). In this study, individual picture frames are considered
as being the minimal unit of analysis. These may be equated to a paragraph (Corcoran)
for purposes of comparing the results to reading comprehension-type assessments.
Orienting Response (OR) is the automatic response to a stimulus that helps an
individual focus attention on that stimulus and is often captured through psychometric
means (i.e., an attentional blink, or a look (Lang & Basil, 1998; Lang et al., 1999; Reeves
et al., 1986; Thorson & Lang, 1992). An OR can also cause an overt arousal response that
may be detrimental to recall. In this study the orienting response is considered to be a
potential confound and is dealt with in the methodology of the study. Subjects will be
provided a short preview of the video presented at the same speed in order for them to get
used to the format.
Presentation speed is the rate of scene changes or edits of a video media
presentation and is the operative term utilized in this study. In many studies used to
support the premises in this study, pacing has been the operative term used to refer to the
speed of message presentation (Corcoran, 1981; Hill & Lang, 1993; Lang et al., 1999;
Lang et al., 2000; McCollum & Bryant, 1999; Reeves et al., 1986; Schale, 1971,
December; Wagely, 1978; Zillman, 1991). The term pacing also carries certain pre
conceived connotations in educational circles that have nothing to do with the premise of
the current study. In order to avoid confusion, all references to pacing were replaced in
the current study with the terms presentation speed, or speed of message presentation.
Screen media are that which use controlled exposure of sequenced images on a
screen (either on a video or movie screen), often in fixed relationships to speech, music,


11
similar instructional outcomes in a more efficient and stimulating manner -something
found by some researchers to be a more practical outcome measurement for using video
media than achievement alone (Kini, 1994; Peck, 1987).
Background of the Study
Younger students live in a media-centric world. Researchers have shown that
youth today spend more time watching television and movies than most any other leisure
time activity (Pearl, 1982). In addition, the presentation speed of passages on commercial
television has increased significantly in the past fifty years (Stephens, 1996). Researchers
have shown that viewers automatically learn to cope with symbolic presentation methods
through repeated exposure to television and visual patterns (Abelman, 1995; Bargh,
1988; Carr, 1982). Because of their increased exposure to rapid sequence and
presentation speed brought on by fast-cuts/montage found in television programs aimed
at youth, it may also be assumed that these individuals might be able to comprehend these
messages on a much wider scale than can their adult counterparts. One cannot continue to
assume that exposure to video, regardless of presentation speed, is simply a passive
viewing activity. These advancements in media technology are changing the way viewers
look at and interpret video media. In addition, the widespread availability of production
techniques provides easy access to capabilities that allow people to use video media to
easily create their own content. It has been widely shown that these acquisition (i.e.,
production) opportunities also increase exponentially ones ability to comprehend content
delivered in like form (Tyner, 1998).


REFERENCES
Abelman, R. (1995). Gifted, LD, and gifted/LD children's understanding of temporal
sequencing on television. Journal of Broadcasting and Electronic Media, 39, 297-
312.
Allison, T., Ginter, EL, MacCarthy, G., Nobre, A.C., Puce, A., Luby, ML, et al. (1994).
Face recognition in human extrastriate cortex. Journal of Neurophysiology, 71,
821-825.
Alwitt, L. F., Anderson, D. R., Lorch, E.P., & Levin, S. R. (1980). Preschool children's
visual attention to attributes of television. Human Communication Research, 7,
52-67.
Anderson, D., Alwit, L., Lorch, E., & Levin, S. (1979). Watching children watch
television. In G. M. L. Hale (Ed.), Children's understanding of television (pp.
331-353). New York: Academic Press.
Anderson, D. B., & Bryant, J. (1983). Research of children's television viewing: The state
of the art. In J. A. Bryant, D. (Ed.), Children's understanding of television:
Research on comprehension and attention, (pp. 331-354). New York: Academic
Press.
Anderson, D. R., & Collins, P. A. (1988). The impact of children's education:
Television's influence on cognitive development. Washington, DC: Office of
Educational Research and Improvement.
Anderson, K. K.., & Revelle, W. (1994). Impulsivity and time of day: Is rate of change in
arousal a function of impulsivity? Journal of Personality and Social Psychology,
67(2), 334-344.
Anderson, L. A. (1986). Attention and visual perception: The availability of features.
Paper presented at The Human Factors Society-30th Annual Meeting, Houston,
Texas.
Archer, E. J. (1965). Concept identification as a function of obviousness of relevant and
irrelevant information. In D. P. R. C. Anderson & Ausubel (Ed.), Readings in the
Psychology of Cognition New York: Holt, Rhinehart & Winston.
124


107
separately. An analysis of variance shows an interesting result when verbatim and gist
memories are separated and compared. Gist memories were affected more by
presentation speed than verbatim memories. The analysis of variance tables also reveal
that gist memory was more affected by cognitive style than is verbatim memory. The
means and analysis of variance tables reveal that those viewing the video at the fastest
rate scored significantly did better on the gist portion of the memory test, whereas the
differences in mean scores on the verbatim portion of the test were not significant. While
the scores appeared to improve for verbatim questions as the presentation rate was
slowed, the differences were not statistically significant at the .05 level and, therefore,
have to be considered random variations.
Because the analysis of variance for presentation speed involved three different
rate comparisons, a Bonferroni follow-up test was used to pin-point the exact location of
the differences. The results confirm that a significant difference lies between the fast and
slow presentation speeds for gist memory, whereas the differences in scores between the
medium and fast and the medium and slow speeds for either gist of verbatim questions
are not significant.
The pertinent source and means table shows that combined total scores (i.e., totals
in which verbatim and gist scores were added together) improved for those in the sample
who were categorized as being reflective over the impulsive subjects, regardless of
presentation speed. The analysis of variance indicates that these mean variances were
significant at the .05 level. However, the interaction differences between speed and
cognitive style were not found to be significant at the same .05 level. A review of the
means table (Table 17) for impulsive versus reflective for verbatim and gist questions


Cognitive Style 53
Impulsive-Reflective Style 54
Factors from Film and Television That May Affect Learning 58
Preconceived Mental Demands 58
Exemplars and Other Formal Features 60
Symbol Systems 66
Evaluating the Educational Impact of New Media 69
New Media and Memory VO
Effect of Fast Cuts on Memory 73
Presentation Speed versus Content 73
Montage 75
Summary 77
3 METHODOLOGY 81
Introduction 81
Population 82
Sample 82
Instrumentation 83
Test descriptions 88
Initial Development 88
Pilot Test 89
Test Administration 90
Stimuli 91
Methodology 92
Research Design 93
Hypotheses 95
Summary 96
4 ANALYSIS OF THE DATA 98
Introduction 98
Results 98
Summary 106
5 CONCLUSIONS AND RECOMMENDATIONS 109
Introduction 109
Findings 110
Discussion Ill
Presentation Speed Ill
Gender 112
Verbatim versus Gist Memory 112
Cognitive Style 113
Interaction Between Speed and Style 114
Implications 114
v


96
Hypothesis 8. There are no significant differences in gist test scores for those
subjects who are determined to possess impulsive or reflective tendencies.
Hypothesis 9. There is no significant interaction between cognitive style and
presentation speed of the video.
Summary
Most of the universally recognized instructional design and development theories
have accurately established that audience/pupil analysis is an important part of the design
process. Cognitive style is one of many different attributes that might need to be
considered when technology that is used in instructional settings. The MFFT-20 used in
this study to categorize subjects as to their reflectivity-impulsivity has been determined
by independent reliability studies to be valid and more powerful than the original MFFT,
producing what is believed to be a more effective measurement tool for analyzing this
form of cognitive style.
The testing instrument developed for this study was tested for reliability and
construct validity through a series of feedback sessions with teachers, peer groups and
pilot testing. The reliability test results during the pilot fell within an acceptable range.
Because the whole test implementation was automated, the statistical outputs were
consistently reported ands made for an ease of analyses along several axes. In addition,
the automated cognitive style instrument and treatment tests allowed them to be
administered in same day, thereby significantly reducing the potential for mortality
among and contamination between the participants. This chapter describes in detail the
implementation and methodology of the test instruments.


APPENDIX A
SCREEN SHOT FROM MFFT-20 COGNITIVE STYLE TEST
122


4
need to update any earlier research that does exist. Lang has conducted more recent
studies that have looked into the effectiveness of fast-cuts (Lang, 1994; Lang, 1996; Lang
& Basil, 1998; Lang, Bolls, Potter, & Kawahara, 1999; Lang, Zhou, Schwartz, Bolis, &
Potter, 2000), but hers deal with television in a casual viewing environment. The present
study was designed to look at the effect of these rapidly presented visual passages in an
educational setting. It is hoped that a new look at rapid visual processing brought on by
this study might lay the groundwork for educational media producers to update their
thinking about rapidly presented video montage by providing a research basis that
appears to be lacking currently. Specifically, the questions that are being reviewed are:
1) Is it possible to remember content solely from rapidly paced visual montage
that is not supplemented with some form of verbal narrative?
2) Is there a difference between gist and verbatim memories for rapidly presented
videos?
3) How do individual characteristics such as gender and leaming/cognitive styles
affect a students ability to process fast-cuts/montage video presentations in a
classroom setting?
Need for the Study
As has been stated already, most of the previous studies into the impact of fast-
cuts/montage have concentrated on commercial television viewing (Bryant & Rockwell,
1991; Lang & Basil, 1998; Lang et al., 1999; Lang et al., 2000; Zillman, 1991). There has
been one attempt to look at the use of fast cuts/montage in an instructional setting (Keller,
1976). However, that investigator looked at viewers as a collective whole with little
regard for individual differences in personalities and capabilities such as cognitive style.
Further, this study is more than twenty-five years old and its subjects were college-level
students enrolled in communications classes, rather than school-aged students studied in


6
television are considered by some to be nothing more than prototypical, trite, and overly-
familiar formulas that reduce attention and concentration because they have been over-
learned (Anderson et al., 1979; Langer & Imber, 1979). This schema also leads one to
revert to a mindless routine in which the material is ignored or receives a low level of
attention (Wetzel et al., p. 169). This alleged over-familiarity with format has lead many
educators to believe that viewers will have difficulty responding appropriately to
educational televised presentations, unless some form of outside instructional
intervention is also inserted (Wetzel et al.). The third reason is that many of the studies
into casual viewing tended to lump all viewers into a single category (Lang and Basil,
1998; Lang et al., 1999; Lang et al., 2000; McLuhan, 1964; Neuman, 1976; Tyner, 1998;
Walma van der Molen & Van der Voort, 2000). Classical instructional models tend to
validate the value of classifying learners by their individual differences (Gentry, 1998;
Joyce, Weil, & Calhoun, 2000).
The fourth and possibly most important reason for the lack of correlation between
studies of casual viewing and those performed in an educational setting is that the
learning environment is thought to present a different set of circumstancesa unique view
as it were. This is based on the importance placed on the medium to be evaluated as to its
unique ability to bring about some type of alteration of intellectual behavior or thinking
process. Although many of the symbolic (i.e., intellectual) combinations found in non
verbal endeavors such as music, painting, and dance can be displayed directly on
television, there has been some question as to the extent to which transformations in the
thinking process are actually created by viewing the medium, and whether any changes
that might occur are of any significance (Ide, 1974). Television has not been given credit


40
method is based on a metaphor of intuitionalism, in which people prefer to think, reason,
and remember using inexact, gist-like traces rather than more precise (but also more
forgettable) verbatim traces. The concept traces memory preferences through age and
personality differences, with younger children preferring verbatim, exact references,
which become fuzzier as they grow older. Miller and Bjorkland describe fuzzy-trace
theory as a hypothesis about representation, which posits that children of different ages
are disposed to use different types of representations that are available to them. What
differs between this theory and other more mainstream conceptualizations is the
propensity to use different types of representations to solve problems, making it more
aligned with Piaget-like concepts of constructivism. According to Miller and Bjorkland,
earlier misconceptions about fuzzy-trace theory caused confusion as to how it differed
from other already established conventional concepts like schema theory. Gradually,
however, fuzzy-trace theory began to gain acceptance with mainstream theorists,
especially when it was applied to issues of age differences in suggestibility and false
memory creation (Miller & Bjorkland, p. 188). It was later adopted outside of the
immediate Brainerd and Reynas [sic] sphere of influence and began to be used to
explain basic and applied phenomena on a wide range of tasks (Miller and Bjorkland,
p. 188).
Independently, Brainerd and Gordon (1994) and Reyna and Kieman (1994) began
to evolve their jointly-developed theories on to other memory domains, concentrating on
the specific differences between gist and verbatim memory traces. The concept has been
subsequently been adopted by others (Cowan, 1998; Tse et al., 1999). The verbatim-gist
classification has more recently been used in a variety of ways, including to


Ill
Hypothesis 6. There are no significant differences in overall test scores for those
subjects from the sample who are determined to possess impulsive or reflective
tendencies.
The null hypothesis was rejected.
Hypothesis 7. There are no significant differences in verbatim test scores for those
subjects who are determined to possess impulsive or reflective tendencies.
This hypothesis was not rejected.
Hypothesis 8. There are no significant differences in gist test scores for those
subjects who are determined to possess impulsive or reflective tendencies.
This hypothesis was rejected.
Hypothesis 9. There is no significant interaction between cognitive style and
presentation speed of the video.
The null hypothesis was not rejected.
Discussion
Presentation Speed
There were no significant differences in overall combined test scores when the
presentation speed of the video was varied. While there were differences in the total
number of correct responses, those differences were not significant, meaning that any
difference between the scores in this portion of the study must be considered random
variations. The lack of significance between the groups due to changes in presentation
speed appear to run counter to many commonly understood ideas about memory
processing and speed of message delivery -that immediate memory could be conversely
affected by the presentation speed of the video. Part of the reason that subjects in the


60
especially because researchers have shown that currently most of younger viewers day-
today encounters with both computers and videos (in particular music videos) are for
amusement purposes (Seward-Barry, 1997).
These ideas about preconceived demands relate directly to the procedures used in
the current study. Care was taken while administering the test to 1) demonstrate the
purpose of the viewing activity using a sample portion of the video and 2) instruct the
subjects to treat the activity just as if they were to be graded on it, even though no
intention to do so was introduced to them beforehand.
Exemplars and Other Formal Features
Although many studies seem to establish televisions reputation as a medium that
promotes mindless non-engagement of its viewers, research has been far from definitive.
As of a recent count (Abelman, 1995), over 4,000 scientific research articles and
government reports have been published examining media effects with special emphasis
on the impact of television on children, and still the literature continues to be imprecise as
to how children learn from media interactions. On the other hand, some studies appear to
be more conclusive. Anderson and Bryant (1983) produced results of studies of specific
formal features of television and showed that the research can be fairly robust in that
subjects were able to define and understand the internal mental processes involved with
interpreting mediated messages. For example, they found that valence in children can be
influenced by specific interpretations of what they see and hear. Particular (i.e., a
womans and/or childs) and/or peculiar voices, sound effects, auditory changes, and
visual movements all had a positive effect on recall and comprehension, whereas men's


13
contrasting theories about memory and cognition (Lang et al., 1999; Lang et al., 2000).
These investigations concluded that individuals have limited capacity for processing
information and indicate that a rapid influx of visual images maintains an individuals
orientation reflex and stimulates cortical arousal. Unlike other forms of arousal, cortical
arousal has been determined to maintain a persons ability to pay attention (Pearl, 1982;
Reeves, Thorson & Schleuder, 1986; Zillman, 1991). The earlier studies also alluded to
conflicting findings about the impact of the content on attenuation and arousal
(Anderson, 1986; Archer, 1965; Basil, 1994), maintaining that arousing content
combined with too fast a pace/message delivery would normally result in loss of retention
and recall. Unlike these previous studies that included occasional highly arousing
content, the current one used an educational topic that might not be considered by many
students as arousing (i.e., American history) in an authentic classroom setting.
The current study also attempted to improve upon some of the various
shortcomings found in the one other earlier study (Keller, 1976) that used very similar
but out-dated methodologies. In his doctoral dissertation, Keller used a self-created
kinestasis film (i.e., rapidly cut video montage) in an attempt to define what effects on
recall are realized by speeding up transmission or information to less than one second per
frame. Standards previous to Kellers study deemed a presentation to be rapid if cuts
occurred every five to ten seconds. Keller also endeavored to derive empirical data to
determine how well the kinestatic process worked on recall and how efficient visually
compressed presentations might be for instruction, trying to find out whether the
kinestasis technique was more than a simple artistic endeavor. However, by his own
admission, Kellers research came up short because some of his methods may have been


42
current study to address this challenge and extend their analyses into the pictorial
memory domain.
The Effect of Arousal on Memory
Another, related topic on memory is the concept of arousal and an attempt to
clarify some of the differences of opinion with regards to its perceived positive and/or
negative effects on memory. Many of the research findings with respect to arousal and
memory appear to be based on the Yerkes-Dodson law (Yerkes & Dodson, 1908) which
predicted an inverted U-shaped relationship between arousal and performance. A certain
amount of arousal can be a motivator toward change (with change equating to learning),
whereas too much or too little can work against the learner. The Yerkes-Dodson law
appears to be the basis for much of the early controversies surrounding arousal in
mediated messages as to whether they impose a negative or positive effect their viewers
(Lang & Basil, 1998). Neil Postman (1986) asserted that the presentation speed of media
messages is a force that increases arousal to the point where most cognitive activity is
negated. He relegated television viewing to pure mindless entertainment with little or no
educative value. On the other hand, Zillman (1991) postulated that arousal can be a
unifying force that intensifies motivated behavior. Further, arousal plays a significant role
in many motivational models, like the ARCS model developed by John Keller (1983).
The A in this acronym stands for gaining the learners attention, which is handled
through arousing or curiosity-seeking cognitive engagement.
The interactions between arousal, motivation, and attention and their affect on
learning have gained considerable focus in educational research in recent years. Posner


114
an average of 7.43 (approximately 49%). These differences were not found to be
statistically significant and, therefore, were considered random fluctuations. In the gist
portion, however, impulsive subjects scored an average of 5.20 out of a possible 10
(52%). The reflective subjects scored an average of 5.82, or a little more that 58% correct
answers, which was shown by the analysis of variance to be statistically significant.
These results show that while cognitive style did affect memories for the video, the gist
memory was affected most.
Interaction Between Speed and Style
No significant differences were found to be caused by an interaction between
presentation speed and cognitive style. This means that, at least for the subjects in this
study, cognitive style mediated immediate memory but it was not further affected by a
change in presentation speed. That portion of the subject group that was found to be
impulsive or reflective was affected by the speed in the same way as the remainder of the
sample group.
Implications
The results of this study indicate that students tend to remember gist information
from rapidly presented videos better than those presented at slower speeds. These results
appear to contradict earlier research that seemed to indicate that the viewers would be
able to remember more information from pictures if they are presented at slower speeds.
One of the reasons might have been because the earlier studies measured memory on a
combined basis, concentrating solely on measuring immediate verbatim memory. The
parsing of gist and verbatim scores in the current study has uncovered a potential new
approach to investigate differences in pictorial cognition. The fact that the significant


54
process of learning. Out of this broad context of research emerged the concept of
cognitive styles (Witkin & Goodenough).
Guilford defined cognitive style as that which conceptualizes intelligence as having a
process dimension (Green, 1985, p. 2). According to this view, learning is not merely an
automatic reaction to a stimulus but a set of operational steps that varies, depending on
individual proclivities. Brumby (1982) asserted the following assumptions regarding
cognitive style:
Ones cognitive style is singular (i.e., an individual has only one) and can
be measured on a bi-polar scale.
While an individual possesses one style, others may be present in varying
degrees.
Individuals can select an appropriate style appropriate to the task at hand.
Earl Messick (1970), a charter member of the 1949 New Look movement,
catalogued nine dimensions of cognitive style, covering research that he and his
colleagues performed, and that of others who came along in the years that followed. His
list included scanning, breadth of categorization, conceptualizing style, levelers versus
sharpeners, distractibility, tolerance for unrealistic experiences, cognitively complex
versus simple, field-dependent versus independent, and impulsive versus reflective. Of
the nine mentioned, meta-research (Green, 1985) has shown the latter two to be the most
commonly accepted as credible sources for investigating how individuals perceive and
process visual patterns. Of these two, impulsive-reflective was determined to be the most
relevant to the current study.
Impulsive Reflective Style
The impulsive-reflective scale was borne out of Jerome Kagans research work


5
their formative educational environments. The current study attempts to resolve some of
these contradictions to see the effect fast video presentations have on immediate
memories for their content.
Educational media researchers have had to contend with a conflicting view that
holds that the relatively fast presentation speed of television programs creates an
environment that may be detrimental to attention and recall (McCollum fe Bryant, 1999;
Neuman, 1976). To the contrary, Intraubs (1999) studies into conceptual masking have
shown that humans are able to recognize and recall pictorial presentations when a
minimal amount of lag time separates individual images. In addition, some of the more
recent successes with childrens programs have shown educational successes that are due
to their rapid and rhythmical presentation speed (Anderson, Alwit, Lorch, Anderson and Bryant, 1983; Anderson <& Collins, 1988; Pearl, 1982). Still others have
proposed that presentation speed in instructional media may actually add interest to
otherwise uninspiring content (Canelos, 1986; Edgar, 1997; Hawkins, Pingree, Bruce, <&
Tapper, 1997; Hill fe Lang, 1993).
Previous examinations into the effects of casual television viewing have not
translated too well into an educational setting (Salomon, 1994). As Kozma (1986) stated,
"viewership should not be confused with learning" (p.14). However, once one delves
deeper into these studies, four possible reasons for this lack of transfer become apparent.
First, the reputation that commercial television has held for being nothing more than an
entertainment device has hurt its standing in educational circles and has caused
detrimental pre-conceived notions about how viewers are to be properly introduced to
televised content (Wetzel et al., 1994). Second, many of the schemas used in commercial


95
whom this portion of the test would be administered. As the main and interaction
effects for presentation speeds were also to be analyzed, a second 2x3 factorial design
ANOVA was designed, as shown in Table 3.
Hypotheses
The hypotheses for this study are stated in the null form. The following hypotheses
were determined for the research design using analyses of variance:
Hypothesis 1. There are no significant differences in overall test scores (both
verbatim and gist scores are pooled together), for the overall sample set based on
presentation speed of the video.
Hypothesis 2. There are no significant differences in overall test scores between
male and female subjects in the overall sample set.
Hypothesis 3. There is no significant interaction between presentation speed and
gender for overall test scores.
Hypothesis 4. There are no significant differences in verbatim test scores, for the
overall sample set, based on presentation speed of the video.
Hypothesis 5. There are no significant differences in gist test scores for the overall
sample set, based on presentation speed of the video.
Hypothesis 6. There are no significant differences in overall test scores for those
subjects from the overall sample set who are determined to possess impulsive or
reflective tendencies.
Hypothesis 7. There are no significant differences in verbatim test scores for those
subjects who are determined to possess impulsive or reflective tendencies.


48
there have been a few well-elaborated schema theories that demonstrate specific
influences of the effect of showing pictures episodically in the absence of personal or
semantic interpretations, a number of independent investigations of perceptions formed
on the basis of televised news reports demonstrate convincingly that there exists a high
degree of correspondence between subject interpretation and recall (Zillman & Brosius,
2000, p. 38). Value judgments also play a role information processing. Several studies
into news viewing (Bums, 1992; Hayes-Roth & Hayes-Roth, 1977) also have shown that
pictorial representations have a positive effect on attention and processing. Likewise, the
value of increasing the number of occurrences of pictorial displays (as opposed to simply
reiterating a fact verbally) has been demonstrated in a number of studies. Zillman and
Brosius showed that an increased number of occurrences of a pictorial message can also
influence perception in the news. For example, in their studies, viewers concept of how
much significance or importance they should appropriate to a story seemed to rest in
direct proportion to the number of times it was shown on television, to the point that
where subjects applied more significance to the story than it actually warranted. Zillman
and Brosius pointed the fact that repeated visual representations tended to slant the news,
suggesting that there was strong evidence that the sheer number of concrete, visible
sources relating their experiences and concerns does exert an influence on issue
perception (p. 110). They further suggested that photographs used in print media and on
televised newscasts, regardless of how innocuous the content, are remembered longer and
the image that was more likely to be retained was of the visualization, rather than the
textual message that accompanied it.


92
help box on every screen and heard an audio reading of the instructions whenever they
changed.
Methodology
Testing began immediately after the video was viewed. Subjects were asked to
click a start button when they were ready to begin. The process of clicking the button
initiated the internal system clock that tracked latency to first response. By providing
instructions in this way, the program avoided counting time reading and understanding
the questions as part of the latency calculations during the administration of the
MFFT-20.
The study utilized a combination of pictorial and textual methods to assess
memory. For the verbatim portion of the test, subjects were shown several groups of
pictures and asked to pick out via a mouse click the one picture among three distracters
that did or did not appear in the video. For the gist memory questions subjects were asked
contextual questions in the program. Subjects were also asked to place sets of pictures
into contextual-chronological order based on their presentation in the video.
These methods of testing visual clues appear to be in accordance with long-
established procedures and findings the literature. Archer (1965) proposed that, if one is
testing for visual processing, evaluative vehicles should also be designed that are closely
aligned with the visual process because something may be lost in the translation to
verbal. Corcoran (1981) noted in his studies into visual perception that testing should
utilize the same mode as the original presentation (i.e., textual to textual, pictorial to
pictorial) because textual and pictorial forms of information are processed at different
rates and in different ways. He also found that reading has the tendency to interfere with


72
to television during shows like Sesame Street or the Electric Company served to also
point out that attention might be better defined as "visual selection" (p. 253). Attention
was measured by a hidden observer who pushed a button when a child's eyes were
directed towards the television screen. This type of eyes on screen method dominated
research for years. However, more recent literature has clearly defined attention as "a
psychological cognitive process that varies within individuals over time" (Reeves et al, p.
254).
Reeves et al. (1986) later began to find that perhaps, attention, as it relates to new
media, is a many-fold process and that it is not necessary an all-or-nothing effect.
Perhaps, different types of attention required different types of responses. Vigilance tasks
are an automatic process that requires people to wait for an event and to respond quickly.
This process decreases over time and only allows a subject to be able to attend them one
at a time. Reeves et al. correlated this to primitive cues that are bom out of humans
animal instincts. Attentionalpreparation tasks involve giving people cues or primes
before a stimulus occurs. Television can be thought of as a continuous priming or cueing
process. This strategy could be used to study the sequencing of various visual devices,
such as pans, zooms, audio silence, screen luminance, or presentation speed and their
ability to prime attention for subsequent program content. Other questions could be asked
about the effects of un-cued or abrupt scene changes on attention and the circumstances
under which cueing increases processing efficiency. Just as Olson viewed symbol
systems, attention appears to require internal devices that are also used differently,
depending on the task that is to be performed. Measuring attention can be added to


78
in rapidly presented montage in commercial programs (Lang, 1994; Lang & Basil, 1998;
Lang et al., 1999; Lang et al., 2000). It is perhaps Langs work in commercial television
that provides the best backdrop for studies that investigate a corresponding effect of these
new media techniques might have in the educational domain. If one accepts the
contentions of Reeves et al. (1986) and Stephens (1996), it should be easy to understand
how todays youth, brought up on MTV, with its fast cuts, rapid movements, and
iconoclastic acoustics, could get used to, prefer, and possibly learn from similar
techniques in an educational setting.
There are those who contend that the rapid format of television segments takes
away from attention and understanding (Anderson et al., 1979; Anderson & Bryant,
1983; Anderson & Collins, 1988). Reeves et al. (1986) demonstrated information in
several research studies that contradict that belief. Stephens (1996) and Meyrowitz (1985)
appear to agree. They concluded that, once its capabilities are fully implemented using
techniques like fast cuts and video montage, television will provide an opportunity to
transcend the time and place continuum by cutting rapidly between images taken from
different contexts. In other words, viewers can find additional meaning in the relationship
between numerous different scenes. Taken in this way, the video image provides
opportunities for creative seeing, making the case that interpreting images requires more
imagination of their viewers, not less.
This does not mean that educators or their students have no obligation with
regards to content. Lang and Basil (1998) and Lang et al. (1999) and Lang et al. (2000)
have suggested that producers who want their messages to be remembered should create
arousing messages that are presented slowly or at medium speed or calm messages that


119
media-centric youth to more quickly perceive and assimilate rapidly
presented visual images was due to casual television viewing habits and
video game usage. Studies could be developed into determining whether
increasing perceptual skills can be a trainable activity, especially using
older subjects who might not have had the same viewing and gaming
opportunities previously.
9. The results of this study align to a certain degree with Stephens (1996)
assertions that video montage can convey composite thoughts in as much as
a paragraph does in text-based communications. Based on the results of the
current study, a further investigation could be developed to look further
into the effects of continued used of video montage to develop gist
memory.
10. Studies could be developed to determine the relationship between numbers
of incorrect choices students make prior making correct selections,
latencies, and learning progress. New interactive technologies provide
novel opportunities for educators to easily build test instruments that track
both these test-taking characteristics. While the results were not included,
the computer software used in the current study to track verbatim and gist
memory was also programmed to track latency as well as correct responses
in the exact manner that was used in the automated MFFT-20 cognitive
style test. Tracking latencies as well as number of responses made prior to
correct selection could be incorporated into more formal studies to more
thoroughly track the progress of student learning.


36
term categorization shares some meaning with the term perception. However,
categorizing also implies a deeper process than simply detecting that some unit of
information has been perceived. It also means that other related characteristics have been
captured (i.e., location, color, or shape, etc.) to apply meaningfulness to them that aides
the process of recalling that information later. Under this definition, immediate memory
is looked upon as an important but insufficient prerequisite for learning. The current
study utilized a cued recall performance test in order to assess the ability of the treatment
to create (and the subjects to immediately remember) an initial memory trace.
Textual versus Pictorial Memory
Although much of the research into memory cited in this review deals with textual
information (Crowder, 1976; Estes, 1994), a great deal of it has direct relevance to the
current study. In his research into memory capacity, Sperling (1963) was able to
determine that his subjects were able to attend to visually perceived textual characters at
a rate of about 100 per second. These same subjects were only able to name (categorize)
these characters at a rate of six per second. This discovery led Sperling to assume that
initial information-gathering and naming (categorizing) may not be the same process, and
that the task of recalling utilizes different resources and techniques for pictorial than for
textual information. Durso and Johnson (1979) found that pictures benefit more when the
tactic used for elaborative processing is to simply name the picture whereas words aid
more when subjects perform some task to categorize them. Gaining an understanding of
these differences between the textual and pictorial memory processes has served two
purposes in the current study. First, it helped to emphasize the role pictorial perception
can play in the learning process. Second, it reinforced the need to assess pictorial memory


37
with like stimuli, that is, to use pictures as targets in the test questions as much as
possible because textual questions and answer choices introduces an additional
translation process.
Recognition and Recall
Over the years, there has been a significant amount of confusion regarding the use
of the terms recognition and recall as it relates to measuring how much learning has taken
place (Sheppard, 1967). While very similar in nature, recognition and recall are not
synonymous. Their difference in meaning suggests that we have different ways of
retrieval once information has been stored. Studies have shown that recognition memory
resides on the right side of the brain whereas recall (especially verbal) takes place on the
left. Recognition implies a simpler intellectual task that does not necessarily require a
significant amount of encoding and carries a longer-term residual. Sheppard describes
studies in which even a quick glance was enough to assure some long-term recognition
value. He showed objects from a Sears Roebuck catalog to Guatemalan peasants who had
no personal knowledge of the items. They were able to recognize the pieces even several
months afterwards. Singer (1980) drew several additional conclusions about the
differences between recognition and recall. He suggested that the human brain is capable
of storing a tremendous amount of visual material much more than information received
in other forms.
The ability of individuals to store so much visual information more rapidly is
perhaps due to the holistic or gestalt qualities of the right brain where visual imagery is
processed. Thus, material seen just a few times (i.e., presented on television) without any
significant effort at learning it, can be recognized if it is re-presented later. The greatest


CHAPTER 5
CONCLUSIONS AND RECOMMENDATIONS
Introduction
This study had several purposes. First it attempted to determine if a change in
presentation speed of a video montage might alter subjects combined memory scores
(i.e., a single total score of the entire memory test was evaluated as a whole in which
verbatim questions and gist questions were pooled together). Second, the study attempted
to see if a change in presentation speed would create any significant differences in the
combined memory scores between males and females. Next, the verbatim and gist scores
were considered separately to determine if presentation speed has a significant affect on
either group. The entire sample set was used in the analyses of variance to investigate
these first two conditions. Lastly, the study attempted to find out if there are significant
differences in combined memory scores for the video presented at three different speeds,
based on subjects cognitive style. Categorizing subjects as to impulsive or reflective
tendencies, by definition, resulted in using only that portion of the overall sample set who
were identified as impulsive or reflective in this portion of the study. To further
investigate any differences, verbatim and gist memories were also considered separately
for this portion of the study.
Analyses of variance were used to investigate the effects of presentation speed,
gender, and cognitive style. Findings are reported below.
109


26
(Seward-Barry, 1997). Characteristics of stimuli cause humans to structure and/or
interpret a visual field in certain ways or groupings. Wertheimer noted the following
primary factors in determining these groups:
Proximity Humans tend to group visual elements together according their
spatial nearness.
Similarity Humans tend to group together items similar in some aspects (i.e.,
their design, size, shape, textures, etc.).
Simplicity Visual items are grouped together by humans into simpler figures,
according to their symmetry, regularity, and/or smoothness.
Closure Humans tend to complete on their own an incomplete picture or
other visual entity.
Wertheimer was particularly concerned with problem solving. While problem
solving per se is outside the scope of the current study, these ideas form an important
basis for some of the assumptions about the affect that rapidly presented video montage
may have on memory and learning. Wertheimer felt that the essence of problem solving
behavior is ones ability to be able to see the overall structure of a problem through the
use of one or more of the above classifications which all allude to the ability to fill in the
gaps, incongruities, or disturbances that prevent one from seeing these overall structures
and use an innate need to do so as an important stimulus for learning. According to Ellis
(1938), other gestalt thinkers like Kofka and Kohler agreed with Wertheimer but
extended these ideas to looking at visual patterns. The laws of visual organization found
in these schools of gestalt thinking (i.e., proximity, similarity, simplicity, and, closure)
form a strong theoretical basis for visual media like film, video, and television to be able
to create its own unique sense of reality (Seward-Barry). The laws of visual organization


134
Nazarro, J. R., & Nazarro, J. N. (1970). Auditory versus visual learning of temporal
patterns. Journal of Experimental Psychology, 84 (3), 472-478.
Nelson, D. D. (1990). Personal tempo as a consideration in the rhythmic training of first
grade students. (Doctoral Dissertation, University of Florida, 1990). Dissertation
Abstracts International, 51 (10), 3356A.
Neuman, O., & Prinz, W. (1990). Relationships between perception and action. New
York: Springer-Verlag.
Neuman, W. R. (1976). Patterns of recall among television news viewers. Public Opinion
Quarterly, 40, 115-123.
Nugent, G. C. (1982). Pictures, audio, and print: Symbolic representation and effect on
learning. Education Communications and Technology Journal, 30, 163-174.
O'Brien, C. R. (1968). Variables of perception in selected individuals viewing
progressively accurate visual stimuli. (Doctoral Dissertation, University of
Florida, 1990). Dissertation Abstracts International, 30 (01), 0173A.
Okun, M. A., Callistus, W. M., & Knoblock, L. B. (1979). Adult age differences in
cognitive tempo. Paper presented at the Annual Meeting of the American
Educational Association, San Francisco.
Olson, D. R. (1974). Introduction. In D. R. Olson (Ed.), Media and symbols: The forms of
expression, communication, and education. The Seventy-Third Yearbook of the
National Society for the Study of Education. Chicago: University of Chicago
Press.
Ong, W. J. (1982). Orality and literacy: The technology of the word. London: Routledge.
Paivio, A. (1986). Mental representation: A dual coding approach. Oxford, England:
Oxford University Press.
Patel, A. D., Peretz, I, Tramo, M., & Labreque, R. (1998, January). Processing prosodic
and musical patterns: A neuro-psychological investigation. Brain and language,
(5/(1), 123-144.
Pearl, D. (1982). Television and behavior: Ten Years of scientific progress and
implications for the eighties. Volume 1: Summary Report. Rockville, MD: U.S.
Department of Health and Human Services.
Peck, K. L. (1987). Adaptive external pacing as an alternative to self-pacing in
computerized instruction. (Doctoral Dissertation, University of Colorado, 1987).
Dissertation Abstracts International, 48 (05), 1105A.


17
It also should be noted that this study has limitations with regards to its ability to
predict actual learning. The study was designed to only define certain rules about
presentation speed of an initial presentation that may become one component of an
overall instructional strategy. By itself, the study does not attempt to make any
predictions or value judgments about those strategies. Rather, it simply attempts to
support the validity of fast-paced video segments that might become building blocks in
sound pedagogical patterns, as delineated in instructional models that might be
subsequently proposed by future educational research.
Definition of Terms
Following are specific definitions of terms used in this study:
Cognitive style is the stable ways in which persons differ in perception and
encoding of information (Wittrock, 1979). Specifically, it is Kagans scale (1966) that
was later modified by Cairns and Cammock (1984) that is used as a basis for the current
study.
Conceptual masking is concept that describes the phenomenon that occurs when a
new visual image is introduced into temporary memory and thereby occludes the
previous one (Intraub, 1999).
Gist Memory has been successfully used as a treatment effect in several studies
looking into prose memory (Biederman, Rabinowitz, Glass, & Stacy, 1974; Brainerd &
Gordon, 1994; Cowan, 1998; Reyna & Kieman, 1994; Voss, Tyler, & Bisanz, 1982). Gist
memory is one of two types of determinations (the other being verbatim memory) as to
the amount and quality of immediate memory. In previous research gist determination
required subjects to try to identify the main idea in a passage or correlate among several


34
visual stimuli. They suggest that short-term memory is based on auditory encoding even
when the stimuli are presented visually. They also found that learning (i.e., memory) is
reduced if the presentation relies mostly on visual inputs alone. Further, many researchers
(Cooper, 2000; Davis et al., 1999; Flannagan, 1998; Flowers, 1995; Moreno & Mayer,
2000; Sterrit, Camp, & Lipman, 1996) have shown how audio tracks can aide visual
encoding, especially where the fidelity of the visual display suffers.
Perhaps the most conclusive studies about the relatively stronger effects of visual
over aural stimulation have been offered by Gavriel Salomon (1979; 1994). He sites
many studies where visualization was more effective in recall, and concluded that visual
actions are often remembered over auditory have been "primarily due to the salience over
action" (Salomon and Cohen, 1977, p. 29), especially moving images. Further, he cites
some investigations in which visual tracks were successfully added to supplement the
auditory. He concluded that there probably is no visual dominance in children but since
actions are generally more memorable, visual information is normally more likely to be
recalled. However, he also cautioned that adding pictures to an already well-formed
auditory presentation may provide some distraction (Salomon and Cohen, p. 29).
Researchers have shown that words and pictures are processed on opposite sides
of the brain and follow different psychological laws (Ederlyi, 1985). This fact had
originally deprecated most traditional ideas about subliminal communication being an
effective means of communication and learning. Most of the research done in subliminal
perception had been done with verbal stimuli, based on the conventional assumptions
about cognition having to precede response. Erdelyis and LeDouxs (1991) more recent
studies using visual stimuli alone have produced more pronounced results. Although their


14
lacking in rigor. For example, he utilized random pictures as opposed to those with
similar contextual content. As schemata have been long shown to be of importance to
learning (Bartlett, 1932) studies looking into recognition and recall need to use a single
context. Additionally, most of the problems Keller ran into appeared to be related to
production and processing techniques that reflected the state of the art of video
production at the time (Keller). His difficulties ranged from dissimilar content, picture
size that was too small, and brightness that caused a deviation in results based on the seat
location of his subjects. Due to technological shortcomings, he utilized what are known
in the industry as wipe-ons/downs as his motion technique. Wipes are production
techniques utilized in television broadcasting in which one picture is replaced by another
that is gradually wiped across the screen. An edit or cut, on the other hand, is where one
picture immediately and completely replaces another. In this technique, the succeeding
picture frame partially masked, or interrupted the previous one and vice versa, causing a
potential confound. There has been some reference in the literature (Goodglass, 1971)
that indicates that visual masking introduces a new, unnecessary variable into studies on
visual processing. Keller also had problems in his test phase because he could not cause
the pictures to replicate the movements that occurred in those same pictures when he
presented them originally. He admitted that this approach was flawed in that the whole
picture could not be viewed until the end of the procedure, causing a degradation of
recognition and recall. The current study attempts to clarify some of those issues. Keller
referred to the need to find another motion procedure (p. 90). The current study used a
commercially-made video montage that refers entirely to the same subject matter, for


116
The results of this study also indicate that using the impulsive-reflective cognitive
style instrumentation may be still a valid measuring tool. While both verbatim and gist
memories were both negatively affected by cognitive style, it was only gist memory for
the rapidly presented videos that was affected significantly. When considering the entire
sample without regard to style, the analysis of variance did not yield any significant
differences. The cognitive style portion of this study uncovered some discrepancies (i.e.
learning difficulties) that might have otherwise gone unnoticed.
Whether the changes found here are the result of differences in casual television
viewing habits or computer usage (or both in combination) was outside the purview of
the current study. However, many of the results indicate that something is different about
the way todays youthful learners perceive visual inputs, creating several interesting
scenarios for future studies.
It is noteworthy that the total correct number of responses for all groups was quite
low (around 50%). The relatively low numbers of correct to total possible responses
serves to reinforce that the purpose for integrating video presentations into teaching and
learning situations has not changed. While a change in presentation style may provide an
essential pre-condition for increasing knowledge, it still needs to be coupled with sound
instructional strategies for any learning to take place.
Recommendations
The results of this study have raised some interesting questions about the nature
of visual perception and it discovered a potential for investigating several new paradigms
for instruction. The following recommendations are made for future studies in this area:
1. When one compares the median latencies and errors and the percentages of


136
Salkind, N. J., & Wright, J. C. (1977). The development of reflection-impulsivity and
cognitive efficiency. Human Development, 20, 377-387.
Salomon, G. (1979). Interaction of media, cognition, and learning. San Francisco: Josey-
Boss.
Salomon, G. (1984). Television is "easy" and print is "tough": The differential investment
of mental effort in learning as a function of perceptions and attributions. Journal
of Educational Psychology, 76, 647-658.
Salomon, G. (1994). Interaction of media, cognition, and learning. Hillsdale, NJ:
Lawrence Erlbaum Associates.
Salomon, G., & Cohen, A. (1977). Television formats: Mastery of mental skills and the
acquisition of knowledge. Journal of Educational Psychology, 69, 612-619.
Schale, F. (1971, December). Measuring degree and rate of visual awareness in rapid
reading on television. Paper presented at the National Reading Conference,
Tampa, FL.
Schank, R. C., & Abelson, R. P. (1977). Scripts, plans, goals, and understanding.
Hillsdale, NJ: Lawrence Erlbaum Associates.
Seidman, S. A. (1981, Spring). On the contributions of music to media productions.
Educational Communications and Technology Journal, 29.
Seward-Barry, A. M. (1997). Visual intelligence. Albany: State University of New York
Press.
Shaffer, R. J., Greenspan, S. I., Tuchman, R. F., Cassily, J. F., Jacokes, L. E. & Stemmer,
P. J. (2000, August 19). Interactive metronome: Effects on motor control,
concentration, control of aggression, and learning in children with attention-
deficit/hyperactivity disorder. Paper presented at the Progress in Motor Control II
Congress, Penn State University, State College.
Shalev, L., & Algom, D. (2000). Stroop and Gamer effects in and out of Posner's beam:
Reconciling two concepts in selective attention. Journal of Experimental
Psychology, 26(1), 997-1017.
Sheingold, K. (1973, August). Developmental differences in intake and storage of visual
information. Journal of Experimental Child Psychology, 16, 1-11.
Sheppard, R. N. (1967). Recognition memory for words, sentences and pictures. Journal
of Verbal Learning & Verbal Behavior, 6, 156-163.


106
investigate these differences. Table 17 shows that the significant differences in test scores
found in Hypothesis 6 were derived from the gist portion of the memory test, which is
consistent with previous findings regarding the significance of gist versus verbatim
memory from the overall combined test scores.
Table 17
Means and Standard Deviation for Gist and Verbatim Scores for Style
STYLE
Mean
Std. Deviation
N
GIST
Impulsive
5.20
1.77
64
Reflective
5.82
1.54
65
Total
5.51
1.68
129
VERBATIM
Impulsive
6.77
2.14
64
Reflective
7.43
2.27
65
Total
7.10
2.23
129
Hypothesis 9. There is no significant interaction between cognitive style and
presentation speed of the video.
The interaction effect for between subjects of cognitive style (STYLE) and
presentation speed (SPEED) resulted in an F (2, 123) of .398 (see Table 14). This F ratio
is not significant at (p<.05). Therefore, this null hypothesis was not rejected.
Summary
The results of the study indicate that combined memory scores (verbatim and gist
added together) tended to improve as the speed of the presentation slowed down.
However, these differences were not statistically significant and are considered random
fluctuations. The analysis of variance for gender also shows small differences in scores
between males and females. These differences, too, were not significant at the .05 level.
In order to more fully review the differences that presentation speed imposes on
memory, the scores for verbatim and gist portion of the test were recorded and analyzed


120
Finally, it should be noted that the technology employed allowed the investigator
to conduct this study with a small number of large groups (30+ at the same time). This
was a bane as well as a blessing. There were times when the interactions between
subjects became somewhat difficult to control. Luckily, there were three proctors in the
room to ensure that subjects remained on task and were correctly following directions.
Future studies might fare better if the number of sessions were increased to allow for
smaller group sizes. Large group interactions did cause a certain number of distractions
and may have had some bearing on the results. While these types of large groups might
indeed be more authentic (i.e., they represent more closely an actual classroom setting),
crowd control is an issue and a potential confound to the results of the research.
Summary
In some regard, readers of this study would be correct if they note that it appears
to have raised more questions than it has answered. This was a part of the original
motivation to do the study. The catalyst to do this study was an inclination on the part of
this investigator to apply a research base to the myriad of recent unsubstantiated reports
found in the literature and personal anecdotal observations about the changing nature of
cognition and communications in todays technology rich society.
As a minimum, it appears that ideas concerning the nature of cognitive style are in
need of updating. The review of the literature revealed that little has been done in this
area for almost twenty years and the median latency and errors have been shown to
decrease significantly since original studies took place.
It should be noted that it was never the intent of this study to discover some new
paradigm for learning. Rather, it simply intended to re-open some discussions and


THE EFFECTS OF COGNITIVE STYLE AND GENDER ON
VERBATIM AND GIST MEMORY FOR
RAPIDLY-PRESENTED MONTAGE VIDEO
i *
l i. V i *
By fj.
ROBERT F. KENNY
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSPOPHY
UNIVERSITY OF FLORIDA
2002

ACKNOWLEDGMENTS
Special appreciation is extended to Jeffry Hurt, David Miller, and David Ostroff,
members of my doctoral committee, for their support in this undertaking, and all their
advice and counsel as I moved through the process of completing my degree. I took
classes from all of them. With each course, 1 became more inspired and awed by their
insight, knowledge, and enthusiasm for their profession. They have inspired me to be just
like them.
I would especially like to thank Lee Mullally, my committee chair, for his
direction and support, and especially his determination not to let me take any shortcuts, or
not to give my best efforts in the many classes and seminars I took from him, the reviews
and edits for the textbook I wrote, and especially this study. He has my unending
appreciation for his extraordinary patience and ability to keep me focused.
I am grateful to Chuck Braverman, whose kinestasis film was used in this study. I
saw it almost 30 years ago and knew then it would somehow become a part of my life. I
am particularly grateful to Randolph Wright from Pyramid Media, copyright holder of
American Time Capsule, who graciously granted me permission to use it in this study. I
am also thankful to Mitchell Stephens, whose book, The Rise of the Image, the Fall of the
Word became the catalyst for this study. Other producers and researchers in new media
like Jeff Scher and Annie Lang have also inspired me with their insights into visual
perception, fast-cuts, and montage video. I am happy and lucky to have met them.
11

I am grateful to all the many teachers and students who helped with the review of
the test instruments, pilot test, and test administration, in particular Rachael Maland,
James Keith, and Marshall Broitman. I am particularly grateful to Tom Wilson and Alan
Smolowe, who were there with me every step of the way, with a constant watchfulness,
keen insights, special interest, and concern for what I was doing.
To my family, I thank my son, Rob, and daughter, Erin, and their spouses who
cheered me on, often wondering in amazement why I would attempt to go down such a
path at this point in my life.
Lastly, I extend a very special thanks to my wife, Sandra, to whom I dedicate this
work. I am extremely grateful for her endurance, sacrifice, and constant encouragement
through this odyssey that started more than six years ago. While I often have wondered
where the time has gone, I never could imagine spending it with anyone else.

TABLE OF CONTENTS
page
ACKNOWLDEGMENTS
LIST OF TABLES vii
ABSTRACT viii
CHAPTERS
1 INTRODUCTION 1
Introduction 1
Statement of the Problem 2
Need for the Study 4
Background 11
Purpose 12
Research Questions 15
Delimitations of the Study 16
Limitations of the Study 16
Definition of Terms 17
Summary 20
2 REVIEW OF THE LITERATURE 22
Perception and Knowledge 22
Neurology of Vision 24
Visual Perception Theories 25
Fast Seeing 28
Encoding 31
Relationship of Memory and Learning 35
Textual versus Pictorial Memory 36
Recognition and Recall 37
Verbatim versus Gist Memory 39
The Effect of Arousal on Memory 42
The Heuristics of Pictures 47
Pacing versus Presentation Speed 50
How Learner Attributes Affect What is Learned 53
IV

Cognitive Style 53
Impulsive-Reflective Style 54
Factors from Film and Television That May Affect Learning 58
Preconceived Mental Demands 58
Exemplars and Other Formal Features 60
Symbol Systems 66
Evaluating the Educational Impact of New Media 69
New Media and Memory VO
Effect of Fast Cuts on Memory 73
Presentation Speed versus Content 73
Montage 75
Summary 77
3 METHODOLOGY 81
Introduction 81
Population 82
Sample 82
Instrumentation 83
Test descriptions 88
Initial Development 88
Pilot Test 89
Test Administration 90
Stimuli 91
Methodology 92
Research Design 93
Hypotheses 95
Summary 96
4 ANALYSIS OF THE DATA 98
Introduction 98
Results 98
Summary 106
5 CONCLUSIONS AND RECOMMENDATIONS 109
Introduction 109
Findings 110
Discussion Ill
Presentation Speed Ill
Gender 112
Verbatim versus Gist Memory 112
Cognitive Style 113
Interaction Between Speed and Style 114
Implications 114
v

Recommendations 116
Summary 120
APPENDICES 122
Appendix A Screen Shot of MFFT-20 122
Appendix B Screen Shot of Sample Test Question 123
REFERENCES 124
BIOGRAPHICAL SKETCH 140
vi

LIST OF TABLES
Table Eage
I.2x3 Two-Way ANOVA (204) Subjects 94
2. Multivariate analysis of variance (204) Subjects 94
3. 2 x 3 two-way ANOVA (129) Subjects 94
4. Source table of analysis of score variance by presentation speed 99
5. Score means and standard deviation for presentation speed 99
6. Source table of analysis of variance by presentation speed and gender 100
7. Score means and standard deviation of scores for gender 100
8. Source table of analysis of variance for speed versus gender 101
9. Score means and standard deviation for speed versus gender 101
10. Source table of analysis of variance for verbatim score for speed 101
II. Source table of analysis of variance for gist score for speed 101
12. Multiple comparisons between speed and gist scores 103
13. Score means and standard deviation for gist versus verbatim test items 103
14. Source table of analysis of variance by presentation speed and cognitive style 103
15. Means and standard deviation for overall scores for style 104
16. Source table of analysis of variance for verbatim and gist and cognitive style 105
17. Means and standard deviation for gist and verbatim scores for style 106
vii

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
THE EFFECTS OF COGNITIVE STYLE AND GENDER
ON VERBATIM AND GIST MEMORY FOR
RAPIDLY-PRESENTED MONTAGE VIDEO
By
Robert F. Kenny
May, 2002
Chair: Lee J. Mullally, Ph.D.
Major Department: School of Teaching and Learning
Younger students live in a media-centric world. Researchers have shown that
todays youth spend more time watching television and movies than most any other
leisure-time activity. Moreover, the presentation speed of passages on commercial
television has increased significantly in the past fifty years. Researchers have shown that
viewers automatically leam to cope with symbolic pictorial presentation methods through
repeated exposure to television and its visual patterns. Increased exposure to rapid
sequence and presentation speed brought on by rapidly-presented montage found in
television programs aimed at youth raises the question as to whether these individuals
remember more from this type of presentation techniques than can their adult
counterparts. Recent advancements in technology have changed the way viewers look at
and interpret video media. Rapidly presented video media place additional attentional

demands on their viewers. Yet these viewers have been found to be able to perceive and
remember content from these messages.
The purpose of the study was to investigate whether youthful viewers are able to
perceive video images that are presented very rapidly and then immediately remember
verbatim information and assimilate the contextual gist of the overall passages. The study
also attempted to determine whether or not there are differences in the way males and
females process these images. In addition, subjects were categorized as being either
impulsive or reflective to determine whether cognitive style has an effect on pictorial
memory.
The study revealed that memory is not significantly affected by changes in
presentation speed when test results were viewed as a composite of verbatim and gist
scores added together. However, when the two scores were separated, gist memories
were positively affected by increases in presentation speed. The study also revealed that
cognitive style had a significant impact on pictorial memory.
IX

CHAPTER 1
INTRODUCTION
Media educators and theorists for years have been analyzing Marshall McLuhans
famous quip, the medium is the message (Meyrowitz, 1985). In some regard, McLuhans
statement may be considered a pre-emptive retort to later critics of educational media like
Richard Clark (1983) who claimed that media are mere vehicles that deliver instruction
but do not influence achievement any more that the truck that delivers groceries causes
changes in nutrition (p. 445). Communications theorists like Walter Ong (1982) not only
agreed with McLuhan, but also extended the meaning of his message to imply that the
types of media people use define the way they think. Ongs notions bring to mind
possible questions as to whether todays media-centric youth perceive differently than
previous generations, with implications as to the kinds of mediated instructional
strategies that might be successful in motivating them to learn as well as providing
perceptual stimuli for recognition and recall.
The predominant types of production techniques used in todays electronic media
are changing. Encouraged by the successes of early pioneers of rapidly presented music
video montages on networks like MTV, VH1, and Nickelodeon, and helped by rapid
advances in technology, todays television producers regularly communicate very
complex messages using fast-cuts and video montage (Stephens, 1996). Significant
increases in viewership of these programs may be providing fertile opportunities for
todays youth to practice their ability to receive and assimilate this fast-paced message

2
delivery approach. According to Seward-Barry (1977), sleeping has become the only
activity that occupies childrens time more than watching television or playing video
games.
For these reasons, a change may be warranted in the way in which educators view
the role video media play in initiating learning opportunities and conveying more
complex or abstract thought. This thinking appears to be incompatible with earlier
communications theorists like Edgar Dale (1969), who felt that television finished
somewhere in the middle of a twelve-point influence scale in its ability to convey
contextual ideas. New digital media that employ visual imagery supplemented by quick
motion, sound, and computer editing (Stephens, 1996, p. 69) may be ready to take on the
same predominance as a communication medium in only fifty years that it took
Guttenbergs great invention five hundred years to accomplish. Corcoran (1981) defined
intelligence as a skill in a particular medium and suggested that the symbolic codes used
in that medium that serve communication purposes and are internalized by a receiver
become an authentic tool of thought. Research studies in which technology (i.e., media)
was successfully used as cognitive mind tools appear to back up this thinking (Hokanson,
2000; Jonassen, 1996).
Statement of the Problem
Researchers and educators looking into the intrinsic instructional value of video
media have presented conflicting views on the role visual perception plays in attention,
motivation, and recall. There have been several studies that have looked into the potential
affect mediated coding systems have on cognition (Davis, Scott, Pair, Hodges, &
Oliverio, 1999; Nugent, 1982; Seidman, 1981; Walma van der Molen & Van der Voort,

3
2000). While many theorists have looked into the use of video, most of them had little
good to say about its ability to bring anything new to the table with regards to using it as
a communicative medium (Berio, 1960; Calvert & Scott, 1989; Ide, 1974; Kozma, 1986).
This may have been due to the limitations imposed by the technology in use at the time.
Recent technological advances in commercial television production techniques allow
todays producers to readily integrate fast-cuts and montage and provide new tools to
more easily communicate complex thought using a pictorial narrative structure. A
correctly constructed rapid-cut montage passage has been found to add clarity because of
the phenomenon that the interpretive whole of a montage segment literally communicates
more than the sum of its parts (Hitchon, Druckler, & Thorson, 1994; Stephens, 1996). In
other words, it is the composite whole of all the visual images in a passage or segment
when considered all at once that gives it extended meaning. In addition, newer editing
techniques have evolved that emphasize the perceptual continuity of a rapidly presented
image-based narrative structure, rather than the classical point of view of editing that
stressed the importance of applying strict rules in order to obtain slow and smooth
transitions between successive shots (dYdewalle & Vanderbeeken, 1990).
One of the specific problems being addressed in the current study is to update the
thinking with regards to using video as an instructional medium in light of these new
technological developments. A review of the literature of the past twenty years has
already yielded some studies into the impact of rapid video editing in an educational
setting (Stephens, 1996; Wetzel, Radtke, & Stem, 1994). However, more recent advances
in technology that have superceded those studies and an increased usage of rapid-cuts and
montage production techniques in commercial television programming have created the

4
need to update any earlier research that does exist. Lang has conducted more recent
studies that have looked into the effectiveness of fast-cuts (Lang, 1994; Lang, 1996; Lang
& Basil, 1998; Lang, Bolls, Potter, & Kawahara, 1999; Lang, Zhou, Schwartz, Bolis, &
Potter, 2000), but hers deal with television in a casual viewing environment. The present
study was designed to look at the effect of these rapidly presented visual passages in an
educational setting. It is hoped that a new look at rapid visual processing brought on by
this study might lay the groundwork for educational media producers to update their
thinking about rapidly presented video montage by providing a research basis that
appears to be lacking currently. Specifically, the questions that are being reviewed are:
1) Is it possible to remember content solely from rapidly paced visual montage
that is not supplemented with some form of verbal narrative?
2) Is there a difference between gist and verbatim memories for rapidly presented
videos?
3) How do individual characteristics such as gender and leaming/cognitive styles
affect a students ability to process fast-cuts/montage video presentations in a
classroom setting?
Need for the Study
As has been stated already, most of the previous studies into the impact of fast-
cuts/montage have concentrated on commercial television viewing (Bryant & Rockwell,
1991; Lang & Basil, 1998; Lang et al., 1999; Lang et al., 2000; Zillman, 1991). There has
been one attempt to look at the use of fast cuts/montage in an instructional setting (Keller,
1976). However, that investigator looked at viewers as a collective whole with little
regard for individual differences in personalities and capabilities such as cognitive style.
Further, this study is more than twenty-five years old and its subjects were college-level
students enrolled in communications classes, rather than school-aged students studied in

5
their formative educational environments. The current study attempts to resolve some of
these contradictions to see the effect fast video presentations have on immediate
memories for their content.
Educational media researchers have had to contend with a conflicting view that
holds that the relatively fast presentation speed of television programs creates an
environment that may be detrimental to attention and recall (McCollum fe Bryant, 1999;
Neuman, 1976). To the contrary, Intraubs (1999) studies into conceptual masking have
shown that humans are able to recognize and recall pictorial presentations when a
minimal amount of lag time separates individual images. In addition, some of the more
recent successes with childrens programs have shown educational successes that are due
to their rapid and rhythmical presentation speed (Anderson, Alwit, Lorch, Anderson and Bryant, 1983; Anderson <& Collins, 1988; Pearl, 1982). Still others have
proposed that presentation speed in instructional media may actually add interest to
otherwise uninspiring content (Canelos, 1986; Edgar, 1997; Hawkins, Pingree, Bruce, <&
Tapper, 1997; Hill fe Lang, 1993).
Previous examinations into the effects of casual television viewing have not
translated too well into an educational setting (Salomon, 1994). As Kozma (1986) stated,
"viewership should not be confused with learning" (p.14). However, once one delves
deeper into these studies, four possible reasons for this lack of transfer become apparent.
First, the reputation that commercial television has held for being nothing more than an
entertainment device has hurt its standing in educational circles and has caused
detrimental pre-conceived notions about how viewers are to be properly introduced to
televised content (Wetzel et al., 1994). Second, many of the schemas used in commercial

6
television are considered by some to be nothing more than prototypical, trite, and overly-
familiar formulas that reduce attention and concentration because they have been over-
learned (Anderson et al., 1979; Langer & Imber, 1979). This schema also leads one to
revert to a mindless routine in which the material is ignored or receives a low level of
attention (Wetzel et al., p. 169). This alleged over-familiarity with format has lead many
educators to believe that viewers will have difficulty responding appropriately to
educational televised presentations, unless some form of outside instructional
intervention is also inserted (Wetzel et al.). The third reason is that many of the studies
into casual viewing tended to lump all viewers into a single category (Lang and Basil,
1998; Lang et al., 1999; Lang et al., 2000; McLuhan, 1964; Neuman, 1976; Tyner, 1998;
Walma van der Molen & Van der Voort, 2000). Classical instructional models tend to
validate the value of classifying learners by their individual differences (Gentry, 1998;
Joyce, Weil, & Calhoun, 2000).
The fourth and possibly most important reason for the lack of correlation between
studies of casual viewing and those performed in an educational setting is that the
learning environment is thought to present a different set of circumstancesa unique view
as it were. This is based on the importance placed on the medium to be evaluated as to its
unique ability to bring about some type of alteration of intellectual behavior or thinking
process. Although many of the symbolic (i.e., intellectual) combinations found in non
verbal endeavors such as music, painting, and dance can be displayed directly on
television, there has been some question as to the extent to which transformations in the
thinking process are actually created by viewing the medium, and whether any changes
that might occur are of any significance (Ide, 1974). Television has not been given credit

7
for yielding any new intellectual construct of its own. Previous studies into the value of
using television as a medium for intellectual change have demonstrated mostly negative
progress in that the results showed television did not actually interfere with learning, or
that it was not less effective than other forms of media (Thompson, Simonson, &
Hargrave, 1996; Wetzel et al. 1994). The current study attempts to clarify where rapidly
presented montage might fit into the overall instructional scheme.
Previous attempts at investigating rapid video and montage presentation were
limited by some incorrect assumptions as well as limitations on technology. For example,
Intraubs (1999) experiments that demonstrated weaknesses in an individuals ability to
understand and remember briefly-glimpsed images dealt with pictures that were not or
only very loosely related. The proper use of montage implies that the pictures included
are at least conceptually related. Intraub indicated that subjects might be able to hold
more than one picture at a time in a conceptual buffer, so long as the series was not too
long (p. 57). Keller (1976) tried to fix this shortcoming in his dissertation but ran into
some serious technical flaws in his overall design. Intraub even admitted that his
experiments might have been more successful if they included pictures that were related
to one another. His intuition may have been correct. It appears that humans may have the
ability to construct meaning from these types of presentations through the use of
interpretive coding (i.e., the process by which meanings are put together from specific
parts of visual communications). In describing their research in teaching Native
Americans how to use film to communicate meaning, Worth and Adair (1997) noted that
the process of coding has been neglected in the study of most of the fine arts, including
film. Their comments alluded to the fact that the form of a medium might be what carries

8
its meaning. In an earlier attempt to extend this notion to television, Pearl (1982)
discussed the relationship between form and content and admitted that it is the form (that
is, the way it uses verbal and linguistic codes) not the content of television that makes
it a unique communicative medium. However, she also cautioned that form and content
cannot always be distinguished no more than grammar and meaning in any verbal
language can (p. 24). She went on to say that some forms are unique to a particular
medium and apply syntactical meaning only in the context of that medium. For example,
slow motion is not real and its meaning must be learned. Other researchers have shown
that these unique formats, once learned, generally become used by people in their own
thinking (Barnett, 2000). That is perhaps why, for example, when one applies slow
motion to a video message, it generally carries some contextual or emotional connotation
each time it is used. The current study tries to apply the same logic to fast cuts and video
montage to see if this presentation format can be interpreted in such a way as to portray
some implied meanings of its own.
Previous research into using increased or compressed presentation speed and
movement in multimedia has been the subject of controversy in the literature.
Some studies of commercial broadcast video have shown that the relatively rapid
presentation speed of programs may have a deleterious effect on recall (Alwitt, Anderson,
Lorch, & Levin, 1980; McCollum & Bryant, 1999). Other research has shown that the
instructional value of video is aided by a systematic presentation speed of information as
a presentation strategy (Comstock, Chafee, Katzman, McCombs, & Roberts, 1978). Still
others have shown that presentation speed and rhythmicity in leisure-time media can
actually heighten enjoyment, enhance motivation, and can play an important part in

9
determining the affective or emotional response of message receivers" (Seidman, 1981, p.
49). Lang et al. (1999) and Lang et al. (2000) looked into presentation speed as its own
construct, comparing/combining it to/with arousing content. They discussed the effect of
adding interesting and arousing content as having a positive effect on cortical arousal,
and therefore, recall and recognition. Lang et al. (2000) also alluded to future research
that should continue to probe the shape of the relationship between presentation rate and
recall and test even faster rates of edits to determine whether there is a point at which
memory begins to decline. They suggested that producers who want their messages to be
remembered should create arousing messages that are presented slowly or at medium
speed, or calm messages that are presented at medium or rapid speed. They concluded
that producers should not create messages that are either calm and slow or arousing and
fast.
The effect of integrating increased message presentation speed directly into
instructional messages in an educational environment whose content might be considered
by some students as less than arousing will be examined in the current study. It should
also be noted that Lang et al. (2000) considered cuts to be fast if they changed at a top
rate of eleven to twelve per thirty-second segment (i.e., one every 2-3 seconds). The
current study looks at a top rate of one cut every one-third to one-half of a second, or
roughly ten times as fast. The slow messages will run roughly equivalent to Lang's top
rate.
It should also be noted that the term cuts. as it was used in connection with
commercial television research, differs from what how it was used in the current study. In
those studies, a cut might simply mean a wipe or a swipe, or a change from one camera

10
angle to another. The content may not change at all, only the viewers perspective. In the
current study, a cut always signifies a change in content. One picture or image was
replaced with an entirely new one.
Lang et al. (2000) indicated that under very broad interpretation, the speed of
message presentation of motion and edits/cuts can be considered a form of cueing.
Downs (1989) also referred to presentation speed in a similar way, referring to it as a
messages domain attribute (p. 3). Salomon (1979) acknowledged that media attributes
are those that are within the mediated stimulus, possibly shared to some extent with
other forms of [sic] media and make the presented information more comprehensible or
better memorized by learners of particular characteristics (pp. 5-6). Under Salomons
definition, symbols include most objects, marks, events, models, or pictures (p. 29). It
is assumed in this study that the rhythmic patterns afforded by fast-cuts are an event.
Where the current study varies from previous research is that it takes the interrogation of
symbols and attributes to another level. The current study aims to show that rapid
presentation speed (also referred to as fast-cuts) may be considered an invaluable
communicative attribute of media and is, therefore, capable of being studied separately to
discover its contribution to learning. This study aims to determine to what extent that
rapid presentation speed of video images either aids or interferes with recognition and
recall (specifically, gist and verbatim memories), taking into consideration the changes
that appear to be taking place in an ever-increasing media-centric society. Even if fast,
moderate, and slowly presented visual stimuli are all shown to be equivalent in their
ability to stimulate recognition and recall, educational research might benefit from
discovering whether visual stimuli presented at an increased delivery pace can provide

11
similar instructional outcomes in a more efficient and stimulating manner -something
found by some researchers to be a more practical outcome measurement for using video
media than achievement alone (Kini, 1994; Peck, 1987).
Background of the Study
Younger students live in a media-centric world. Researchers have shown that
youth today spend more time watching television and movies than most any other leisure
time activity (Pearl, 1982). In addition, the presentation speed of passages on commercial
television has increased significantly in the past fifty years (Stephens, 1996). Researchers
have shown that viewers automatically learn to cope with symbolic presentation methods
through repeated exposure to television and visual patterns (Abelman, 1995; Bargh,
1988; Carr, 1982). Because of their increased exposure to rapid sequence and
presentation speed brought on by fast-cuts/montage found in television programs aimed
at youth, it may also be assumed that these individuals might be able to comprehend these
messages on a much wider scale than can their adult counterparts. One cannot continue to
assume that exposure to video, regardless of presentation speed, is simply a passive
viewing activity. These advancements in media technology are changing the way viewers
look at and interpret video media. In addition, the widespread availability of production
techniques provides easy access to capabilities that allow people to use video media to
easily create their own content. It has been widely shown that these acquisition (i.e.,
production) opportunities also increase exponentially ones ability to comprehend content
delivered in like form (Tyner, 1998).

12
Purpose of the Study
This study was designed to determine if there are differences in the way visual
montage messages that vary only in their presentation speed (i.e., fast, medium, or slow)
are perceived and immediately remembered in an educational setting. Recent reports
found in the literature, (Brainerd & Reyna, 1990; Brainerd & Gordon, 1994; Reyna &
Kieman, 1994) have suggested that researchers have been able to successfully parse
memory into verbatim (i.e., precise/literal) memory for specific details and gist memory
(i.e., contextual or contextual remembrances similar to that which is assessed in reading
comprehension tests). This study looks at immediate memory as a whole and separately,
using these same categorizations. The educational setting was determined to be several
ninth grade classrooms that were categorized by gender and also further delineated by the
subjects individual cognitive/leaming style. The overall sample set for this study came
from that population group. For purposes of this study the identification of cognitive style
was limited to the reflective-impulsive scale, as originally developed by Jerome Kagan
(1965; 1966), and later refined and re-catalogued by Cairns and Cammock (1984).
Identical videos that varied only in their speed of message delivery were presented to
three randomly assigned groups of students. The procedural requirements for
implementing the cognitive style test instrument indicated that the analysis was to be
performed on a smaller subset of the sample base to see if the variance in delivery speed
also affects gist and verbatim memory, and if there might be any interaction with their
prospective cognitive styles.
Previously, presentation rate had been found to affect immediate memory in
examinations of viewers in a commercial viewing environment that compared several

13
contrasting theories about memory and cognition (Lang et al., 1999; Lang et al., 2000).
These investigations concluded that individuals have limited capacity for processing
information and indicate that a rapid influx of visual images maintains an individuals
orientation reflex and stimulates cortical arousal. Unlike other forms of arousal, cortical
arousal has been determined to maintain a persons ability to pay attention (Pearl, 1982;
Reeves, Thorson & Schleuder, 1986; Zillman, 1991). The earlier studies also alluded to
conflicting findings about the impact of the content on attenuation and arousal
(Anderson, 1986; Archer, 1965; Basil, 1994), maintaining that arousing content
combined with too fast a pace/message delivery would normally result in loss of retention
and recall. Unlike these previous studies that included occasional highly arousing
content, the current one used an educational topic that might not be considered by many
students as arousing (i.e., American history) in an authentic classroom setting.
The current study also attempted to improve upon some of the various
shortcomings found in the one other earlier study (Keller, 1976) that used very similar
but out-dated methodologies. In his doctoral dissertation, Keller used a self-created
kinestasis film (i.e., rapidly cut video montage) in an attempt to define what effects on
recall are realized by speeding up transmission or information to less than one second per
frame. Standards previous to Kellers study deemed a presentation to be rapid if cuts
occurred every five to ten seconds. Keller also endeavored to derive empirical data to
determine how well the kinestatic process worked on recall and how efficient visually
compressed presentations might be for instruction, trying to find out whether the
kinestasis technique was more than a simple artistic endeavor. However, by his own
admission, Kellers research came up short because some of his methods may have been

14
lacking in rigor. For example, he utilized random pictures as opposed to those with
similar contextual content. As schemata have been long shown to be of importance to
learning (Bartlett, 1932) studies looking into recognition and recall need to use a single
context. Additionally, most of the problems Keller ran into appeared to be related to
production and processing techniques that reflected the state of the art of video
production at the time (Keller). His difficulties ranged from dissimilar content, picture
size that was too small, and brightness that caused a deviation in results based on the seat
location of his subjects. Due to technological shortcomings, he utilized what are known
in the industry as wipe-ons/downs as his motion technique. Wipes are production
techniques utilized in television broadcasting in which one picture is replaced by another
that is gradually wiped across the screen. An edit or cut, on the other hand, is where one
picture immediately and completely replaces another. In this technique, the succeeding
picture frame partially masked, or interrupted the previous one and vice versa, causing a
potential confound. There has been some reference in the literature (Goodglass, 1971)
that indicates that visual masking introduces a new, unnecessary variable into studies on
visual processing. Keller also had problems in his test phase because he could not cause
the pictures to replicate the movements that occurred in those same pictures when he
presented them originally. He admitted that this approach was flawed in that the whole
picture could not be viewed until the end of the procedure, causing a degradation of
recognition and recall. The current study attempts to clarify some of those issues. Keller
referred to the need to find another motion procedure (p. 90). The current study used a
commercially-made video montage that refers entirely to the same subject matter, for

15
which permission has been granted by the copyright holder to modify for use in this study
(R. Wright, personal communication, March 21, 2001).
Finally, Kellers (1976) subjects were college level communications students who
were segregated by grade point average. He identified the need to find a more rigorous
randomization variable than GPA, noting that there was a significant interaction effect
among the subjects, who accounted for almost 24% of variance in picture memory
recognition, while the treatment effect only accounted for 6%. The current study
attempted to look at an additional classification method (i.e., cognitive style) that had
been identified and validated in research subsequent to Kellers study (Green, 1985;
Kagan, 1965; Okun, Callistus, & Knoblock, 1979; Ridberg, Parke, & Hetherington, 1970;
Salkind & Wright, 1977; Witkin & Goodenough, 1981).
Research Questions
Following are research questions raised in this study:
1. Is there a difference between the amount remembered in terms of verbatim
(specific details) and gist (contextual content) memory when changes in
presentation speed occur?
2. Are there gender differences in ones ability to remember information presented
in rapid video presentations?
3. Are verbatim and gist memory affected by a learners cognitive learning style?
These specific questions are intended to lead to an over-riding question as to whether if
the amount of information retained through fast-cuts/montage is found to be equivalent to
more moderately paced instruction, is a faster delivery more educationally efficient, as
determined by the fact that the same amount of material can be successfully presented in
a shorter time interval?

16
Delimitations of the Study
This study was conducted with certain limitations. In previous normative data
validation and reliability studies for the cognitive style testing instrument being used in
this study, subjects were limited to ninth grade students from a specific geographical
area. Only those students who agreed to participate and whose parents signed a consent
form are included in the study. The study only looked into the amount of information that
could be immediately remembered from the video passages. While immediate memory
has been deemed an important prerequisite to learning, this study makes no
representation with regards to whether the material is subsequently learned (i.e., retained
for longer periods of time, or able to be utilized in subsequent tasks) by the subjects. The
fact that learning may or not actually take place implies that additional pedagogical and
cognitive techniques are subsequently employed, and is beyond the scope of this study.
Limitations of the Study
This study contains limitations and assumptions that may affect the
generalizability of the results. The study is only generalizable to the population from
which the sample is taken. In this case, subjects came from one high school of a mixed
ethnographic and demographic make-up in North Central Florida. While it is understood
that students who have previous knowledge of the topic (i.e., American history) might be
more likely to recognize and recall the pictorial representations in the videos, it is also
assumed that this knowledge base was most likely equally distributed throughout the
sample set because subjects were selected at random. Also, as the students were
randomly assigned to the three presentation groups, it was determined that any previous
history knowledge would not materially affect the results.

17
It also should be noted that this study has limitations with regards to its ability to
predict actual learning. The study was designed to only define certain rules about
presentation speed of an initial presentation that may become one component of an
overall instructional strategy. By itself, the study does not attempt to make any
predictions or value judgments about those strategies. Rather, it simply attempts to
support the validity of fast-paced video segments that might become building blocks in
sound pedagogical patterns, as delineated in instructional models that might be
subsequently proposed by future educational research.
Definition of Terms
Following are specific definitions of terms used in this study:
Cognitive style is the stable ways in which persons differ in perception and
encoding of information (Wittrock, 1979). Specifically, it is Kagans scale (1966) that
was later modified by Cairns and Cammock (1984) that is used as a basis for the current
study.
Conceptual masking is concept that describes the phenomenon that occurs when a
new visual image is introduced into temporary memory and thereby occludes the
previous one (Intraub, 1999).
Gist Memory has been successfully used as a treatment effect in several studies
looking into prose memory (Biederman, Rabinowitz, Glass, & Stacy, 1974; Brainerd &
Gordon, 1994; Cowan, 1998; Reyna & Kieman, 1994; Voss, Tyler, & Bisanz, 1982). Gist
memory is one of two types of determinations (the other being verbatim memory) as to
the amount and quality of immediate memory. In previous research gist determination
required subjects to try to identify the main idea in a passage or correlate among several

18
scenes. The objective of assessing gist memory in this study was to determine whether
the subjects comprehend the basic contextual themes presented on the video and to
compare it to subjects ability to immediately remember specific (i.e., verbatim) details
from the video presentation.
Kinestasis is a term coined by Charles Braverman (1969) that is derived from
combining the Greek words for moving and static. The term refers to a film or video
process that presents still and/or moving images at speeds that create the illusion that they
are moving in a continuous flow. The treatment video used in this study is a video copy
of a kinestasis film developed by Braverman, and whose copyrights are owned by
Pyramid Media.
The term media includes films, television, and compressed motion pictures found
in digital multimedia. Previous researchers into presentation speed (Wetzel et al., 1994)
and compressed motion pictures (Arnold, 1996) delved into one or more of these types of
media and have referred to them as virtually interchangeable in meaning. This study also
refers to the term new media. This term connotes an inference about the convergence (the
merging of television, computers, and the telephone) that is taking place. The current
study uses screen media and video media interchangeably when referring to its
capabilities to present a pictorial story. It uses new media to mean traditional media that
are being technologically advanced due to convergence.
Minimal unit of analysis is that product of semiotic analysis which is the smallest
interchangeable element that still retains the basic properties of the whole symbol system
but cannot be further subdivided without losing those properties (Corcoran, 1981). Hill
(1981) referred to this concept as a symbol, or the basic unitary element of intellectual

19
activity. In the assessment of reading comprehension, a word is often associated with the
unit of analysis (Murdock, 1982). In this study, individual picture frames are considered
as being the minimal unit of analysis. These may be equated to a paragraph (Corcoran)
for purposes of comparing the results to reading comprehension-type assessments.
Orienting Response (OR) is the automatic response to a stimulus that helps an
individual focus attention on that stimulus and is often captured through psychometric
means (i.e., an attentional blink, or a look (Lang & Basil, 1998; Lang et al., 1999; Reeves
et al., 1986; Thorson & Lang, 1992). An OR can also cause an overt arousal response that
may be detrimental to recall. In this study the orienting response is considered to be a
potential confound and is dealt with in the methodology of the study. Subjects will be
provided a short preview of the video presented at the same speed in order for them to get
used to the format.
Presentation speed is the rate of scene changes or edits of a video media
presentation and is the operative term utilized in this study. In many studies used to
support the premises in this study, pacing has been the operative term used to refer to the
speed of message presentation (Corcoran, 1981; Hill & Lang, 1993; Lang et al., 1999;
Lang et al., 2000; McCollum & Bryant, 1999; Reeves et al., 1986; Schale, 1971,
December; Wagely, 1978; Zillman, 1991). The term pacing also carries certain pre
conceived connotations in educational circles that have nothing to do with the premise of
the current study. In order to avoid confusion, all references to pacing were replaced in
the current study with the terms presentation speed, or speed of message presentation.
Screen media are that which use controlled exposure of sequenced images on a
screen (either on a video or movie screen), often in fixed relationships to speech, music,

20
and other sound (Corcoran, 1981). Much of the literature presented in Chapter 2 refers to
screen media. For purposes of this study screen media and video media are considered to
be identical.
Verbatim/rote memory refers to subjects being able to identify whether specific
objects within a scene, or overall scenes occurred in the video that has just been seen
(Brainerd & Gordon, 1994; Reyna & Kieman, 1994). In some studies, verbatim memory
also has been referred to as object recognition (Tse, Vegh, Marchionini, & Shneiderman,
1999). For purposes of this study, verbatim memory is assessed positively and falsely.
Subjects were asked to select an image that was presented in the video display (positive
identification) from a group of pictures immediately after presentation of the test video.
Subjects were also are asked to decide whether any picture or series of pictures in a group
actually appeared in the video (false identification).
Summary
Meyrowitz (1985) posited the idea that new media, through its ability to merge
many formerly distinct knowledge situations, appear to be breaking down the
boundaries among various disciplines, opening new dialogues, and fostering the
development of cross-disciplinary areas of study (p. 327). He also speculated that these
new kinds of electronic media may be "introducing our children to a different way of
thinking that involves the integration of multiple variables and overlapping lines of
simultaneous actions" (p. 326). This assertion proposes, among other things, that
electronic video media have already greatly reduced the influence that time and location
used to have on what people know. It is not, therefore, unreasonable to agree with
Stephens (1996) proposal that a transition may be occurring in the way the youth of

21
today think. This appears to be at odds with the linear thinking processes associated with
print media. The current study, through the questions posed in the current chapter and a
review of the literature found in the next, attempts to find out how significant these
changes in perception of new media are.

CHAPTER 2
REVIEW OF THE RELATED LITERATURE
Perception and Knowledge
When humans learn from the world around them, they receive stimuli
(information) via one or more of their five senses: audition (hearing), gustation (taste),
olfaction (smell), kinesis (touch), and vision (sight) (Murch, 1973). Processing received
stimuli can be automatic or be the result of some literal attempt to react to it. For
example, when one smells food, it may make him or her feel hungry. This may happen
with or without that person actually realizing what is happening.
Besides being the catalyst for automatic processing of bodily functions,
perception has also been considered by educational researchers as a fundamental building
block for knowledge and learning. Murch (1973) posited that humans must be able to
perceive something before they are able to learn about it. This information may be
firsthand, or via an artificial artifact (also known as a medium). Gibson (1969) defined
perception as a process of receiving stimuli and consciously reacting to them by
assigning some meaning. Accordingly, humans tend to react to perceptions individually,
based on previous personal experiences, making this experience very subjective. Winn
(1982) described the process of learning as a taxonomy of three procedural steps. After a
person first receives information to be stored, and later when he or she attempts to
retrieve it from memory, a long process of transformation, abstractions, and elaborations
takes place. The first step is perception, generally considered an automatic operation, in
22

23
which features are integrated into complete visual displays. Assimilation is the next level
of complexity and involves the recognizing and integrating of information into schemata
for later recall. Analogy is similar to assimilation but is an even more complex process
that involves the temporary assimilation of new information into abstract schemata that
embody other concepts that are peripherally related to it. Explained in this way, Winn
alluded to perception as a necessary but insufficient step in the whole learning process.
Bruning, Shraw, and Ronning, (1999) posited that the body holds perceptions in a
series of registers called memory so that perceptual analysis can occur before that
information is lost. When all is working well, perception allows humans to detect the
incoming stimuli and allocate attention to them. Through pattern recognition, individuals
may recognize a stimulus because of their previous experiences. However, on occasion,
pattern recognition may be limited, thereby preventing any learning. Marr (1982)
proposed a structural approach to the relationship between perception and intellectual
endeavors. He imagined that knowledge is stored in the brain as a set of statements,
templates, prototypes, or models of distinctive features of a particular object or class of
objects that has been perceived. Bruning et al. (1999) appeared to agree, stating that using
specific, structured descriptors as an approach to recall and recognize patterns may be
very helpful for teaching because it suggests that perception can be a guided activity.
Teachers can instruct students the proper knowledge needed for an accurate structural
description (p. 27). Although the visual field may be partially obstructed, the mind is
still able to make determinations about what it sees, using clues and mental mappings
(visual patterns) from previous experiences (Seward-Barry, 1997). Gestalt psychologists
refer to this process as visual intelligence and base their theories on this emphasis of the

24
mind being able to discriminate parts to the whole and make other spatial determinations
that are described further later in this chapter. But first, to understand more fully the
Gestaltist view of visual intelligence and how it relates to this study, it might be best to
consider some of the literature concerning the neurological process involved with vision.
Neurology of Vision
According to Murch (1973), biologists define vision as a process where the eye
sees the world through a biological process of photic radiation, an energy source that
stimulates the eye. The process begins as reflected light that bounces off objects in the
environment. This optic array is focused by the cornea and lens onto the visual field on
ones retina that lines the back of the eye (Seward-Barry, 1997). These images are then
transmitted to the brain through various routes, one being the cortex/limbic system. This
system provides stimulus to the brain, including the amygdala: the sub-cortical region
within the temporal lobe (LeDoux, 1991). According to Seward-Barry, researchers have
identified four parallel systems involved in the different attributes of vision: one each for
motion and color, and two for form. Goldstein (1989) points out that perception is based
not on direct contact with the environment, but on the brains contact with electrical
signals the represent the environment. We can think of these electrical signals as forming
a code that signals various properties of the environment to the brain (p. 50). In other
words, perception as a biological function may be described as being based on a coding
system that takes in the visual signal and is interpreted by the brain. LeDoux (1991)
viewed this interpretation process as a dual operation. He proposed that sensory
perceptual processing takes two routes. The first travels along the amygdala and readies
the body to react, even before it recognizes the need to do so. The second route is through

25
the neo-cortex where the signal can be analyzed and then sent to the amygdala and
emotional response added after cognition (Seward-Barry). This dual processing theory
provides an insight as to why visual perception takes on such a personal meaning. It also
introduces the idea that will be further developed later in this chapter that there can be
two very different emotional reactions to perception: one helpful to cognition, and one
that can be detrimental.
Visual Perception Theories
For some theorists, seeing is more than a simple biological function. According to
Seward-Barry (1997), visual perception is actually taking the visual image, combining it
with other data from the other senses, and synthesizing it with previous experiences.
Perceptual psychologist J. J. Gibson (1979) made a distinction between the visual field,
the image that appears on the retina, and a mental creation of what comprises our world,
giving further credence to the fact that to an individual, reality is very interpretive and
subjective. To constructivist theorists, visual reality is simply a map-like image, the end
product of a process that begins with light refraction in the environment and ends in the
intricate and complex dynamics of the mind (Seward-Barry, p. 15). Viewed this way,
perception is what intercepts an outside stimulus for a person to create an individual
reality. When video producers present their images in unique ways (such as utilizing
rapidly presented montage) they simulate this mapping process and create opportunities
for viewers to construct their own views and interpretations.
Gestalt theory, as outlined by Ellis (1938) and later by Wertheimer (1959), shares
some common beliefs with constructivists. However, this theory proposes a more
consistent view of perception among individuals through the concept of idea grouping

26
(Seward-Barry, 1997). Characteristics of stimuli cause humans to structure and/or
interpret a visual field in certain ways or groupings. Wertheimer noted the following
primary factors in determining these groups:
Proximity Humans tend to group visual elements together according their
spatial nearness.
Similarity Humans tend to group together items similar in some aspects (i.e.,
their design, size, shape, textures, etc.).
Simplicity Visual items are grouped together by humans into simpler figures,
according to their symmetry, regularity, and/or smoothness.
Closure Humans tend to complete on their own an incomplete picture or
other visual entity.
Wertheimer was particularly concerned with problem solving. While problem
solving per se is outside the scope of the current study, these ideas form an important
basis for some of the assumptions about the affect that rapidly presented video montage
may have on memory and learning. Wertheimer felt that the essence of problem solving
behavior is ones ability to be able to see the overall structure of a problem through the
use of one or more of the above classifications which all allude to the ability to fill in the
gaps, incongruities, or disturbances that prevent one from seeing these overall structures
and use an innate need to do so as an important stimulus for learning. According to Ellis
(1938), other gestalt thinkers like Kofka and Kohler agreed with Wertheimer but
extended these ideas to looking at visual patterns. The laws of visual organization found
in these schools of gestalt thinking (i.e., proximity, similarity, simplicity, and, closure)
form a strong theoretical basis for visual media like film, video, and television to be able
to create its own unique sense of reality (Seward-Barry). The laws of visual organization

27
guide the senses to transform individual visual inputs into composite visual units made up
of real and imagined stimuli, forcing one to make inferences about them (Murch, 1973).
These views of perception lend credence to a belief that a standardized or
continuous view of the perceptual process can be formulated and forms the basis to the
theoretical approaches used in this study as they relate to an individuals ability to
remember and aggregate more complex meaning from video montage. Studies by gestalt
psychologists Richard Gregory and Semir Zeki into concepts, such as phi phenomenon
(processing movement as individual sequences separated by brief instances) and apparent
movement (the impression of movement from two stationary stimuli), reveal that there
might be two distinct systems of visual perception in the brain (Seward-Barry, 1997).
They also show how specialized areas of the visual cortex work together to create a
unified perception. One is responsible for perceiving movement between brief flashes and
the other takes care of perceiving movement between long flashes. These two concepts
provide further credence to the idea that the visual whole is made up of several
independent parts which, taken together, can furnish additional meaning, as afforded by
fast-cut montage found in new media technologies.
The visual process, as efficient as it appears to be, is not perfect (Seward-Barry,
1997). Much of gestalt theory is based on the law of pragnanz which stipulates that
psychological organization will always be good (i.e., simple, regular, symmetrical) as
the prevailing conditions will allow (Koffka, 1935/1963, p. 110). It is also the early
twentieth century counterpart to Aristotles concept of common sense that posited that
efficiency is achieved through simplicity, regularity, and symmetry (Seward-Barry,
1997, p. 47). These same theoretical concepts (proximity, similarity, simplicity, and,

28
especially, closure) explain how humans can often be visually fooled by slights of hand
or other artistic manipulations of visual stimuli, how they are able to see paintings
differently, or be taken in by virtual reality. The mind tends to close off partially
completed geometrical patterns, and/or create its own sense of reality based on what it
thinks it sees (Seward-Barry). These same four laws also explain how montage video
provides the continuity needed to be able to tell a story visually.
Fast SeeinR
Evidence points to the ability of humans to be able to process visual patterns very
rapidly (Coltheart, 1999; Tovee, 1998). Theory suggests that impulses processed in the
visual cortex are able to process almost 87% of perceived information within 400
milliseconds. According to Tovee, clinical researchers also found that this operation is
not linear. In fact, the majority of information (around 67%) is actually available within
the first 20-50 milliseconds of the initial spike train (p. 143). The investigation of what
is understood from rapidly presented sequences of visual stimuli began with the research
of Potter and Levy, in the late nineteen sixties and Forster in the early seventies
(Coltheart). Their studies suggest that most information encoded in an impulse is
available to the brain to act on at the very beginning moments of reception. Forster
developed a technique of rapid presentation of sequences termed RSVP (rapid serial
visual presentation). Potter and Levy provided a comprehensive review of RSVP
methodology, which proposed that humans possessed a natural short-term memory
system (Coltheart). These researchers confirmed their studies by placing constraints on
how long the visual system has to process stimulus in both forward and backward
masking functions in which the masking stimulus follows very quickly the onset of the

29
initial pattem/object stimulus. Through studies on rhesus monkeys, researchers have been
able to point out which cells within the brain are responsible for certain functions such as
pattern recognition. Once identified, these brain cell studies were translated to human
subjects, who were tested to determine how fast the brain is able to react, recognize, or
otherwise discriminate between faces of individuals. Researchers found that after only a
20-millisecond delay, test subjects could recognize and discriminate face identity,
indicating that the minimum amount of time that neurons need to be active in order to
mediate recognition and discrimination is in the 20-30 millisecond range (Tovee). The
next stage to processing this information is the brain reacting to these stimuli and
initiating reactionary motor commands. Independent follow-up research, based on
clinical tests performed on epileptic patients in preparation for brain surgery, indicates
that this process takes somewhere in the 150-200 millisecond range (Allison, et al.,
1994). These researchers confirmed that the whole process of pattern recognition and
reaction takes in the neighborhood of 400-500 milliseconds, indicating that, not only can
the visual system process information rapidly, it can also rapidly update and modify
responses, based on previous experiences (Tovee). For example, people can rapidly and
accurately reconstruct meaningful objects out of fragmentary or ambiguous evidence
(Ramachandran, 1994). These findings appear to denigrate traditional assumptions about
perception that awareness must mediate between stimulus and response. In fact,
researchers now know that many responses are automatic and emotional, and may bypass
cognitive processing all together (Seward-Barry, 1997). The fact that initial responses can
precede cognitive processing has caused a renewed interest in subliminal research, which
deals with subconscious message registration. However, results from research in this area

30
have been inconclusive due to the inability to date of researchers to quantitatively
delineate and measure perceptual responses (Seward-Barry).
Most previous research into limits on perception and recognition of rapidly
presented stimuli dealt with verbal information (Coltheart, 1999). However, Intraub
(1999) worked extensively with scene recognition and memory. Citing experiments of
Potter, she suggests that abstract representations of scenes are used in the interpretation of
pictures and scenes. Using a technique called conceptual masking, she was able to show
that the attentional demands of new and meaningful pictures interfere with ones ability
to continue processing their predecessors. Perhaps, sharing the gestaltist view of closure,
Coltheart further demonstrates this phenomenon, referring to it as boundary extension
(p. 6). Even in rapidly presented stimuli, subjects indicated that they tended to remember
information that was not shown but was likely to have been present outside the pictures
boundaries. Further, Intraub produced experiments that showed that photographs, when
shown in rapid succession, (at a rate that mimicked ones ability to scan them) were
unable to be recognized by their viewers, even a few moments later. These studies seem
to report limitations on memory, rather than perception, shown with Intraubs studies of a
series of rapidly presented pictures that were separated by blank screens of varying
lengths of time. It appears that the longer time lags between meaningful pictures
permitted viewers to properly process, categorize, and consolidate their meaning. Potter
argued (Intraub) that a picture is identified within about 100 milliseconds. Until the item
is identified, it is vulnerable to visual masking by a new visual event (p. 52). Once
identification is complete, however, the pictorial representation is maintained in short
term memory for a few hundred milliseconds, long enough for it to be consolidated and

31
encoded into memory. Intraub reported that the process of stabilizing integration errors
can take place when pictures are presented at a rate of about three to four pictures per
second. This is roughly equivalent to the presentation rate of the pictures presented to the
fast-paced group of the current study.
Encoding
Perception might have been conceived of as a totally individual operation was it
not for the fact that there exists some type of standardized analytical coding thesis on
which to base a more stable view of the process. In reality, humans process information
by breaking perceptual units down into sub-units called symbols and icons (Neuman &
Prinz, 1990). Corcoran (1981) proposed the idea that symbols and icons comprise the
very basic subset of cognition. He referred to them as the minimal units of perceptual
analysis or that product of semiotic analysis which is the smallest interchangeable
element that still retains the basic properties of the whole symbol system but cannot be
further subdivided without losing those properties (p. 118). He identified several forms
of icons and symbols, including (but not limited to) words, pictures, and auditory
patterns. He also referred to rhythmic patterns and control as symbols of perception that
can be translated to enhance learning but he did not limit his conceptualization to
audition. He also claimed that visual rhythmicity is a form of symbolic representation that
can be learned as a cueing system for attention and recall (Corcoran).
Humans process symbols and icons on more than one plane, depending on the
structure of the input. Words, pictures, sounds, etc. are perceived and treated in similar,
but distinct ways. Paivio (1986) proposed three different procedural methods:
representational, or the direct activation of verbal (i.e., text-based) or non-verbal (i.e.,

32
pictorial) representations, referential, the activation of the verbal system by non-verbal
system or vice-versa, and associative, the activation of representations within the same
verbal or non-verbal system. These differences in method of processing of stimuli
contrast in many ways. While researchers like Carr (1982) believe that a common
semantic system exists that provides similar representations for words and pictures,
others have suggested that differences in processing exist or that one form of symbol
takes precedence over the others. For example, the structural organization of a perceived
sequence can play an analogous part in its perceptual segmentation (Corcoran, 1981).
Corcoran (1981) also proposed that verbal and non-verbal forms of information are
processed at different rates. Still others (Archer, 1965; Gummerman & Gray, 1972; Huba
& Velletino, 1980; Rettenbach, 1999) posited the idea that increasing the speed in
perceptual encoding might be a learned activity that varies based on age and/or sex.
Perhaps the most significant debate with regards to the role perception has on
memory has taken place between those who believe in the superiority of verbal coding
over non-verbal, or vice versa. In their studies comparing the two, Smith and Haviland
(1972) introduced two complimentary assumptions. The influence hypothesis argued that
the perceptual superiority of words is due to a subjects opportunity to use the
redundancy in words for making perceptual inferences. The unitization hypothesis
proposed that words are more perceptible because there are fewer units to process in a
word than a non-word. Other researchers concluded that visual coding takes precedence.
Pezdek (1977) performed an experiment in which the sentence the bird is perched atop of
the tree was shown repeatedly at the same time as a picture of an eagle atop of a tree.
After time, the participants responded to a follow-up questionnaire that the sentence

33
actually was that an eagle is perched atop of a tree. In this case, visual perception
changed the actual perceived wording of the sentence. His studies were later confirmed in
additional studies performed by Gwen Nugent (1982) who also found that visual images
are recalled more readily than verbal texts. Her studies have been confirmed several times
by those evaluating the so-called Stroop effect where subjects incorrectly name a text
message by the actual font color of the text, rather than the color spelled out by the text
message (Shalev & Algom, 2000). Plass, Chun, and Mayer (1998) were less conclusive.
They developed a generative theory stated that learners actively selected relevant verbal
and visual information, organized the information into coherent mental representations,
and then integrated with one another these newly constructed visual and verbal
representations. Mayer, Bove, Bryman, Mars, and Tapangco, (1996) reinforced this idea
in studies of science curricula and found that there existed a coherence between the two,
so long as they complimented each other and that each was specific to their purpose.
Another way to look at the coding requirements between verbal and non-verbal
representations is that they are simply different from one another, with neither taking
precedence. In his studies, Singer (1980) found that verbal coding is a sequential process,
whereas the everyday sights and sounds of everyday life is a parallel operation that
encompasses a range of events within the same time. While its strengths make verbal
encoding more efficient for memory, they also limit its functionality in non-textual (i.e.,
non-reading) experiences.
Another comparison in the research of relative strengths or weaknesses of the
various encoding methods occurs between vision and audition; with neither being the
unanimous choice of researchers. Nazarro and Nazarro (1970) tried to equate aural and

34
visual stimuli. They suggest that short-term memory is based on auditory encoding even
when the stimuli are presented visually. They also found that learning (i.e., memory) is
reduced if the presentation relies mostly on visual inputs alone. Further, many researchers
(Cooper, 2000; Davis et al., 1999; Flannagan, 1998; Flowers, 1995; Moreno & Mayer,
2000; Sterrit, Camp, & Lipman, 1996) have shown how audio tracks can aide visual
encoding, especially where the fidelity of the visual display suffers.
Perhaps the most conclusive studies about the relatively stronger effects of visual
over aural stimulation have been offered by Gavriel Salomon (1979; 1994). He sites
many studies where visualization was more effective in recall, and concluded that visual
actions are often remembered over auditory have been "primarily due to the salience over
action" (Salomon and Cohen, 1977, p. 29), especially moving images. Further, he cites
some investigations in which visual tracks were successfully added to supplement the
auditory. He concluded that there probably is no visual dominance in children but since
actions are generally more memorable, visual information is normally more likely to be
recalled. However, he also cautioned that adding pictures to an already well-formed
auditory presentation may provide some distraction (Salomon and Cohen, p. 29).
Researchers have shown that words and pictures are processed on opposite sides
of the brain and follow different psychological laws (Ederlyi, 1985). This fact had
originally deprecated most traditional ideas about subliminal communication being an
effective means of communication and learning. Most of the research done in subliminal
perception had been done with verbal stimuli, based on the conventional assumptions
about cognition having to precede response. Erdelyis and LeDouxs (1991) more recent
studies using visual stimuli alone have produced more pronounced results. Although their

35
work found that the signal is sufficient to cause an emotional response, it wasnt strong
enough to reach conscious thought, and that the emotional responses are autonomous and
primal, similar to those that humans, have needed for survival throughout their
evolutionary process (Dixon, 1981). This inherent automatic preparation of their response
to stimuli possibly explains why humans might likely be unconsciously biased towards
accepting or rejecting ideas, and occasionally acting on them. It is also where researchers
believe the most significant contributions to research into subliminal message processing
might lie (Seward-Barry, 1997).
Relationship of Memory and Learning
Studies involving learning and memory date back more than a century (Crowder,
1976). Beginning with the Ebbinhaus preliminary studies into brain associations during
the late nineteenth century, the psychology of learning principles evolved by the mid
twentieth century into Kohlers framework of analyses that addressed three processes
associated with learning and memory: stage analysis, coding analysis, and task analysis.
While learning may actually encompass many more of these analyses processes (such as
habituation, acquisition strategies, operant training, etc.) all of which are significant
when taken in their entirety they are beyond the scope of the current study. This study
deals with simple recognition and comprehension assessed in stage analysis, an essential
pre-condition to learning, and certain aspects of coding analysis, a subset of the memory
process that has been found to be an adequate premise on which to base assumptions as to
whether learning may be taking place (Estes, 1994). Estes related the ability of
individuals to complete the process of recalling information from memory to their ability
to classify or categorize that information in a meaningful way. Under this definition, the

36
term categorization shares some meaning with the term perception. However,
categorizing also implies a deeper process than simply detecting that some unit of
information has been perceived. It also means that other related characteristics have been
captured (i.e., location, color, or shape, etc.) to apply meaningfulness to them that aides
the process of recalling that information later. Under this definition, immediate memory
is looked upon as an important but insufficient prerequisite for learning. The current
study utilized a cued recall performance test in order to assess the ability of the treatment
to create (and the subjects to immediately remember) an initial memory trace.
Textual versus Pictorial Memory
Although much of the research into memory cited in this review deals with textual
information (Crowder, 1976; Estes, 1994), a great deal of it has direct relevance to the
current study. In his research into memory capacity, Sperling (1963) was able to
determine that his subjects were able to attend to visually perceived textual characters at
a rate of about 100 per second. These same subjects were only able to name (categorize)
these characters at a rate of six per second. This discovery led Sperling to assume that
initial information-gathering and naming (categorizing) may not be the same process, and
that the task of recalling utilizes different resources and techniques for pictorial than for
textual information. Durso and Johnson (1979) found that pictures benefit more when the
tactic used for elaborative processing is to simply name the picture whereas words aid
more when subjects perform some task to categorize them. Gaining an understanding of
these differences between the textual and pictorial memory processes has served two
purposes in the current study. First, it helped to emphasize the role pictorial perception
can play in the learning process. Second, it reinforced the need to assess pictorial memory

37
with like stimuli, that is, to use pictures as targets in the test questions as much as
possible because textual questions and answer choices introduces an additional
translation process.
Recognition and Recall
Over the years, there has been a significant amount of confusion regarding the use
of the terms recognition and recall as it relates to measuring how much learning has taken
place (Sheppard, 1967). While very similar in nature, recognition and recall are not
synonymous. Their difference in meaning suggests that we have different ways of
retrieval once information has been stored. Studies have shown that recognition memory
resides on the right side of the brain whereas recall (especially verbal) takes place on the
left. Recognition implies a simpler intellectual task that does not necessarily require a
significant amount of encoding and carries a longer-term residual. Sheppard describes
studies in which even a quick glance was enough to assure some long-term recognition
value. He showed objects from a Sears Roebuck catalog to Guatemalan peasants who had
no personal knowledge of the items. They were able to recognize the pieces even several
months afterwards. Singer (1980) drew several additional conclusions about the
differences between recognition and recall. He suggested that the human brain is capable
of storing a tremendous amount of visual material much more than information received
in other forms.
The ability of individuals to store so much visual information more rapidly is
perhaps due to the holistic or gestalt qualities of the right brain where visual imagery is
processed. Thus, material seen just a few times (i.e., presented on television) without any
significant effort at learning it, can be recognized if it is re-presented later. The greatest

38
limitation of assessing learning based on recognition alone includes the fact that it is not a
useful measurement when the responder doesn't have the object in front of him or her for
identification. Perhaps the most significant contribution to the learning process that
recognition plays relates to what Singer (1980) called the differential affect. In short,
differential affect is an original startle that occurs when new and novel information is
first presented. The brain processes this information on the right side of the brain in the
same way it processes information to be later used in recognition activities. Singer noted
that the startle effect takes some getting used to, but once the initial startle is over, the
brain is better able to process subsequent information it receives. Theories surrounding
the startle effect suggest that increasing recognition from stimuli used in the quick-cut
imagery in todays new media is both a trainable act and is sufficient to gain viewers
attention so that they may recognize it again during future stimuli engagements. As such,
it helps to formulate a theoretical basis for certain hypotheses in the current study as well
as suggestions for potential future research found in Chapter 5.
Another assertion about the differences between recognition and recall is that they
are basically the same with the exception that the former may be a somewhat easier
process because it involves a lower capacity threshold requirement (Crowder, 1976).
Crowder found that this idea about recognition and recall has been discredited by some
due to its tie to a concept that recognition and recall are basically the same processes.
Crowder also reports that researchers have more recently looked upon the two as linked
but separate operations. Recall involves the generation of an additional process that can
be tested separately. For example, a variable might improve recall while, at the same
time, damage recognition. Additionally, there is evidence in experiments involving

39
intentional versus incidental learning that shows that recall improves when the subjects
are warned of an ultimate memory test, where recognition does not change significantly
and may even decrease. According to Estes (1994), recognition, might be better described
as two distinct processes: absolute and recognition in context. The former relates to
whether a subject recognizes a particular person, or he or she remembers ever
encountering that person before. The current study attempts to elaborate on Estes work,
comparing the ability to remember verbatim facts with subjects ability to remember
contextual (i.e., gist) sequences. Estes further suggests that recognition does not in
general, provide a direct window to memory (p. 230). Rather, it may be said that
recognition provides the best available basis for estimating memory storage, as long as
interpretations of data are guided by appropriate models (p.230). It is the problems
associated with these model interpretations where much of the on-going controversy
about the recognition-recall debate appears to reside.
Verbatim versus Gist Memory
The apparently unresolved confusion with regards to the use of the terms
recognition and recall led to the need to find another means to assess immediate memory
in the current study. A further review of the literature led to a discovery of relatively new
research on two complementary memory processes that have proven useful to the current
study. The foundation of these newer ideas about describing memory in terms of a dual
relationship in a bi-polar representational system lies in Brainerd and Reynas (1990)
fuzzy-trace theory. Fuzzy trace is consistent with many of the ideas about childhood
memories but runs counter to Piagets developmental/structuralist approach that uses the
mind as a computer comparison (Miller & Bjorkland, 1998). Brainerd and Reynas

40
method is based on a metaphor of intuitionalism, in which people prefer to think, reason,
and remember using inexact, gist-like traces rather than more precise (but also more
forgettable) verbatim traces. The concept traces memory preferences through age and
personality differences, with younger children preferring verbatim, exact references,
which become fuzzier as they grow older. Miller and Bjorkland describe fuzzy-trace
theory as a hypothesis about representation, which posits that children of different ages
are disposed to use different types of representations that are available to them. What
differs between this theory and other more mainstream conceptualizations is the
propensity to use different types of representations to solve problems, making it more
aligned with Piaget-like concepts of constructivism. According to Miller and Bjorkland,
earlier misconceptions about fuzzy-trace theory caused confusion as to how it differed
from other already established conventional concepts like schema theory. Gradually,
however, fuzzy-trace theory began to gain acceptance with mainstream theorists,
especially when it was applied to issues of age differences in suggestibility and false
memory creation (Miller & Bjorkland, p. 188). It was later adopted outside of the
immediate Brainerd and Reynas [sic] sphere of influence and began to be used to
explain basic and applied phenomena on a wide range of tasks (Miller and Bjorkland,
p. 188).
Independently, Brainerd and Gordon (1994) and Reyna and Kieman (1994) began
to evolve their jointly-developed theories on to other memory domains, concentrating on
the specific differences between gist and verbatim memory traces. The concept has been
subsequently been adopted by others (Cowan, 1998; Tse et al., 1999). The verbatim-gist
classification has more recently been used in a variety of ways, including to

41
counterbalance the more traditional ideas about encoding, and the verbal versus
nonverbal comparisons, and more importantly for this study, in analyses of the ability of
subjects to recognize and utilize information perceived in visual presentations (Haber,
1970; Loftus, 1979).
According to fuzzy-trace theory (Brainerd & Reyna, 1990), both gist and
verbatim memories are based on similar inputs, but are stored separately. Verbatim
memory may be associated with higher ability students, but some researchers have shown
that reasoning power can often be disassociated with strict recall of specific details.
Others have indicated that gist determination is associated with reasoning powers,
making it higher-level activity events (Reyna & Kieman, 1994). It follows then that gist
memory can be assessed independently without requiring that the subjects also score well
at verbatim memory and vice versa. Reyna and Kieman did point out that the only
exception to the verbatim-gist independence rule is the fact that success with verbatim
recall has been shown to be higher with linear text-based presentations as opposed to
pictures.
Although it is a relatively new theory, fuzzy trace serves to explain an important
concept of recognition and recall from non-verbal displays as used in the present study.
As Brainerd and Gordon (1994) pointed out, so little is known about the developmental
and functional relationships between verbatim and gist memories that it is necessary for
researchers to proceed step by step, investigating these relationships within delimited
classes of inputs in the hope that general conclusions will ultimately emerge (p. 163).
This gives rise to questions that investigate the effect cognitive style and presentation
speed might have on these two forms of memory determination. It is the intent of the

42
current study to address this challenge and extend their analyses into the pictorial
memory domain.
The Effect of Arousal on Memory
Another, related topic on memory is the concept of arousal and an attempt to
clarify some of the differences of opinion with regards to its perceived positive and/or
negative effects on memory. Many of the research findings with respect to arousal and
memory appear to be based on the Yerkes-Dodson law (Yerkes & Dodson, 1908) which
predicted an inverted U-shaped relationship between arousal and performance. A certain
amount of arousal can be a motivator toward change (with change equating to learning),
whereas too much or too little can work against the learner. The Yerkes-Dodson law
appears to be the basis for much of the early controversies surrounding arousal in
mediated messages as to whether they impose a negative or positive effect their viewers
(Lang & Basil, 1998). Neil Postman (1986) asserted that the presentation speed of media
messages is a force that increases arousal to the point where most cognitive activity is
negated. He relegated television viewing to pure mindless entertainment with little or no
educative value. On the other hand, Zillman (1991) postulated that arousal can be a
unifying force that intensifies motivated behavior. Further, arousal plays a significant role
in many motivational models, like the ARCS model developed by John Keller (1983).
The A in this acronym stands for gaining the learners attention, which is handled
through arousing or curiosity-seeking cognitive engagement.
The interactions between arousal, motivation, and attention and their affect on
learning have gained considerable focus in educational research in recent years. Posner

43
(1982) described four basic ideas about attention. Each is listed below with parenthetical
comments regarding relevance to the current study:
1. Mental operations related to attention take time to perform and the amount
of time required will correspond to fixed qualities of the stimuli (such as
presentation speed) and the discretionary strategies of individuals (such as
cognitive style).
2. Mental events occurring closely in time are processed successively.
(Montage video attempts to replicate these closely related mental events.)
3. Internal events can be studied by observing the amount of facilitation or
inhibition they cause and this process is hierarchical. (Immediate memory
feeds longer term recollection.)
4. Attentional processing favors stimulus change. An orienting reflex biases
people toward fresh or novel sources of stimulation. (Presentation speed
can act like an orienting reflex.)
The works of Reeves et al. (1986), Reeves and Geiger (1994), and Reeves and
Nass (1996) have centered on the individuality of the attentional reflex. They clearly
defined attention as "a psychological cognitive process that varies within individuals over
time" (Reeves et al., p. 254). These more modem theories look at attention as more of a
filtering device that is not exactly an all or nothing gateway. Further, there might be
different types of attention requiring different types of responses. For example, vigilance
tasks require people to wait for an event and then respond quickly. They decrease over
time and require individuals to attend to only one at a time. These are primitive orienting
cues that stem from primal man. Attentional preparation tasks give people cues or primes
before stimulus occurs. Perceptual intrusion tasks are those so arranged that people
cannot avoid attending to them. Finally, attention switching tasks are those in which
subjects must switch their attention from one task to another. Reeves et al. elaborated
further. They posited an Orienting Response Theory (OR) that suggests, in part, that

44
attention is subject to habituation that explains, for example, how people become
accustomed to living in cities and with noises and after a time becomes accustomed to
them to the point of eventually being able to block them out without loss of focus on
other tasks at hand. OR Theory might explain how youths of today might have become
accustomed to fast-paced media so that they can able to learn from it more efficiently
than their elders who are generally less exposed to these new media techniques.
These new ideas about arousal also encompass the concept that attention (and
subsequent arousal) in humans involves mechanisms that are used differently depending
on the task to be performed. Lang et al. (1999) differentiated between cortical and
emotional arousal in which the former can cause the body to bring more attention to a
situation rather than less. However, even cortical arousal may have its limits. In
reviewing Zillmans limited capacity model, Lang proffered four related principles:
1. A viewer will allocate an overall level of processing resources to the
complete viewing task based on goals, interests, etc.
2. A viewer's goals influence the proportion of resources allocated to the
various sub-processes (such as storage & retrieval).
3. The structural and content features of the message elicit orienting behavior
and the automatic allocation of resources to encoding.
4 The content and structural attributes of a message can also elicit arousal,
which results in the automatic allocation of resources to encoding and to
storage.
Zillman (1991) admitted that when there are insufficient resources available (i.e.,
cognitive overload) to carry out all the sub-processes, some aspects of cognitive
processing will be performed less well. This idea of finding optimum levels of arousal to
enhance motivation and performance was further developed by Winifred Hill (1985) who
developed a series of laws and theories that included several about stimulating students

45
interest through arousing content. Among her assumptions were that optimal arousal
involves three factors: novelty, complexity, and intensity. The level of arousal (i.e.,
attentional demands) is function of an inter-reaction among these three factors. Highly
intensive and novel stimuli are more arousing than those that are weaker or more
familiar. Moderate intensity levels of arousal are more reinforcing than either high or low
levels of either the newness or novelty of content. According to Lang et al. (1999), it is
important to find a consistency and/or balance between the level of external arousal
applied by the presentation vehicle and any arousal that might be inherent in the content.
If task or conceptual complexity outweighs its novelty, a more moderate stimulus would
be more reinforcing for cognition. If, on the other hand, intensity and complexity are
relatively equivalent in strength, a change in content (or novel form of content) is what
will regain a viewers attention. Hill summarized by stating that a highly novel stimulus
will [sic] be more reinforcing if it is mild and simple, whereas among very familiar
stimuli the more intense and complex ones [sic] tend to be preferred (p. 176).
Lang et al. (1999) and Lang et al. (2000) elaborated further on the correlation
between intensity and complexity of content, relating the former to presentation speed and
the number of structural features occurring in a message. Again, referring to the limited
capacity model she suggested that, as the pacing in a message increases, the number of
orienting responses (ORs) elicited by that message (and the amount of information
available to be encoded) will also increase. Because of these increased ORs, a viewer will
automatically allocate more resources to encoding fast paced messages. She also
proposed that viewers control some aspects of the resource allocation process by making
decisions such as whether to watch, how carefully to watch, how hard to try based on

46
how interesting the subject is, how relevant the information is, or simply whether the they
want to remember it at all. Furthermore, Lang proposed that the medium itself can control
some aspect of the automatic allocation of processing resources by also eliciting it own
orienting responses (ORs) from the viewer. The prime example Lang used in her research
was an increase in presentation speed. Lang found that these ORs are automatic,
reflexive, and attentional responses to changes in the environment or to stimuli, and
people have learned that they signal important information. She also found that this
automatic allocation of resources is a relatively short-tem response, occurring over
seconds. She concluded limited, elicited arousal results in the automatic allocation of
resources to both encoding and storage.
Kahneman (1973) proposed a variable capacity model that predicts that arousal
elicited by a message will increase one's overall pool of available cognitive resources.
While he failed to specify some type of upper limit, he did suggest that secondary
reaction time might slow down as presentation speed increases, especially if content is
also arousing. Conversely, as speed increases for messages with calm content, rapidity
alone might be sufficient to increase the size of the cognitive pool allocated to respond to
the message. Basil (1994) proposed a fixed capacity model that predicts that higher
resources may be committed to the message but the overall size of the cognitive pool
remains constant. Overall, arousing messages might cause slower reaction times than
calm messages if the presentation speed also increases because resources are
automatically allocated to the encoding sub-processes as the result of orienting response.
Conversely, if more resources are committed to creating a faster response, resources
available for encoding will be reduced.

47
Sheingolds (1973) work further complicates the ability to unwind these
contrasting views of resource allocation by showing that the age and sex of the viewers
tend to confound things. Lang et al. (2000) later tried to unravel these variables by
eliminating content as one of them. Using a measuring device that calculated skin
conductance and heart rate, she gauged secondary task reaction times during several
different studies and came to the conclusion that the limited capacity model most
accurately reflects what occurs (at least in a casual viewing environment) with the
following two hypotheses:
1. Arousing content is likely to elicit feelings of arousal and measurable
sympathetic nervous system activation. As pacing increases, viewers
encode more and, therefore, recognize more of specific content.
2. Arousal results in the automatic allocation of processing resources to
encoding and storage.
Langs conclusions suggest that media producers who want their messages to be
remembered should create arousing messages that are presented slowly or at medium
speed, or calm messages that are presented at medium or fast speed. These producers
should not create messages that are at the same time calm and slow or arousing and fast.
While Langs work has been a watershed for casual television viewing, questions remain
as to how much of their work relates to student viewers and in an educational
environment where content might simply be considered boring. In spite of these
reservations, the limited capacity model was used to develop a portion of the hypotheses
for the current study.
The Hueristics of Pictures
To the extent that processing of viewed information is controlled by highly
varying individual past experiences, extracting information also varies greatly. While

48
there have been a few well-elaborated schema theories that demonstrate specific
influences of the effect of showing pictures episodically in the absence of personal or
semantic interpretations, a number of independent investigations of perceptions formed
on the basis of televised news reports demonstrate convincingly that there exists a high
degree of correspondence between subject interpretation and recall (Zillman & Brosius,
2000, p. 38). Value judgments also play a role information processing. Several studies
into news viewing (Bums, 1992; Hayes-Roth & Hayes-Roth, 1977) also have shown that
pictorial representations have a positive effect on attention and processing. Likewise, the
value of increasing the number of occurrences of pictorial displays (as opposed to simply
reiterating a fact verbally) has been demonstrated in a number of studies. Zillman and
Brosius showed that an increased number of occurrences of a pictorial message can also
influence perception in the news. For example, in their studies, viewers concept of how
much significance or importance they should appropriate to a story seemed to rest in
direct proportion to the number of times it was shown on television, to the point that
where subjects applied more significance to the story than it actually warranted. Zillman
and Brosius pointed the fact that repeated visual representations tended to slant the news,
suggesting that there was strong evidence that the sheer number of concrete, visible
sources relating their experiences and concerns does exert an influence on issue
perception (p. 110). They further suggested that photographs used in print media and on
televised newscasts, regardless of how innocuous the content, are remembered longer and
the image that was more likely to be retained was of the visualization, rather than the
textual message that accompanied it.

49
In conveying any mediated information, it has long been deemed important to
separate the flow into manageable chunks and to isolate and focus on some events at the
expense of attention to occurrences in between (Kozina, 1986). According to Zillman and
Brosius (2000), a mediated narrative (even an instructional message) broken down into
exemplifying chunks may jump from event to event, irrespective of the events proximity
in location or time, so long as it shares enough similarity that makes it sufficient to
provide reliable information about other events in the group (or the group itself). A
picture or series of pictures, then, may become a form of an exemplar to be used to aid
recall and recognition as long as they formulate a schema for their viewers. While it is
has been shown that learners remember more when they attach personal meaning to an
idea or concept (Brown & Kulick, 1977; Tulving, 1972), research into casual television
viewing has also demonstrated that viewer recognition and recall for content of news
shows (Zillman & Brosius), music videos (Hitchon et al., 1994), and/or advertisements
(Hill & Lang, 1993) can be modified, either through a form of exemplification of the
information by means of lashing loosely related content or structure in associated
pictorial representations. All three of these studies, taken together, show that by using a
combination of stratifying content into meaningful and related chunks and employing
certain cueing strategies can overcome some of the memory-related limitations imposed
by presenting unfamiliar content. Mayer et al. (1996) suggest that attention and memory
for pictures (especially moving pictures) can be controlled by the use of formal
exemplifying features such as zooms, presentation speed, sound effects, and music.
Likewise, the format of the instrument used in the current study employs many of these
visual exemplification techniques (such as similar coloration, zooming, and differing the

50
length of time images are displayed) to stratify presented information into manageable
chunks to aid in the recognition and recall of its viewers.
Pacing versus Presentation Speed
When referring to the speed of presentation, non-education researchers often use
the terms, rhythmicity, tempo and pacing, interchangeably. However, their definitions do
not always equate. Educational multimedia developers often refer to pacing as the rate
that a learner proceeds from one place to the next, as in self-paced instruction (Canelos,
1986; Kozma, 1986). In this connotation, pacing is more associated with interactive
video, where interaction also implies some degree of learner control of the rate and
sequence of the events that take place. On the other hand, in broadcast or instructional
video, where the events are presented to viewers at a pre-defined rate, pacing may be
better defined as the rate of information presentation (Kozma, p. 14). With this latter
definition, pacing is more associated with rhythmicity or tempo. It is within the latter
context that the term pacing is referred to in the current study. In recognition of the fact
that, in educational circles, the term pacing has long been associated with Canelos
connotation, the term presentation speed is being used exclusively in the current study,
even though pacing was the operative term used most often in the literature, especially in
studies conducted in non-educational settings.
There is one point in the research where both casual viewing and educational
researchers appear to agree: that changing the speed in, and/or adding musicality or
rhythmicity to a pictorial presentation adds to memory for its content (Flowers, 1995;
Hitchon et al., 1994; Luckett, 1996; Patel, Peretz, Tramo, & Labreque, 1998; Shaffer,
Greenspan, Tuchman, Cassily, Jacokes, & Stemmer, 2000; Wagely, 1978). There are

51
also indications (Corcoran, 1981) that rhythmic control can be learned and translated as a
coding structure in its own right to enhance learning and can be made to act as a cueing
system. Zillman (1991) found that rapidly presented visual programs foster superior
attention, and potentially superior learning. While he admitted that much of the successes
of faster speeds may be attributable to arousal, he also quoted studies that showed that the
creation of transient alertness, even by primitive means such as the frequent instigation
of the orienting reflex, tends to facilitate information acquisition in audiences for which
high levels of attentiveness cannot be (otherwise) expected (p. 126). Apparently,
alternating the speed of an instructional message tends to increasingly arouse the cortex,
which, in turn, as many researchers have suggested, tends to make a viewer more alert
(Lang et al., 1999; Pearl, 1982; Reeves et al., 1986; Zillman, 1991). Nelson (1990)
quoted studies that seem to indicate that rhythm serves both organizational and
expressive functions that can also be shown to significantly help to organize thinking and
cognition. While investigating the effects of training using tachistoscopes Peck (1987)
found that by increasing the speed of a lesson slightly, it may be possible to increase
attention and, by doing so, also increase learning. Peck also found that the effects of
altering the speed are more pronounced for slow than for fast readers, something later
confirmed by Walters (1983). In fact, a review of the literature into the use of
tachistoscopes in remedial and rapid reading training programs (Dick, 1973; Long, 1982;
Schale, 1971; Sheingold, 1973; Woodley, 1984) appears to confirm the positive effect a
more rapid presentation speed can have on increasing attention and cognition, especially
when the content includes redundant clues and it represents conceptual information.
These rapid reading training techniques have been shown to also increase visual

52
awareness of larger spans of information, and processing rates can be accelerated without
loss of comprehension.
Zillman (1991) showed that the tempo of a message can be influenced by more
than its structural or production process. Message content, for example, may have a
carry-over effect. He showed that the interspersion of fast-paced humor is often mediated
by cortical arousal. Semi-attentive children in the audience may be made alert by the
inserted humorous tidbits, and this alertness may have carried over into exposure to the
immediately subsequent educational material (Zillman, p. 126). These content studies
were not limited to humor. Other studies looked at attentional reflexes caused by
fireworks (Reeves & Nass, 1996; Zillman, 1991), and negative or violent content (Pearl,
1982; Reeves et al., 1986; Reeves & Nass, 1996). Hill and Lang (1993) used a
combination of image complexity and the number of structural or formal features in a
message to manipulate effects of pacing. Other studies looked into the interaction
between pacing and content. In his studies into the levels of arousal found in fast cuts in
music videos, Hitchon et al. (1994) found that fast-paced video montage did not
contribute negatively to the complexity or ambiguity issues as long as the story line was
not confusing or ambiguous. Theoretically speaking, if increasing the speed of message
delivery does not necessarily cause attentional overload, one might be able to make a
stronger case by looking at a corollary question (i.e., does the slowing of the pace that an
image is presented necessarily increase its retention?). In video productions, presentation
speed has a direct relationship to the number of frames an image appears, with 30 frames
per second being the standard (Stephens, 1996). Potter and Levy (Keller, 1976) and
others (Goodglass, 1971) found a significant correlation between retention and duration

53
of images displayed for 200 to 300 and milliseconds, which equates to approximately 1/3
of a second, and those displayed for more than a second. Optimal retention occurred
between 200-300 milliseconds and one to two seconds. Conversely, in other studies, it
was found that images displayed for more than one to two seconds actually caused a
decrease in cognition due to loss of attention (Thorson & Lang, 1992). It might be that
duration becomes a variable of lesser import as the size of the learning set increases and
the number of categories decreases (Keller, p.82). In short, this research appears to
indicate that longer displays do not necessarily signify a relative increase in recall, and
can actually cause degeneration. In the current study, images appear on screen for as little
as 1/3 of a second to a maximum of one second, with the majority appearing for less than
1/2 of a second, well within the range indicated by Potter and Levy (Keller) and
Goodglass.
How Learner Attributes Affect What is Learned
Cognitive style
The concept of cognitive style has endured for more than fifty years. It came out
of the New Look movement in perception that was bom during a symposium sponsored
by the American Psychologist Association held in New York in 1949 (Witkin &
Goodenough, 1981). Participants in the New Look movement were a loose confederation
of psychologists who became critical of the dominant approaches to perception then in
vogue. Their main criticism was that most current approaches tended to ignore the person
doing the perceiving. Out of that symposium came a flood of new studies aimed at
looking at the personality traits of individuals (i.e., individual differences) during the

54
process of learning. Out of this broad context of research emerged the concept of
cognitive styles (Witkin & Goodenough).
Guilford defined cognitive style as that which conceptualizes intelligence as having a
process dimension (Green, 1985, p. 2). According to this view, learning is not merely an
automatic reaction to a stimulus but a set of operational steps that varies, depending on
individual proclivities. Brumby (1982) asserted the following assumptions regarding
cognitive style:
Ones cognitive style is singular (i.e., an individual has only one) and can
be measured on a bi-polar scale.
While an individual possesses one style, others may be present in varying
degrees.
Individuals can select an appropriate style appropriate to the task at hand.
Earl Messick (1970), a charter member of the 1949 New Look movement,
catalogued nine dimensions of cognitive style, covering research that he and his
colleagues performed, and that of others who came along in the years that followed. His
list included scanning, breadth of categorization, conceptualizing style, levelers versus
sharpeners, distractibility, tolerance for unrealistic experiences, cognitively complex
versus simple, field-dependent versus independent, and impulsive versus reflective. Of
the nine mentioned, meta-research (Green, 1985) has shown the latter two to be the most
commonly accepted as credible sources for investigating how individuals perceive and
process visual patterns. Of these two, impulsive-reflective was determined to be the most
relevant to the current study.
Impulsive Reflective Style
The impulsive-reflective scale was borne out of Jerome Kagans research work

55
(1965) with cognitive tempo (i.e., rate of cognition) in younger aged children in which
their speed and attention to detail was indexed. Children deemed to be impulsive tend to
react very quickly and make quick decisions (i.e., they select the first answer that occurs
even though it may be wrong), while reflectives tend to take more time to consider
various options but are also generally more accurate with their interpretations. Contrary
to many stereotypes about bright children thinking quickly, Kagan (1966) found that
neither tendencies for fast or slow decision times were significantly related to verbal
ability or innate intelligence. According to Block, Block, and Harrington, (1974) some
have criticized Kagans findings, noting that although response times were positively
correlated with performance, overall, the correlations were often quite low (ranging from
close to zero to around .45). However, according to others (Ayabe, 1973; Bridgeman,
1980) theirs and others research appeared to ignore the converse of the measurement
scale that compares slow-accurate to fast-accurate, which states that accuracy and not
speed counts the most in complex problem-solving situations. Bridgeman points out that,
even with timed tests, the negative correlation between speed and accuracy can also be a
plus, suggesting that there are many children who are fast and accurate (p .212). The
problem, then, appears to be taking an extremely bi-polar classification and make
subjective determinations in overall psychological and/or personality evaluations that
may require less polarization and more sub-categorization. These classifications (or close
derivations thereof) have been used quite successfully and deemed to be valid
classifications in specific educational evaluations. Campbell and Davis (1982) found that
the reflection-impulsivity style construct emerges as an ecologically valid and
parsimonious descriptor of a component of student behavior (p. 8) where it is

56
determined, for example, to hinder learning performance. Conversely, Boyden and Gilpin
(1978) found latency and error rates to be independent of measures of distractibility. The
relationship between impulsivity and academic achievement is not necessarily tied to
aptitude or intelligence but to ones ability to attenuate and/or process specific types of
inputs (Kogan, 1971; Leino, 1981; Messer, 1970). Cooper (1982) suggested that
differences in processing speeds may well be accurate indicators of ones ability to
process mostly visual information. Ridberg et al. (1970) found that cognitive style is
predictive of performance in a variety of measurement tools, including those for reading
recognition, secondary learning, and reasoning. Hedberg and McNamara (1985) found
that when using visual information the tendency to reflection or impulsivity might be an
important predictor of performance, particularly in relation to time and error under
conditions of response uncertainly and time pressure. Merrinboer (1990) was able to use
the classification as a predictor of academic performance and use it to pre-arrange
feedback strategies to increase effective computer usage, especially in younger students.
It appears as though impulsivity is a characteristic that mitigates with age. Wright
(1979) found that impulsivity lessens over time as one grows into adulthood. Okun et al.
(1979) found that adults have significantly higher latencies than middle-aged and
younger adults. In other words, as one grows older an individual tends to become more
logical in his or her outlook, incorporating familiarization and planning into perceptual
processes. However, an increased ability to exert a reflective response to stimuli can also
work in reverse. OBrien (1968) discussed effects of becoming stimulus bound,
especially when exposed to substandard information. This phenomenon, which occurs
mostly in adults, causes individuals to so over-structure their internal response that they

57
were unable to subsequently correctly perceive a very clear and undistorted image.
Although adults tend to be more reflective in response to stimuli, these same reflective
tendencies could work against them when comparing the correctness of their responses to
younger individuals, especially if the younger person participates more often in visually
oriented past times. There appears to be a crossover of the effectiveness of ones
increased developmental reflective abilities and ones ability to effectively develop a
corresponding facility to realize perceptual closure from visual data that is either less than
highest in fidelity and/or is displayed at a very rapid rate.
Anderson and Revelle (1994) looked at the effect changes in daily arousal rhythm
patterns play in causing similar alterations in impulsivity-reflectivity. In this context,
arousal is meant to portray the processes that mediate non-specific alertness, or liveliness.
Impulsives tend to demonstrate high alertness and sense of arousal, a key element in
learning preparedness. Highly impulsive children have also been linked to certain
attention deficit syndromes, as portrayed by their lacking in the ability to sustain longer
periods of attention (Anderson & Revelle). Anderson and Revelles research indicates
that the impulsivity rates vary (i.e., are more prominent) by the time of day, with
impulsivity in those individuals with impulsive tendencies being more pronounced in the
morning hours.
Using impulsivity-reflection to categorize individual differences has not been
without a certain amount of controversy. While the impulsive-reflective scale can be very
beneficial in predicting certain academic outcomes and has significance in timed tests,
there are shortcomings in making a completely accurate independent identification with
regards to whether a student fits this classification in all contexts and under which

58
conditions (Bridgeman, 1980). For sure, the impulsivity label was considered powerful
by some (Kagan, 1965; Kagan, 1966) and appealing by others (Hedberg & McNamara,
1985). However, some researchers (Ayabe, 1973; Bridgeman, 1980; Campbell & Davis,
1982) found that some weakness exists in the outcome of using classification as it relates
to discriminant validity and generalizability. In other words, there appears that some
evidence of loss of power, due to a corresponding loss of independence of traits in the
subjects tested in certain studies. The criterion used to justify discriminate validity in the
identification of a classification method or methods, which measure a discrete trait to the
exclusion of others (Campbell & Fiske, 1959). An additional problem in using this
classification is that it appears to be more heavily weighed down by the errors subjects
made than latency itself (Ayabe; Block et al., 1974). Further, there is an indication that
many studies failed to yield consistent findings in an ecological (i.e., classroom) setting
due to inconsistencies with the exact definition of impulsivity (Campbell & Davis).
Lastly, previous studies systematically excluded females. Some researchers went
so far as to suggest that gender differences were so extreme that they would skew the
results (Caimes & Cammock, 1984). The current study looked at gender differences to
see if the same rationale is still justified.
Factors From Film and Television That May Affect Learning
Preconceived Mental Demands
Salomons (1984) with mental effort appears to confirm the idea that learning that
is or is not realized by a medium can be influenced by ones previous experiences with
that medium by coloring student preconceptions about the mental demands being placed
on them by that medium. In his studies, students reported that they exerted more effort

59
towards reading than with television, but they rated the latter as more realistic and more
efficient. Greater mental effort requirements attributed to reading seemed to result in
greater perceived demand (PDC) towards that medium. Conversely, as these same
students perceived less mental effort was required towards television, they exerted a
correspondingly less effort and paid less attention. The amount of effort exerted was also
in direct proportion to preconceived notions about ones perceived self-efficacy about
either medium. The more comfortable they were with the medium, they less effort was
exerted. While Salomons studies showed some overall weakness due to some
discrepancies between self-reports and actual performance, his studies were backed up by
several others. For example, Ksobiech (1976) reported that pre-conceived task demands
affected the proportion of time students requested to see the visual portion of a video
presentation or only the audio portion, based on whether they were told that they were
going to be examined on the material, or were only asked to observe it for evaluation.
Students sought the video or audio source that maximized the purpose for which they
were to receive the material. Krendl and Watkins (1983) found that students viewing
television for entertainment value differed in the number of items they were able to recall
from it from those who were asked to view a presentation to learn from it.
Accordingly, some researchers (Langer, Blank, & Chanowitz, 1978; Langer &
Imber, 1979; Schank & Abelson, 1977) hold the view that most encounters with
television cause viewers to generally approach it with a sort of mindlessness that can
undermine its role in an educational setting. It was very important for the success of the
current study that the subjects do not consider the activity of viewing the video and
subsequently answering the questions that followed as an entertainment activity,

60
especially because researchers have shown that currently most of younger viewers day-
today encounters with both computers and videos (in particular music videos) are for
amusement purposes (Seward-Barry, 1997).
These ideas about preconceived demands relate directly to the procedures used in
the current study. Care was taken while administering the test to 1) demonstrate the
purpose of the viewing activity using a sample portion of the video and 2) instruct the
subjects to treat the activity just as if they were to be graded on it, even though no
intention to do so was introduced to them beforehand.
Exemplars and Other Formal Features
Although many studies seem to establish televisions reputation as a medium that
promotes mindless non-engagement of its viewers, research has been far from definitive.
As of a recent count (Abelman, 1995), over 4,000 scientific research articles and
government reports have been published examining media effects with special emphasis
on the impact of television on children, and still the literature continues to be imprecise as
to how children learn from media interactions. On the other hand, some studies appear to
be more conclusive. Anderson and Bryant (1983) produced results of studies of specific
formal features of television and showed that the research can be fairly robust in that
subjects were able to define and understand the internal mental processes involved with
interpreting mediated messages. For example, they found that valence in children can be
influenced by specific interpretations of what they see and hear. Particular (i.e., a
womans and/or childs) and/or peculiar voices, sound effects, auditory changes, and
visual movements all had a positive effect on recall and comprehension, whereas men's

61
voices, long zooms, and long periods of inactivity were found to be less comprehensible
and less interesting.
Other studies used foreign (Salomon, 1994) or indigenous (Worth & Adair, 1997)
subjects unfamiliar with certain features and/or function common to film and television
production in the United States confirmed these findings. Salomon found that Israeli
children unfamiliar with the format or content were able to recognize and recall
information from various episodes of Sesame Street just as well as American children
who were well exposed to the shows. In teaching film-making young Navajo Indians who
had never seen movies before, Worth and Adair were able to demonstrate in qualitative
and longitudinal studies that certain film communication contains universal that do not
appear to vary, even in light of cultural biases. The Navajo films showed, as a group,
similar functional patterns in their visualizations to those of other cultures. Worth and
Adair found that how we interact and become social through our manipulations of a
variety of symbols and symbol forms (such as those associated with film) enables people
to communicate more fully and fruitfully with each other. For example, while observing
general movement patterns in a film, a viewer can also know exactly what the producer is
trying to communicate. Worth and Adair (p. 8) wrote, you can actually see what is being
done -how it moves. If you write an entire book about it, then it is still. You give it to
someone and he reads it and he does not get the same picture in his mind. What I want to
see is how something can move in front of my eyes." Worth and Adair were suggesting
that a movie is a movie, regardless of the language and culture in which the maker lives.
There are subtle differences in the use of specific techniques, content, and so forth, but
overall, most features and patterns were the same (i.e., the linear nature of episodic

62
sequences). While the study with the Navajo tribe dealt with making films, Worth and
Adair found that much of their findings would carry over to video. In later experiments
several years later, their results with video followed a parallel course. For example, they
found an aversion on the part of their subjects to the use of zooming and close-ups while
teaching them to use the medium. These same proclivities carried over, even with
advanced technologies and easier access found on the video cameras.
Salomon (1979; 1994) suggested that a proper review of media effects of
television involves the examination of the subtle interactions among components of the
medium, individual characteristics of its viewers, and the proposed outcomes for which
the medium is being used. Zillman and Brosius (2000) make a further differentiation. To
them, many of the effects attained by mass media are due to the redundant, stereotypical,
and cyclical attributes by which viewers learn by extracting personally relevant material
(i.e., episodic memory per Tulving (1972)). According to Zillman and Brosius, rather
than creating a sense of mindlessness, this redundancy is actually what creates the power
of the medium. They pointed out that all media (especially mass media) create learning
through the use of specialized media effects (i.e., exemplars) that may be best realized
when used over and over again. For example, most successful mass media producers
understand that, in conveying information, it is important to cut the flow into manageable
chunks and to isolate and focus on some events at the expense of attention to occurrences
in between. In other words, a story line may jump from event to event irrespective of the
events' proximity in location or time. It is an effective strategy so long as the succeeding
events share enough similarity and are sufficient to imply that every individual event is
capable of representing the group of events at large (Zillman & Brosius). In other words,

63
a scriptwriter shows a man getting into his car and driving out of the driveway. Then in
the next scene he is on the highway on the outskirts of a big city. Finally, he pulls up into
his parking space at work. Enough information has been presented to allow viewers to
imagine that time has passed and the man has driven all the way to work and are able to
fill in the information in between. Bryant and Rockwell (1991) found that viewers well-
versed in television and/or film techniques can fill in gaps of omitted information, even
after missing as much as five or more minutes of a television show simply because they
are familiar with the repetitive formats used by screenwriters. Abelman (1995) showed
that viewers can learn how to build an internal concept map for the most often-used
devices in films and teleplays (including time leaps, fall backs, dream sequences, scenes
showing characters recalling past events, etc.) even without comparable real-world
experiences. This appears to further reinforce the theory that television viewing is a
learned activity and highlights the inter-relatedness of childrens linguistic, cognitive, and
perceptual skills for accurate comprehension of televisions most basic narrative device:
temporal sequencing. Corcoran (1981) suggested that intelligence may be defined as a
skill in a particular medium and that symbolic codes that serve communication purposes
can be internalized by a receiver to serve in a cognitive capacity. In other words,
filmed/televised techniques such as zooming, slow motion, or rapid montage may
actually be thought of as tools of thought. Viewers appear to learn how to use these tools
naturally without being trained, because these types of story-building and post-production
editing devices are used redundantly in one show after another (Abelman; Amheim,
1974; Corcoran; Jankowski & Fuchs, 1995; Zillman & Brosius).

64
Pearl (1982) discussed the relationship between form and content and admitted
that her research showed that it is the form (that is, the way it uses verbal and linguistic
codes), not the content of television that is unique. However, she also cautioned that form
and content cannot always be distinguished -no more than grammar and meaning in
any verbal language can (p. 24). Like grammar, some forms are unique to the medium
and apply syntactical meaning only in the context of that medium. For example, slow
motion is not real and its meaning must be learned. But once learned, these formats are
generally used by people in their own thinking. Pearl outlined a taxonomy for this coding
structure of formal features. For example, entropy or form complexity is a combination of
the number of different scenes in a show, the number of characters, and the number of
times the scene and characters appear. Dynamism is the rate of change in scenes and
characters. In this view, content variation is merely determining which features appear
together, their tempo or rate, and/or their variability within scenes.
Unfortunately, the link back to attribution or lack of perceived mental demand has
been caused by television producers who tend to use these same technical features over
and over again as a means of convenience in an entertainment industry that tends to value
volume over quality (Jankowski & Fuchs, 1995). However, redundancy of format and
content can also play a role in causing viewers to make value judgments about what they
see. Zillman and Brosius (2000) cite studies showing that an increased number of
occurrences sometimes influences viewers perception of the news story. For example, if
the number of occurrences of a particular story in a newscast is increased, viewers tend to
be influenced by the story in that the event was probably more significant than it actually
was. (i.e., it tends to slant the news). There is strong evidence that the sheer number of

65
concrete, visible sources relating their experiences and concerns does exert an influence
on issue perception (Zillman & Brosius, p. 110). They go on further by stating that the
incidental use of image exemplars that add non-redundant, specific information to the
text of a news report does influence issue perception. Specifically, the incidental nature
of a pictorial supplement to a story goes unrecognized and, as a result, is integrated with
the narrative information in fostering perceptions and judgment. Over time, if presented
enough times, the photographs are actually remembered longer and the retained image of
the incident was of the photograph, not of the text. Furthermore, if the image that
supplements a news story presents a negative or positive aspect, that image can actually
influence ones value judgments. This is especially so if the presentation speed of that
image is varied (Barnett, 2000; Corcoran, 1981; O'Brien, 1968; Pearl, 1982).
However, presentation speed has not always been consistently viewed as a
positive media effect. Pearl (1982), for example, talked of television being a magnet on
viewers that "shapes their behavior" (p. 1). In an attempt to investigate the ties to violence
and television viewing she found herself in conflict with her own research. First she
claimed that "the rapid form of presentation characterizing American television in which
novelty piles upon novelty in short sequences may well be counterproductive for
organized and effective learning sequences" (Pearl, p.20). But she then goes on to state
that this may be more of a problem with younger children. Kozma (1986) appears to
support this view. He points out that a fast tempo does not necessarily sustain visual
attention, and that an increased tempo is not likely to provide enough time for the viewer,
especially one inexperienced with television formats, to extract that which is needed
bring it into working memory. In addition, others have indicated that as the amount of

66
information increases with the increase speed, short-term memory limitations are soon
reached and information is lost (Lang et al., 2000). The fact that broadcast medium
presents information linearly and does not permit a child to return and review content
puts the onus on the producer to utilize commonly recognized techniques such as
maintain scenes for longer periods, to limit the objectives of the program, slow its speed,
and/or to use slow motion. Kozma points out that shows like Sesame Street accommodate
these limitations by constraining the objectives of each episode and representing
information in different formats. However, simply slowing the speed down may not be
the total solution either. Cronbach and Snow (1977) point out that viewers who are
familiar with the content might actually need a faster speed to maintain their attention.
Salcedo (1985) looked at such features as focal length and found that close-ups
were regarded as more intimate emotionally by viewers. Other features elicited similar
varied impressions (Barnett, 2000; O'Brien, 1968): time on screen (the longer characters
appeared on screen, the better impression they left on viewers), types of transitions (fades
are considered to be more positive or emotional than cuts), and slow motion among them.
In addition to its presentation speed, the video used in the current study utilizes several of
the above features to increase comprehensibility and emotional involvement of its
viewers.
Symbol Systems
Symbol systems differ along several dimensions, which Salomon (1979; 1994)
believed are to be primarily those of notationality, repleteness (density) and resemblance.
A notational symbol system is one in which there is a strong, clear, and consistent
correlation between a set of symbols and a set of objects or concepts, such as a written

67
language, musical notes, or mathematical symbols. Repleteness or density refers to the
relative richness or number of dimensions that the information conveys, like a sketch as
compared to a full-color painting, a textual passage as compared to a picture (even more
so if that picture is moving). Salomon distinguished among symbol systems with respect
to the degree that they resemble the object or concept being represented. Symbol systems
tend to vary in terms of their depictive and/or descriptive reference, from a one-to-one to
a totally abstract representation. Salomon also distinguished between psychological and
real resemblance. Depicting an object with a high degree of realism may or may not be
required for understanding, depending on an individuals pre-conceived ideas as to that
something should look like. As all cognition and learning are based on internal symbolic
representations that are central to all systems of mediated communication and thinking,
actual resemblance (i.e., fidelity) may or may not be required in individual cases.
Gardner, Howard, and Perkins (1974) agreed with this thinking and proposed that
concepts like fidelity should be viewed in its broadest context. That is, the symbolic
process should be compared to how one imagines or conceives something, not
necessarily how true to form it is (i.e., colors, lines etc). In other words, a medium might
be said to be faithful to the real world if it presents information in a way that is true to the
way one thinks (i.e., in a streaming sense).
Salomon (1979) explains that it is not the medium itself that makes the difference
in message processing. Rather, he held that film and television are generally not symbol
systems themselves, but rather are a place where multiple symbol systems are used.
Because they manifest no one system, television and film adopt the symbol system of the
content they depict. For example, television and film may use photography, print, speech,

68
dance, music, etc. and can be replete, resemble reality, and approximate notions of
notational ability, borrowing their power from other media. This same power to employ a
wide range of symbol systems may also define film and televisions potential
shortcomings, reflecting a fallacy in the belief that they always deliver to their potential
(Wetzel et al., 1994). For example, if a video screen is simply placed in front of the class
that displays nothing more than a talking head lecture covering an irrelevant content,
there will be little hope for realizing a valuable instructional delivery. All media carry
with them a systematized methodology for symbolic processing, but that varies in how
they are used, in what context, and by which individuals.
Symbol systems theory plays a significant role in formulating some of the
hypotheses for the current study. First, the pictures found in the depiction of historical
events are iconic in nature and, as such may be considered primary symbols. Second,
although the speed of delivery is considered rapid or fast (at about 200-300 millisecond
per image), the speed within this context is varied, which manipulates the amount of
mental effort required to comprehend the conceptual context. As the images are presented
in combination in a montage format, they represent reality to viewers in a way similar to
the way they think (i.e., in a streaming sense). Lastly, as outlined by Salomon (1979),
fidelity not withstanding, the images in the videos used in the current study are indeed
symbols in that they represent the sole source of reality about the historical events to the
viewers who were not live witnesses to those events. In addition, as per the
conceptualizations of Snow et al. (1965), the amount of information the subjects are able
to extract from the videos may well depend on the amount of previous experience they
have with this medium (television).

69
Evaluating the Educational Impact of New Media
The technologies which are helping video and computing converge, and with
which it is now possible to more easily create visual productions full of rapidly presented
montages, are often referred to as new media. In spite of its inference, the term new
media is not new. Researchers have been alluding to television in this way for almost
fifty years. In writing his introduction to the Seventy-Third Yearbook of the National
Society for the Study of Education, David Olson (1974) noted that new
technologies/media are not the panacea to all possible educational reform needs that
researchers were looking for:
It would be of much greater promise to discover the areas in which
media diverge and hence serve different purposes. Perhaps, the
function of the new media is not primarily that of providing more
effective means for conveying the kinds of information evolved in the
last five hundred years of a book or literate culture but, rather, it is that
of using new media as a means of exploring and representing our
experience in ways that parallel those involved in that literate culture.
(p. 8)
Olsons ideas have been backed up by other researchers. Gross (1974) suggested
that new media does not reduce the vital importance of competence in the basic modes of
intelligence and communication. Chu and Schramm (1968) found in earlier studies that
screen size, using animation, changing the aspect ratio or size of the screen made few, if
any, significant difference in learning, even though these changes all were seen to be
preferred by students in follow-up questionnaires. While the current study also takes
advantage of many of the features of new media, it is not new media by itself that can
take credit for any positive (or negative) outcomes with regards to memory. The basic
premises for how to evaluate the use of media in educational settings remain. Olson
reaffirmed the need to base research of new media using a symbol systems approach. He

70
resolved that previous research showed the lack of a strong theoretical base, and that new
media would not reach their potential until some scheme is shown through empirical
evidence that they can re-specify information, how that this re-structuring would be
influenced by the media that presents it, and what the psychological consequences of
relying on that re-structuring might be. Media are similar in the knowledge they
communicate, but they differ as to the skills needed and developed by using them.
Evidence appears to show that educational media cannot be chosen simply because of
their ability to convey certain kinds of content. Rather, they must be chosen based on
their ability to develop the intellectual processing skills that they help to develop.
Olson (1974) outlined several criteria for analyzing the theoretical potential of
various media for use in educational settings which became a basis some of the goals of
the study and several of the recommendations in Chapter 5. To evaluate the instructional
impact of media one must:
Identify the specific symbol system it uses.
Identify the specific basic skills that are required and how much of literacy
is required for use of this medium.
Identify the intellectual consequences of exposure to that medium (i.e., is
the ability to imagine an action in slow motion dependent upon having
seen slow-motion film?).
Identify the scholastic goals for which the medium is most appropriate.
Determine how knowledge and skill (intelligence) in the medium can be
evaluated.
New Media and Memory
New media make it very much easier to create new an interesting pictorial
storylines using rapidly presented montages, as evidenced by the exponential use in

71
commercial television. Market and psychological research (Reeves & Nass, 1996;
Stephens, 1996) has demonstrated its effectiveness on memory in commercials and
marketing campaigns. Further, researchers (Reeves et al., 1986; Reeves & Nass) appear
to concur with Olsons assertions in light of a rapidly changing technological arena.
Rapid changes are also in process for the way in which researchers view media
assessment. Several things gave rise to an increased interest in looking at the
psychological and sociological aspects of television viewing during the late 1980s. An
evaluation of the then-current research by Reeves et al. found that there was an increased
understanding about attention. First, was the fact that recognition was not always present,
and second, and most important, that continued attention was not always a necessary
prerequisite for learning and understanding. Until that time, most evaluation of attention
was limited to measuring direct eye contact. However, Reeves et al. discovered that due
to increased changes in visual displays, with their rapid scene changes, quick movements,
and zooming, that viewers appeared to have developed a sense of being able to fdter
stimuli and, in essence multi-process. At the same time, much attention was paid to the
potential harm this newer, faster paced television was having on its younger viewers
(increased proclivity for violence, poor attitudes and moral judgment, etc.). The question
Reeves et al. began to look into was why these same effects could not be implemented
into instructional programs in order to accomplish increased educational goals. The
answer may or may not be obvious and requires that one take a deeper look at the aspects
of the new characteristics of message delivery brought on by newer technological
developments. In other words, attention is only one of the variables. Reeves et al. also
found that earlier studies to determine which program attributes caused children to attend

72
to television during shows like Sesame Street or the Electric Company served to also
point out that attention might be better defined as "visual selection" (p. 253). Attention
was measured by a hidden observer who pushed a button when a child's eyes were
directed towards the television screen. This type of eyes on screen method dominated
research for years. However, more recent literature has clearly defined attention as "a
psychological cognitive process that varies within individuals over time" (Reeves et al, p.
254).
Reeves et al. (1986) later began to find that perhaps, attention, as it relates to new
media, is a many-fold process and that it is not necessary an all-or-nothing effect.
Perhaps, different types of attention required different types of responses. Vigilance tasks
are an automatic process that requires people to wait for an event and to respond quickly.
This process decreases over time and only allows a subject to be able to attend them one
at a time. Reeves et al. correlated this to primitive cues that are bom out of humans
animal instincts. Attentionalpreparation tasks involve giving people cues or primes
before a stimulus occurs. Television can be thought of as a continuous priming or cueing
process. This strategy could be used to study the sequencing of various visual devices,
such as pans, zooms, audio silence, screen luminance, or presentation speed and their
ability to prime attention for subsequent program content. Other questions could be asked
about the effects of un-cued or abrupt scene changes on attention and the circumstances
under which cueing increases processing efficiency. Just as Olson viewed symbol
systems, attention appears to require internal devices that are also used differently,
depending on the task that is to be performed. Measuring attention can be added to

73
Olsons list of criteria for analyzing the theoretical potential of various media for use in
educational settings.
Effect of Fast Cuts on Memory
Reeves et al. (1986) re-defined the result of attentional changes in humans,
referring to them Orienting Responses (ORs). These types of responses help to explain,
for example, how humans become habituated to living in large cities with lots of loud
noises, or how one becomes so used to the common sounds found in his or her home (like
a cuckoo clock) and begins to tune them out. Reeves et al. identified several
characteristics of ORs that have implications for evaluating the effectiveness of fast cuts
in an educational setting. Reeves et al. also related these changes to cortical arousal, in
which an individual may have some type of control over and learn how to manipulate for
his or her own purposes. They noted that there is some evidence to suggest that those
who habituate slower perform better on audio and visual tasks than those who habituate
quickly. They also documented studies that showed that high cortical arousal in adults is
related to an increased ability to remember information both of which are premises of
the current study. These ideas about ORs also formed the basis of some of the
recommendations found in Chapter 5.
Presentation Speed versus Content
Lang et al. (1999) more recently discussed adding interesting and arousing
content as having a positive effect on cortical arousal and, therefore, recall and
recognition. She also noted that too much arousal can be harmful. In describing a limited
capacity model, she outlined a sequence where viewers allocate an overall level of mental
processing resources to the complete viewing task based on their goals, interests, etc. In

74
turn, these goals influence the proportion of resources allocated to the various mental
sub-processes (such as storage and retrieval). The structural and content features of the
message elicit orienting behavior (OR) and an automatic allocation of resources. When
there are insufficient resources available to carry out all the sub-processes, some aspects
of processing are performed less well, indicating that too many inputs are overloading the
system. Fast-cuts can be this overloading stimulus, unless the content is delivered in a
structured and continuous way (i.e., it builds its story through a series of related schema).
Lang et al. (1999) further noted that it is the medium itself, by calling upon
certain processing requirements (i.e., its symbol system) that can elicit an Orienting
Response. The Limited Capacity Theory suggests that stimuli that elicit arousal result in
the automatic allocation of resources to encoding and to storage. Thus, an increase in
presentation speed by itself can increase a viewers arousal levels and result in the need
to allocate additional resources to encoding and storage. Lang and Basil (1998) and Lang
et al. (1999) and Lang et al. (2000) found that content alone can also elicit arousal.
Arousal will create the cognitive process to reach its top limit much more quickly if
content is also arousing. On the other hand, as the presentation speed increased for
messages containing only calm content, subjects were able to take on additional cognitive
tasks. Langs studies provided evidence that one should blend the types of content with
the speed of the pace it is presented. Lang et al. (2000) found that memory capacity
problems can be ameliorated if the content is structured so that the amount of new
information provided is limited, irrespective of the presentation speed of the video
presentation. Further, they found that, if the pieces of information are related to some
larger overall schema, the presentation speed also did not matter as much. These concepts

75
were supported by earlier research (Amheim, 1974; dYdewalle & Vanderbeeken, 1990)
into recalling visual sequences where individual frames of information were best
remembered when they related to an overall schema of events. An important factor in
memory enhancement in all these experiments was the cohesiveness of the contextual
content of the messages being presented.
Montage
Mitchell Stephens (1996) elaborated on the potential for an increase in memory
capacity for related schemes of events in his chronicles of the development of montage in
cinema and television. He delineated the beginnings of the use of montage from its use in
the early films of Sergei Einstein (1949), to Chuck Bravermans (1969) use of kinestasis
editing, and its later extensive use in MTV music videos. Stephens demonstrated how
rapid presentation rates have not only enhanced montage videos. He also provided
several examples how producers have been able to replace textual messages as a primary
form of communications. His work appears to support the same type of technological
determinism (p. 21) found in McLuhans (1964) the medium is the message principle:
that the means we use to express our thoughts also change our thoughts. Stephens
suggests that, conceptually, most of thinking with regards to mankinds current system of
logic is based on the literacy of the written word. He even predicts that new media, with
its ability to empower individuals to regularly use vast communicative strategies like
montage, will create a new, unique form of abstract logical thought, similar to what Alan
Kay (1999) had in mind.
In order to provide full understanding of the concept of montage, Stephens (1996)
traces the etymology of the term. Montage was first introduced by Russian filmmakers,

76
who referred to a concept of permitting producers to record movement in the most
complex combinations and [sic] to place points wherever they wanted (p.102). The
Russians began to produce short films that placed short cuts of still photography and
pasted them together to form a surreal photomontage in which "the meaning of a shot is
dependant upon the shots that surround it" (Stephens, p. 102). It is this view of montage
as a collection of highly related thoughts expressed visually that provided the catalyst for
its use in the current study as a way to provide gist memories in videos without the need
for textual supplements.
Montage, in combination with fast cuts, has become a whole new form of
intellectual discourse that Stephens (1996) called complex seeing (p. 178), a technique
that was not available to early television producers. Much of early television lingered
much longer over scenes, allowing the narrative to play a significant role in telling the
story. According to Stephens, new media technologies are changing the paradigm. Now
that the cuts and montage have increased the speed of the instances we see, we are no
longer dwelling on the individual instance but, rather, are being persuaded by the
onslaught and the continuity of the entire series. Not only are we affected by each
individual image but their composite total (p. 180). In essence, the ability of an
individual to become accustomed to the fast-cut montage may be a newer, higher level
form of persistence of vision -the gestalt-like concept that forms the basis of being able
to watch films. The mind closes the gaps in between pictures in a stop-action series of
movements to form one continuous motion. The current study attempts to see if montage
has the potential to follow a similar track in an educational setting.

77
Summary
In researching the current trends of the so-called media effects, one thing has
become evident. It is the newer structures and the form (of which rapid presentation is a
part) of the montage video segments that continue to play an ever-increasing role in how
messages are delivered. This view is not a new one. McLuhan (1964) and Salomon
(1979) predicted media effects long before technology caught up with televisions
promise.
Flowers (1995) cited research as early as the 1970s that contended that educational
environments might be too visuo-centric (p. 570), indicating that looking into visual
processing may have been overdone in many previous studies and that its role in learning
might have been exaggerated. In some respects, Flowers might be correct. Singer (1980)
admitted that many previous assertions regarding textual versus visual image processing
had yet to be supported directly by research evidence, and suggested the need for further
experiments on length of video sequences and its effect on retention of content (or at least
intelligent comprehension of content). Intraubs (1999) experiments appear to answer that
need. She found that humans possess a remarkable capability (p. 67) to understand
scenes that are presented at a pace far more rapidly than normal scanning, so long as the
scene changes overlap with at least somewhat related content.
Over the past decade, media researchers have started to look at the impact new
media have on cognition in a casual viewing environment (Basil, 1994; Hawkins et al.,
1997; Hill & Lang, 1993; Hitchon et al., 1994; Walma van der Molen & Van der Voort,
2000; Watt & Krull, 1977; Zillman, 1991). Some are also beginning to conduct more
extensive research on the effects on memory of editing techniques and the content found

78
in rapidly presented montage in commercial programs (Lang, 1994; Lang & Basil, 1998;
Lang et al., 1999; Lang et al., 2000). It is perhaps Langs work in commercial television
that provides the best backdrop for studies that investigate a corresponding effect of these
new media techniques might have in the educational domain. If one accepts the
contentions of Reeves et al. (1986) and Stephens (1996), it should be easy to understand
how todays youth, brought up on MTV, with its fast cuts, rapid movements, and
iconoclastic acoustics, could get used to, prefer, and possibly learn from similar
techniques in an educational setting.
There are those who contend that the rapid format of television segments takes
away from attention and understanding (Anderson et al., 1979; Anderson & Bryant,
1983; Anderson & Collins, 1988). Reeves et al. (1986) demonstrated information in
several research studies that contradict that belief. Stephens (1996) and Meyrowitz (1985)
appear to agree. They concluded that, once its capabilities are fully implemented using
techniques like fast cuts and video montage, television will provide an opportunity to
transcend the time and place continuum by cutting rapidly between images taken from
different contexts. In other words, viewers can find additional meaning in the relationship
between numerous different scenes. Taken in this way, the video image provides
opportunities for creative seeing, making the case that interpreting images requires more
imagination of their viewers, not less.
This does not mean that educators or their students have no obligation with
regards to content. Lang and Basil (1998) and Lang et al. (1999) and Lang et al. (2000)
have suggested that producers who want their messages to be remembered should create
arousing messages that are presented slowly or at medium speed or calm messages that

79
are presented medium or fast. Producers should not create messages that are, at the same
time, calm and presented slowly or arousing and presented too fast. Carrying this thought
into an educational setting, Verhagen (1992) found that increasing the presentation speed
of video images can be fully interpreted and remembered if learners start their viewing
task with a realistic expectancy of its demand characteristics and are motivated to tune
their mental effort accordingly. Further, research in tachistoscope training (Schale, 1971)
showed that rapid reading training increases visual awareness of larger spans of print, and
that the processing rates can be accelerated without significant loss of comprehension. As
a minimum, even if comprehension is the same, or even if it suffers only slightly, fast cut
montage may not be a negative influence.
Media choices may be as much about cost and efficiency as about cognition and
learning (Cobb, 1997). If it is accepted that one type of efficiency is cognitive efficiency,
it follows that media choices can profit from an understanding of cognitive processes
much in the same way as the latter can be positively leveraged by the type of media that
is ultimately selected.
In this chapter, a theoretical basis was laid out for the methodologies, procedures
and analyses used in the current study. An analogy may be made to describe the way that
video montage derives its ability to convey meaning through the use of closely-related
pictures and rapid presentation. One should view montage similarly to the concept of
chunking in speed reading of text-based passages. In montage, the individual pictures
take the place of words and sentences. Several related pictures become the paragraphs.
Presentation speeds may vary, just as one might vary reading rates, causing variations in
the ability of individuals to remember specifics and derive contextual meaning from the

80
passages. What remains is trace memory for remembering the gist of the storyline, even if
the specific words or phrases (i.e., verbatim memory for the specific pictures) cannot be
recalled exactly.
The next chapter outlines the specific methodologies used in this study to
investigate some of the affects that variation of presentation speed, personal cognitive
style, and gender may have on memories for information presented in the passages. These
methods follow closely the ideas of looking at combined memories, and then breaking
these down into verbatim and gist recollections to see if there are any differences. After
analyzing the results, implications are reviewed and recommendations are made for this
initial look into memories for pictorial representations.

CHAPTER 3
METHODOLOGY
Introduction
In this study the researcher attempted to elicit information about how differing the
presentation speed might affect immediate memory for the pictorial representations found
on the video montage. For this portion of the study, scores for verbatim and gist
memories were pooled together. The researcher then investigated whether the same
pooled scores differed between males and females taking the test. Next, the scores for
verbatim and gist memories were segregated and reviewed in light of the differing
presentation speeds. It is recognized that gist memory is a less exact indication of
immediate memory about individual pictures than is verbatim memory. However, it was
anticipated that a review of gist memory might provide additional overall insights to
memory in that it requires viewers to draw certain conclusions about the contextual
meanings much like a typical reading comprehension test of text-based materials. Lastly,
the researcher investigated the effect cognitive style might have on the verbatim scores
and gist scores taken together, as well as for each of them individually.
A computerized instrument was specifically prepared for this study that displayed
the video at one of three different presentation speeds and then immediately tested
subjects to determine what they remembered from the video. The program was encoded
to score verbatim and gist memory scores separately and also combined into a single
81

82
score (referred to henceforth as the total score or total combined score) in order that
comparisons may be made for speed, gender, and style.
Verbatim memory was tested in two ways. First, subjects were asked in a series of
questions to pick out from groups of four the one picture that appeared in the video. They
were then asked to identity the picture in each of a series of four still images that did not
appear on the video. In order to test whether the subjects were able to comprehend (i.e.,
get the gist of) contextual meaning conveyed by the video, subjects were asked to place
pictures in chronological sequence, to identify by way of multiple choice questions some
of the major techniques used in the video to project moods and attitudes, and to answer
some questions about the circumstances (i.e., storyline) presented in the video.
Population
The subjects for this study came from a population of ninth grade students in a
high school in North Central Florida. Ninth graders were selected due to their closeness
in age to the norms developed for the instrument selected to categorize cognitive style so
that some historical comparisons may be made. That instrument provided normative data,
which was very useful in helping to categorize and compare the current group of students
as to their impulsive-reflective tendencies and current trends.
Sample
The overall sample set consisted of 204 subjects. The entire sample was used in
the analyses to determine the effect of presentation speed on combined test scores as well
as verbatim and gist scores taken separately. The entire sample was also used to
investigate differences between males and females. However, the nature of the instrument
used to categorize this sample set into one of two cognitive styles dictated the creation of

83
a subset of subjects (n=129) from the original sample set of subjects (n=204) for the
second portion of the study, as demonstrated in Figure 1. In the identification of subjects
as to being either impulsive or reflective, a resulted in a certain portion of participants
Impulsive
Slow -Inaccurate
N = 65
N = 37
Fast-Accurate
Reflective
N = 38
$
it
Z
Figure 1
Subjects Included in the MFFT-20 Cognitive Style Test
did not qualify as being in either category. As per the instructions provided with the test
instrument, these outliers (i.e., the fast-accurates and slow-inaccurates) were excluded
from this portion of analysis to determine the effects of cognitive style on immediate
memory. Subjects were categorized by cognitive style as determined by a computerized
program specifically designed for this study. The program scored the subjects on each
test question according to latency to first choice and the total number of errors. The
overall sample and the subset were both randomly placed into one of three groups and
shown the video at one of the three different presentation speeds.
Instrumentation
To categorize the subjects by cognitive style, a 20-item version (MFFT-20) of the
Multiple Familiar Figures Test (MFFT) developed by Cairns and Cammock (1984) at the
University of Northern Ireland was administered. The original MFFT is an instrument
developed by Jerome Kagan (1965; 1966) that was subsequently evaluated for validity
and reliability and adapted over the years by several individuals (Arizmendi, Paulsen, &

84
Domino, 1981; Block et al., 1974; Watkins, Lee, and Erlich, 1978) to determine
impulsive-reflective tendencies. The original format of the MFFT-20 was developed as a
paper version in which participants made their choice of selecting the matching figure
from a set of six distracters by pointing to their choice. The investigator was responsible
for keeping track manually of the number of choices made and timing latency to first
response using a stopwatch. For the current study, the paper copies of the figures and
alternative choices that were to be matched were scanned into a computer and imported
into a program written in Macromedia 8.5 that was specifically developed for this study.
The program presented the pictures and their alternatives on one screen and allowed
subjects to click on their selected picture to indicate their response. The computer
program automatically kept track of the total number of choices made by each participant
and the amount of time to first choice for each of the item sets.
As with the paper versions of the MFFT, subjects in this study were presented
with 12 sample pictures of familiar items and are then asked to identify which one of six
alternatives is identical to the sample. If an error was made the subjects were
subsequently asked by the computer to retry until a correct response was found. Subjects
were also automatically timed as to how quickly or slowly they made their initial choices
(latency) and how many total errors they made. The dividing line between impulsive and
reflective was determined by calculating a median split score for both latency and total
number of errors. The scores were placed into quadrants made up of two intersecting
axes. Those who made very quick but inaccurate decisions ended up in a quadrant labeled
impulsive. Those who were more deliberate (i.e., showed an increased latency to first
response) and made fewer errors than the calculated median were to be determined

85
reflective. Subjects found to be fast-accurate (i.e., faster and more accurate than the
calculated medians) or slow-inaccurate are placed in two other cells. As per the
instructions provided, subjects who landed in either one of these latter two categories
were excluded from the study.
According to the literature, (Berry, 1991; Green, 1985) the impulsive-reflective
test, in addition to leveling-sharpening and dependence-independence, has been one of
the most commonly used and more accurate means to test for cognitive style and to show
how individuals perceive and process visual patterns. Because the treatment in the current
study was to show accuracy of visual processing as well as processing speed, it was
determined that impulsive-reflectivity would be the most closely related of the three
cognitive style classifications to use.
Another goal of this study was to determine if there might be an interaction
between a subjects reaction to rapidly-presented content and his or her cognitive styles.
Rapid visual presentation has been found in several studies to increase cortical arousal.
Further, impulsivity has been reported by some as a stable mediator of the rate of change
in arousal states. Anderson and Revelle (1994) have demonstrated that impulsives are
susceptible to attentional lapses that are, in turn, mediated by impulsivity-related phase
differences in diurnal arousal patterns (p. 334). There is no one correct style. There
could well be a proper time and a place for being impulsive or being reflective. It was
determined that a test for the interaction between cognitive style and how well subjects
receive and process rapid visual presentations could result in some very interesting
findings that could be quite useful in helping to determine the proper use of rapid visual
presentations in the educational environment.

86
The MFFT has been subject to several attempts to refute it as a valid
categorization test (Salkind & Wright, 1977; Watkins et al., 1978). Further, Ikegulu and
Ikegulu (1999) found that the notion of a generalized visual processing rate may be
questionable, based on the fact that there have been few repeated measurement studies to
test the generalizability of the dimension. Other research has indicated that impulsive-
reflective designation might be best depicted on a continuous (i.e., from low to high),
rather than a bi-polar scale, as reported in the impulsive-reflective array (Salkind &
Wright). On the other hand, Salkind and Wright demonstrated in other studies that
continuous scaling seems to contradict a basic definitional premise of a cognitive style
(that is style by its very nature is bi-polar). This apparent anomaly appears to some to
create a potential lack of power for the impulsive-reflective scale to be useful in
accurately classifying a cognitive style. Ault, Mitchell, and Hartmann, (1967) contributed
a loss of power due to Kagans possible over-reliance on latency rather than number of
errors to determine reflective versus impulsivity. The findings of Ault et al. seem to
contradict Kagans original hypothesis, that stated categorization is the result of both
speed and error-rate considered together (Kagan, 1965).
In spite of these and other attempts to dispute it, Kagans MFFT instrument has
been supported in several research studies that more than reinforced its validity
(Anzmendi et al., 1981; Green, May, 1985), making a strong enough case to justify its
use it in the current study. To further strengthen power of the MFFT as a categorization
technique, a 20-item variation of the original instrument was chosen for the current study
because it had been validated by its authors in five separate reliability tests, and who

87
strongly assert that their instrument overcame most of the objections that arose in the
previous studies.
Caims and Cammock (1984) developed the 20-item MFFT (henceforth referred to
as the MFFT-20) and presented five case studies that asserted an increased reliability and
accuracy in that subjects were more accurately categorized into one of the four quadrants
(impulsive, reflective, fast-accurate, or slow-inaccurate). Their instrument used 20 sets of
pictures (instead of the 12 in Kagans original test) that were reduced down from an
original list of 32 items that was, in turn, concatenated and prioritized by several earlier
studies. The authors then ran a test on 90 subjects chosen at random. The 32 items (2
samples and 30 test items) were then paired down to a final list of 20, and chosen on the
basis of their being the most commonly missed. The authors then performed four separate
reliability tests of over 300 total additional subjects that developed sets of norms, and
established strong correlations between order position (i.e., the order in which the picture
sets are presented), error rates, and interactions between age and sex. They then reviewed
the selected items for highest item-total error correlation. Subjects were classified as
reflective or impulsive using the double median split criterion. The median error and
latency scores were total 34 errors and 11.7 seconds respectively. The coefficient alpha
for their 30-item test was .98 for latency and .81 for errors, both at (p< .01). The authors
then developed their 20-item instrument, and re-tested it in three additional reliability
studies. The first study aimed at obtaining reliability coefficients using corrected
correlation between split-halves given two weeks apart. This study also yielded product-
moment correlations for errors and latency of .80 and .83 respectively. Using Spearman-
Brown, the authors then determined that the complete 20-item test would have

88
reliabilities of .89 for errors and .91 for latency. A third study was then conducted to
investigate test-retest reliability of the MFFT-20. The results were favorably compared to
the most reliable form (form F) of original MFFT for error and latency when
administered to a similar age group (Egeland & Weinberg, 1976). Based on the
information discovered in their studies, the authors determined that the MFFT-20 was a
superior instrument with regards to reliability and validity to earlier tests.
A computerized instrument was developed by the current investigator to test
subjects memories immediately following the presentation of a video. The video was a
copy of a kinestasis film depicting a snapshot history of the United States first produced
by Charles Braverman (1969). The current holder of the copyright to this video has
provided permission to the current investigator to utilize a copy for use in this study (R.
Wright, personal communication, March 21, 2001).
Test Descriptions
Initial Development
Two sets of question banks for verbatim memory were developed from which the
final multiple choice test items were derived. The first bank consisted of 15 sets of four
multiple choice questions from which participants would be asked to first pick out the
one picture in each group that actually appeared in the video. The second bank, also
consisting of 15 questions, was made up of similar groupings of still images, but the
participants would be asked to select the one picture in each group of four that did not
appear on the video. Care was taken to insure that all pictures were presented in identical
size, shape and fidelity. To develop a final set of questions concerning gist memory,
several questions were written that asked viewers about the overall content and major

89
themes presented in the video. Some of these were pictorial questions and some were
text-based. These three question banks (consisting of a total of 45 questions) were then
presented to three social studies teachers and one English teacher who evaluated them for
appropriate content, duplication of pictures, and to check for potential fidelity problems
that might distinguish correct responses from distracters. The overall set of 45 questions
was pared down to 25, based on the feedback from the teachers. The questions were then
presented to a several student test viewers who provided additional feedback. As a result
of this phase, the program was changed to randomly place the responses on the screen so
as to randomize the placement of correct answers on the screen and to eliminate the
opportunity for participants sitting next to another to share answers. In addition, the color
schemes for the screen background and text were changed to reflect those combinations
deemed more pleasing to the younger age group of participants for whom this test was
designed.
Pilot Test
A pilot test was conducted with a group of students from a different high school
in the area consisting of subjects of identical age and similar demographic make-up as the
final test group. The pilot group consisted of 97 subjects. They were given the test in a
group setting that was arranged in such a way that no one could see any on elses screens.
So as to supplement the viewers understanding of the instructions, an audio instructional
track of a female reader was added to the computerized program so that the participants
could both see and hear the directions on how to take the test. Headphones were used to
reduce background noises and distractions. To determine reliability, two alpha reliability
scale tests were run. The 15 items dealing with verbatim recall and the 10 items asking

90
gist (i.e., comprehension) questions were analyzed both together and separately.
Adjustments were made to overcome any potential reliability shortcomings prior to
administering the actual test to the selected subjects. A reliability co-efficient on the
combined question set yielded an alpha level of .9028. The co-relational alpha tests to
determine reliability of the test instrument to assess verbatim and gist memory
individually resulted in co-efficient ratios of .7548 and .8662 respectively. These results
indicate that the reliability of the test instrument fell within reasonable ranges of
acceptability for its intended use in this study.
Test Administration
To test for cognitive style, a computerized version of the MFFT-20 test was
developed for use in the current study. The same pictures from the original MFFT-20
were scanned into digital form and placed on an 800 x 600 computer screen format using
Director 8.5, an interactive software program developed by Macromedia. Multiple copies
of an executable form of the program were made so that each subject was able to view his
or her own individual screen in a lab containing 25 Dell Pentium IV and 5 Apple G-4
Computers. Controls were put in place so that no subjects were able to see anothers
screens. Subjects also wore headphones in order to hear the indirections that were also
displayed on the screens. As with the pilot study, the headphones also add the
convenience of providing a more focused environment by eliminating surrounding stray
noises and disruptions. The program presented two sets of sample items and 20 sets of
actual pictures in the exact order as the paper version of the MFFT-20.
The computer program for both the MFFT-20 and the verbatim and gist questions
automatically kept track of the responses by the participants and outputted that

91
information to text files. These files were then imported into Microsoft Excel and
concatenated to consolidate the results for each test instrument. These results were then
exported into an SPSS version 10.0 data file for analysis.
Stimuli
Subjects were shown a video entitled American Time Capsule, a kinsestatic visual
portrayal of a chronological history of the United States which is presented very rapidly
through a montage of approximately 1,300 still photographs, portraits, and paintings. The
original video was imported into a Macromedia Director 8.5 program written specifically
for this study in which it could be played back at one of three different speeds (fast,
medium, and slow), based on a specific password that was randomly provided to each
subject. A spreadsheet was developed to track which password was assigned to each
subject and to ensure an equal distribution of assignments over the sample base, for each
gender, and for each cognitive group. At the fast rate, subjects saw the video at its
original speed (averaging 300 milliseconds per picture) that lasted approximately four
minutes. The medium speed presented the video in about six minutes. The slowest
speed presented the 1,300 pictures in about eight minutes, or one every 500-600
milliseconds. The subjects were first shown a sample portion of the video presented at the
same rate of speed for the purposes of gaining their attention, accustoming them to the
presentation style and speed, and providing an overview of the subject matter. The
subjects were able to begin viewing the entire video when they felt they were ready by
clicking a button on the screen. In order to facilitate correct interpretation of the
directions supplied, the subjects were provided the same set of instructions in a small

92
help box on every screen and heard an audio reading of the instructions whenever they
changed.
Methodology
Testing began immediately after the video was viewed. Subjects were asked to
click a start button when they were ready to begin. The process of clicking the button
initiated the internal system clock that tracked latency to first response. By providing
instructions in this way, the program avoided counting time reading and understanding
the questions as part of the latency calculations during the administration of the
MFFT-20.
The study utilized a combination of pictorial and textual methods to assess
memory. For the verbatim portion of the test, subjects were shown several groups of
pictures and asked to pick out via a mouse click the one picture among three distracters
that did or did not appear in the video. For the gist memory questions subjects were asked
contextual questions in the program. Subjects were also asked to place sets of pictures
into contextual-chronological order based on their presentation in the video.
These methods of testing visual clues appear to be in accordance with long-
established procedures and findings the literature. Archer (1965) proposed that, if one is
testing for visual processing, evaluative vehicles should also be designed that are closely
aligned with the visual process because something may be lost in the translation to
verbal. Corcoran (1981) noted in his studies into visual perception that testing should
utilize the same mode as the original presentation (i.e., textual to textual, pictorial to
pictorial) because textual and pictorial forms of information are processed at different
rates and in different ways. He also found that reading has the tendency to interfere with

93
the retrieval of some internal visual representations. Mayer et al. (1996) indicated that
words may not serve as proper feedback format to visual motion cues. For these reasons,
verbal questions were kept to a minimum in the current study.
In spite of the initial intent to eliminate text-based questions, it was eventually
determined that the gist portion would require some of these types. The gist questions
were intended to simulate reading comprehension-like assessments where subjects are not
asked to memorize every word they read from a book. Rather, they are asked about their
overall impressions and understandings about its content. Therefore, it was determined
that it was not important that the subjects remember every picture in the gist portion of
this the test used in this study, but, like the reading comprehension test, they would be
asked to recall contextual information from of the video. So as to further ameliorate the
potential confound imposed by the additional reading requirements imposed by the text-
based questions, all instructions and questions for the textual portion of the test were
presented in both visual and auditory form.
Research Design
Three separate analyses of variance were utilized in this study. First, it was
determined that to accurately obtain overall test results for the effects of presentation
speed and to detect possible genders differences, the entire sample set had to be analyzed.
To analyze the effect of gender and presentation speed on the overall sample, a two-way
2x3 factorial design ANOVA (2 gender categories and three presentation speeds) was
administered to the entire sample of 204 subjects, as shown in Table 1. Next, it was
determined that testing for two separate score results required a separate multivariate

ANOVA (verbatim and gist) x (fast, medium, and slow). The design for this analysis is
shown in Table 2.
94
Table 1
2x3 Two-way ANOVA (204 Subjects)
PRESENTATION SPEED
GENDER
Fast
Medium
Slow
Male
(N= 107)
34
38
35
Female
(N = 97)
41
31
25
Table 2
Multivariate Analysis of Variance (204 Subjects)
PRESENTATION SPEED
QUESTION TYPE
Fast
Medium
Slow
VERBATIM vs GIST
75
69
60
Last, the procedures for administering the MFFT-20 specified that a certain
number of the subjects, by definition, would be systematically excluded from analysis.
The administration of the cognitive style test yielded a subset of the sample of 129
subjects who were categorized as either impulsive (n=64) or reflective (n=66) and to
Table 3
2x3 Two-way ANOVA (129 Subjects)
PRESENTATION SPEED
STYLE
Fast
Medium
Slow
Impulsive
(N = 64)
22
25
17
Reflective
(N = 66)
24
21
20

95
whom this portion of the test would be administered. As the main and interaction
effects for presentation speeds were also to be analyzed, a second 2x3 factorial design
ANOVA was designed, as shown in Table 3.
Hypotheses
The hypotheses for this study are stated in the null form. The following hypotheses
were determined for the research design using analyses of variance:
Hypothesis 1. There are no significant differences in overall test scores (both
verbatim and gist scores are pooled together), for the overall sample set based on
presentation speed of the video.
Hypothesis 2. There are no significant differences in overall test scores between
male and female subjects in the overall sample set.
Hypothesis 3. There is no significant interaction between presentation speed and
gender for overall test scores.
Hypothesis 4. There are no significant differences in verbatim test scores, for the
overall sample set, based on presentation speed of the video.
Hypothesis 5. There are no significant differences in gist test scores for the overall
sample set, based on presentation speed of the video.
Hypothesis 6. There are no significant differences in overall test scores for those
subjects from the overall sample set who are determined to possess impulsive or
reflective tendencies.
Hypothesis 7. There are no significant differences in verbatim test scores for those
subjects who are determined to possess impulsive or reflective tendencies.

96
Hypothesis 8. There are no significant differences in gist test scores for those
subjects who are determined to possess impulsive or reflective tendencies.
Hypothesis 9. There is no significant interaction between cognitive style and
presentation speed of the video.
Summary
Most of the universally recognized instructional design and development theories
have accurately established that audience/pupil analysis is an important part of the design
process. Cognitive style is one of many different attributes that might need to be
considered when technology that is used in instructional settings. The MFFT-20 used in
this study to categorize subjects as to their reflectivity-impulsivity has been determined
by independent reliability studies to be valid and more powerful than the original MFFT,
producing what is believed to be a more effective measurement tool for analyzing this
form of cognitive style.
The testing instrument developed for this study was tested for reliability and
construct validity through a series of feedback sessions with teachers, peer groups and
pilot testing. The reliability test results during the pilot fell within an acceptable range.
Because the whole test implementation was automated, the statistical outputs were
consistently reported ands made for an ease of analyses along several axes. In addition,
the automated cognitive style instrument and treatment tests allowed them to be
administered in same day, thereby significantly reducing the potential for mortality
among and contamination between the participants. This chapter describes in detail the
implementation and methodology of the test instruments.

97
The next chapter will report the results of these tests. The statistics will be
presented in both narrative and table form, leading to the subsequent chapters where the
results will be analyzed and discussed with regards to the kinds of inferences that may be
made.

CHAPTER 4
ANALYSIS OF THE DATA
Introduction
This study had four overall purposes. The first was to determine whether
differences in presentation speed of a video montage would have an effect on a memory.
Second, this study attempted to determine whether the presentation speed affect males
and females differently. A two-way analysis of variance was used in these two portions
of the study. Third, the study attempted to determine whether there would be differences
between immediate verbatim and gist memory for the content of the video caused by
changes in presentation speed. The design of this portion of the study was a multivariate
analysis of variance. Last, an investigation was made to determine whether the immediate
memories for subjects classified as either impulsive or reflective would be influenced by
changes in presentation speed. The design of this portion of the study was a 2 x 3
factorial design ANOVA.
Results
The data were analyzed with analyses of variance and a multivariate ANOVA.
The relevant statistics are presented in ANOVA summary tables. In addition, descriptive
statistics are included for means and standard deviations.
Hypothesis 1. There are no significant differences in overall test scores (in which
verbatim and gist scores are pooled together), for the overall sample set based on
presentation speed of the video.
98

99
The analysis resulted in an F (2, 201) of .317 (see Table 4). An F ratio of .317 is
not significant at (p <.05), therefore, this null hypothesis was not rejected.
Table 4
Source Table of Analysis of Score Variance by Presentation Speed
Source
SS (Type III)
df
MS
F
sis-
Between Groups
SPEED
6.377
2
3.189
.317
.729
Error
2022.329
201
10.061
Total
2028.706
203
The means and standard deviations for presentation speed are presented in Table
5. In Table 5, Fast represents the results for those who watched the video presented at
its original speed (averaging 300-400 milliseconds per picture), Medium at 1/3 of the
original (approximately 500-700 milliseconds), and Slow (at approximately 1/2 the
original speed (each picture was presented at approximately 1 second each).
Table 5
Score Means and Standard Deviation for Presentation Speed
SPEED
Mean
SD
N
Fast
12.99
3.18
75
Medium
12.70
3.05
69
Slow
12.57
3.30
60
Total
12.76
3.16
204
Flypothesis 2. There are no significant differences in overall test scores between
male and female subjects in the overall sample set.
The analysis resulted in a between subjects effect F (1, 202) of 1.542 (see Table
6). Because an F ratio of 1.566 for this main effect is not significant at (p <.05), this null
hypothesis could not be rejected. The means and standard deviations are presented in

100
Table 6
Source Table of Analysis of Variance for Gender
Source
SS (Type III)
df
MS
F
Sig.
Between Groups
GENDER
15.604
1
15.604
1.566
.212
Error
2013.101
202
9.966
Total
2028.706
203
Table 7
Score Means and Standard Deviation for Gender
GENDER
Mean
N
SD
Male
13.03
107
3.34
Female
12.47
97
2.94
Total
12.76
204
3.16
Table 7. The mean scores and standard deviations are shown by gender (i.e.,
males and females). The table shows a difference in scores and standard deviations
between males and females, with males obtaining a higher average score. However, as
the differences are not significant, they are considered statistically as random variations.
Hypothesis 3. There is no significant interaction between presentation speed and
gender for overall test scores.
Table 8 shows the interaction effect for between subjects of SPEED GENDER
resulted in an F ratio of .435. This F ratio is not significant at (p<.05). Therefore, this null
hypothesis was not rejected.
Table 9 shows the means and standard deviations comparing presentation speed
with gender. Males had higher memory scores than females but, these differences are
shown to be not significant. As such, they are considered random fluctuations.

101
Table 8
Source Table of Analysis of Variance of Speed versus Gender
Source
SS (Type III)
df
MS
F
Sig.
Between Groups
SPEED
8.437
2
4.219
.419
.659
GENDER
15.540
1
15.540
1.542
.216
SPEED GENDER
8.766
2
4.383
.435
.648
Error
1995.428
198
10.078
Total
35268.000
204
Table 9
Score Means and Standard Deviation for Speed versus Gender
SPEED
GENDER
Mean
SD
N
Fast
Male
13.56
3.10
34
Female
12.51
3.20
41
Total
12.99
3.18
75
Medium
Male
12.97
3.46
38
Female
12.35
2.48
31
Total
12.70
3.05
69
Slow
Male
12.57
3.46
35
Female
12.56
3.12
25
Total
12.57
3.30
60
Total
Male
13.03
3.34
107
Female
12.47
2.94
97
Total
12.76
3.16
204
Hypothesis 4. There are no significant differences in verbatim test scores, for the
overall sample set, based on presentation speed of the video.
The analysis of verbatim scores resulted in an F (2, 201) of 1.082 (see Table 10).
Table 10
Source Table of Analysis of Variance for Verbatim Scores for Speed
SS (Type III)
df
MS
F Sig.
Between Groups
12.035
2
6.018
1.082 .341
Error
1117.592
201
5.560
Total
1129.627
203

102
Because the F ratio for verbatim recall was not significant at (p<.05), this null
hypothesis could not be rejected. The means and standard deviation are presented in
Table 14.
Hypothesis 5. There are no significant differences in gist test scores for the overall
sample set, based on presentation speed of the video.
The analysis of gist scores resulted in an F (2,201) of 5.491 (see Table 11).
Because the F ratio for verbatim recall was significant at (p<.05), this null hypothesis was
rejected.
Table 11
Source Table of Analysis of Variance for Gist Scores for Speed
Source
SS (Type III)
df
MS
F
Sig.
Between Groups
32.462
2
16.231
5.491
.005*
Error
594.127
201
2.956
Total
626.588
203
* Significant at (p<.05)
Because the results were found to be significant, and because there were more
than two groups to be compared, a Bonferroni test was run to compare individual pairings
of gist scores to each of the individual presentation speeds. Table 12 shows that the pair
wise comparisons are significant for gist scores when comparing Slow and Fast
speeds, but not significant between Medium and Slow or Medium and Fast.
In order to further compare verbatim and gist scores, a means and standard
deviation table (Table 13) is presented. Table 13 shows a mean score for fast speeds of
6.01 out of 10 possible gist responses for those viewing the video at the fast speed and
5.05 (out of 10) mean score for those viewing it at the slow speed. Table 13 shows that,

103
as a percentage, those subjects who watched the video at the two faster speeds also
tended to get more gist questions correct than those watching at the slow speed. For the
verbatim questions, there were no significant differences based on changes in speed.
Table 12
Multiple Comparisons Between Speed and Gist Scores
SPEED
SPEED
Mean
Difference
Std.
Error
Sig.
95%
Confidence
Interval
Lower
Bound
Upper Bound
Fast
Medium
.25
.287
1.000
-.45
.94
Slow
.96*
.298
.004
.24
1.68
Medium
Fast
-.25
.287
1.000
-.94
.45
Slow
.72
.303
.057
-.01
1.45
Slow
Fast
-.96*
.298
.004
-1.68
-.24
Medium
-.72
.303
.057
-1.45
.01
* Significant at (p<.05)
Table 13
Score Means and Standard Deviation for Gist versus Verbatim Test Items
SPEED
Mean
Std. Deviation
N
VERBATIM
Fast
6.93
2.47
75
(15 questions)
Medium
6.97
2.26
69
Slow
7.48
2.33
60
Total
7.11
2.36
204
GIST
Fast
6.01
1.79
75
(10 questions)
Medium
5.77
1.64
69
Slow
5.05
1.72
60
Total
5.65
1.76
204
Hypothesis 6. There are no significant differences in overall test scores for those
subjects from the overall sample set who are determined to possess impulsive or
reflective tendencies.
As seen in Tables 14 and 15, the sample size is smaller (n=129 versus n=204) due
to the procedures involved in determining impulsive and reflective tendencies. A portion

104
of the sample was systematically excluded due to these subjects being cast as either fast-
accurate or slow-inaccurate, which placed them outside the parameters set forth by the
administrative instructions that accompanied the MFFT-20. An analysis of variance was
performed and obtained an F (2,123) of 6.560 for the main effect for cognitive style. The
F ratio for between subjects was significant at (p<.05), therefore, the null hypothesis was
rejected. The means and standard deviations are presented in Table 15. This table shows
Table 14
Source Table of Analysis of Score Variance by Presentation Speed and Cognitive Style
Source
SS (Type III)
df
MS
F
Sig.
SPEED
2.968
2
1.484
.160
.852
STYLE
60.870
1
60.870
6.560
.012*
SPEED STYLE
7.380
2
3.690
.398
.673
Error
1141.340
123
9.279
Total
21733.000
129
* Significant at (p<.05)
Table 15
Means and Standard Deviation for Overall Scores for Style
SPEED
STYLE
Mean
Std. Deviation
N
Fast
Impulsive
11.50
2.09
22
Reflective
13.50
3.31
24
Total
12.54
2.94
46
Medium
Impulsive
12.36
3.86
25
Reflective
13.24
2.14
21
Total
12.76
3.19
46
Slow
Impulsive
11.82
3.30
17
Reflective
13.10
3.02
20
Total
12.51
3.18
37
Total
Impulsive
11.92
3.17
64
Reflective
13.29
2.85
65
Total
12.61
3.08
129

105
that impulsive subjects had significantly lower correct scores than reflective subjects,
regardless of presentation speed.
Hypothesis 7. There are no significant differences in verbatim test scores for those
subjects who are determined to possess impulsive or reflective tendencies.
In order to further investigate the differences in scores obtained for impulsive or
reflective styles, a one-way analysis variance was developed (Table 16). An analysis of
Table 16
Source Table of Analysis of Score Variance for Verbatim and Gist and Cognitive Style
SS (Type III)
df
MS
F
Sig.
GIST
Between
12.089
1
12.089
4.410
.038*
Groups
Within Groups
348.144
127
2.741
Total
360.233
128
VERBATIM
Between
14.267
1
14.267
2.925
.090
Groups
Within Groups
619.423
127
4.877
Total
633.690
128
* Significant at (p<.05)
was performed and obtained an F (1, 127)of 2.925 for verbatim scores. The F ratio for
between subjects was not significant at (p<.05), therefore, the null hypothesis was not
rejected.
Hypothesis 8. There are no significant differences in gist test scores for those
subjects who are determined to possess impulsive or reflective tendencies.
The same analysis of variance used for Hypothesis 7 was used to show both
verbatim and gist scores (Table 16). The analysis obtained an F (1, 127) of 4.410 for
verbatim scores. The F ratio for between subjects was significant at (p<.05), therefore,
the null hypothesis was rejected.
A means and standard deviation table (Table 17) was developed to further

106
investigate these differences. Table 17 shows that the significant differences in test scores
found in Hypothesis 6 were derived from the gist portion of the memory test, which is
consistent with previous findings regarding the significance of gist versus verbatim
memory from the overall combined test scores.
Table 17
Means and Standard Deviation for Gist and Verbatim Scores for Style
STYLE
Mean
Std. Deviation
N
GIST
Impulsive
5.20
1.77
64
Reflective
5.82
1.54
65
Total
5.51
1.68
129
VERBATIM
Impulsive
6.77
2.14
64
Reflective
7.43
2.27
65
Total
7.10
2.23
129
Hypothesis 9. There is no significant interaction between cognitive style and
presentation speed of the video.
The interaction effect for between subjects of cognitive style (STYLE) and
presentation speed (SPEED) resulted in an F (2, 123) of .398 (see Table 14). This F ratio
is not significant at (p<.05). Therefore, this null hypothesis was not rejected.
Summary
The results of the study indicate that combined memory scores (verbatim and gist
added together) tended to improve as the speed of the presentation slowed down.
However, these differences were not statistically significant and are considered random
fluctuations. The analysis of variance for gender also shows small differences in scores
between males and females. These differences, too, were not significant at the .05 level.
In order to more fully review the differences that presentation speed imposes on
memory, the scores for verbatim and gist portion of the test were recorded and analyzed

107
separately. An analysis of variance shows an interesting result when verbatim and gist
memories are separated and compared. Gist memories were affected more by
presentation speed than verbatim memories. The analysis of variance tables also reveal
that gist memory was more affected by cognitive style than is verbatim memory. The
means and analysis of variance tables reveal that those viewing the video at the fastest
rate scored significantly did better on the gist portion of the memory test, whereas the
differences in mean scores on the verbatim portion of the test were not significant. While
the scores appeared to improve for verbatim questions as the presentation rate was
slowed, the differences were not statistically significant at the .05 level and, therefore,
have to be considered random variations.
Because the analysis of variance for presentation speed involved three different
rate comparisons, a Bonferroni follow-up test was used to pin-point the exact location of
the differences. The results confirm that a significant difference lies between the fast and
slow presentation speeds for gist memory, whereas the differences in scores between the
medium and fast and the medium and slow speeds for either gist of verbatim questions
are not significant.
The pertinent source and means table shows that combined total scores (i.e., totals
in which verbatim and gist scores were added together) improved for those in the sample
who were categorized as being reflective over the impulsive subjects, regardless of
presentation speed. The analysis of variance indicates that these mean variances were
significant at the .05 level. However, the interaction differences between speed and
cognitive style were not found to be significant at the same .05 level. A review of the
means table (Table 17) for impulsive versus reflective for verbatim and gist questions

108
identified the source of the significant variations. Of the two types of memory tested, the
gist comparison proved to be significant, whereas the differences between reflective
subjects on the verbatim portion of the memory test were found to be random variations
at the .05 level.
The role of this chapter was to report on the specific statistics as they related to
the current study. The results of the tests show that many of the variances that were
uncovered were not significant at the .05 level. However, two of the comparison tests (the
one contrasting gist memories and presentation speed and the one that evaluated
cognitive style and presentation speed) revealed significant differences. The impact of
these variations and considerations as to how they relate to the current and potential
future studies will be discussed in the next chapter.

CHAPTER 5
CONCLUSIONS AND RECOMMENDATIONS
Introduction
This study had several purposes. First it attempted to determine if a change in
presentation speed of a video montage might alter subjects combined memory scores
(i.e., a single total score of the entire memory test was evaluated as a whole in which
verbatim questions and gist questions were pooled together). Second, the study attempted
to see if a change in presentation speed would create any significant differences in the
combined memory scores between males and females. Next, the verbatim and gist scores
were considered separately to determine if presentation speed has a significant affect on
either group. The entire sample set was used in the analyses of variance to investigate
these first two conditions. Lastly, the study attempted to find out if there are significant
differences in combined memory scores for the video presented at three different speeds,
based on subjects cognitive style. Categorizing subjects as to impulsive or reflective
tendencies, by definition, resulted in using only that portion of the overall sample set who
were identified as impulsive or reflective in this portion of the study. To further
investigate any differences, verbatim and gist memories were also considered separately
for this portion of the study.
Analyses of variance were used to investigate the effects of presentation speed,
gender, and cognitive style. Findings are reported below.
109

110
Findings
All hypotheses are stated in the operational null form and were tested at the .05
level.
Hypothesis 1. There are no significant differences in overall test scores (both
verbatim and gist scores are pooled together), for the overall sample set based on
presentation speed of the video.
The null hypothesis was not rejected.
Hypothesis 2. There are no significant differences in overall test scores between
male and female subjects in the overall sample set.
The null hypothesis was not rejected.
Hypothesis 3. There is no significant interaction between presentation speed and
gender for overall test scores.
The null hypothesis was not rejected.
Hypothesis 4. There are no significant differences in verbatim test scores, for the
overall sample set, based on presentation speed of the video.
This hypothesis was not rejected.
Hypothesis 5. There are no significant differences in gist test scores for the overall
sample set, based on presentation speed of the video.
The null hypothesis was rejected. Further, a Bonferroni post-hoc test indicates that those
watching the video at the fastest speed scored significantly better on the gist portion of
the test than did those watching the video at the slow speed.

Ill
Hypothesis 6. There are no significant differences in overall test scores for those
subjects from the sample who are determined to possess impulsive or reflective
tendencies.
The null hypothesis was rejected.
Hypothesis 7. There are no significant differences in verbatim test scores for those
subjects who are determined to possess impulsive or reflective tendencies.
This hypothesis was not rejected.
Hypothesis 8. There are no significant differences in gist test scores for those
subjects who are determined to possess impulsive or reflective tendencies.
This hypothesis was rejected.
Hypothesis 9. There is no significant interaction between cognitive style and
presentation speed of the video.
The null hypothesis was not rejected.
Discussion
Presentation Speed
There were no significant differences in overall combined test scores when the
presentation speed of the video was varied. While there were differences in the total
number of correct responses, those differences were not significant, meaning that any
difference between the scores in this portion of the study must be considered random
variations. The lack of significance between the groups due to changes in presentation
speed appear to run counter to many commonly understood ideas about memory
processing and speed of message delivery -that immediate memory could be conversely
affected by the presentation speed of the video. Part of the reason that subjects in the

112
slower groups did not score better might have been because they became disinterested
(the videos lasted six and eight minutes versus four for the fastest group) as the duration
of the video grew longer -a phenomenon observed by the principle investigator while
conducting this study. The poorer scores may have been due to a lack of attention that
could have counter-balanced any positive gain in memory that the subjects might have
enjoyed by being able to look at the pictures for longer periods of time.
Gender
The total mean score for males (13.03) was larger than that for females (12.47).
However, the analysis of variance did not indicate that this difference was significant at
the .05 level, indicating that any differences in scores that did exist between males and
females were random variations. Therefore, it was determined that gender, as a variable,
was not a factor and need not be considered separately in the remainder of this study.
Verbatim versus Gist Memory
The mean scores for both verbatim (7.11 out of 15 possible correct responses) and
gist (5.65 out of 10) both represent an accuracy level of about 50% of the total number of
questions for each type. For verbatim memory questions, subjects who watched the video
at the slowest rate performed better than did those who watched the video at the medium
or fastest rates. The scores appeared as if they would improve as the presentation rate
slowed down, but the differences between the fast and medium speeds were virtually
identical (6.93 and 6.97 respectively). Further, these differences were not significant at
the .05 level at any speed, therefore were considered to be random fluctuations. In
summary, verbatim memory was not affected by presentation speed.

113
When considering gist memory, the means table indicates that the scores
improved as the presentation speed got faster. The analysis of variance reveals that these
differences were significant at the .05 level. Because there were three presentation groups
being compared, a follow-up test was run to pin-point where the significant differences
existed. The Bonferroni test shows that the significant difference between test scores
exists between the fasted and the slowest presentation speeds, while the differences
between the other two groups (fast-medium and medium-slow) were not significant. The
difference only mattered when one compares the two extremes of the presentation speeds.
Cognitive Style
By definition, reflective individuals were expected to score better on tests than
impulsives. The means and standard deviation table used confirms that this assumption
held true in this study as well. Reflective subjects did score significantly better in the
overall test scores in which the verbatim and gist questions were combined. However, in
a separate analysis in which the verbatim and gist scores were viewed separately, the
differences for verbatim memory were found to be not significant at the (p>.05) level. On
the other hand, there was a significant difference in gist memory when considering
cognitive style.
The mean number of overall correct responses for impulsive subjects was 11.92
out of a possible 25 (approximately 48%). For reflective subjects, the overall mean was
13.29 correct responses out of 25 questions (approximately 53%). The descriptive
statistics are more revealing when verbatim and gist scores are separated. When looking
at the verbatim scores alone, impulsive subjects scored an average of 6.77 out of 15
possible verbatim responses (approximately 45%). Reflective subjects correctly answered

114
an average of 7.43 (approximately 49%). These differences were not found to be
statistically significant and, therefore, were considered random fluctuations. In the gist
portion, however, impulsive subjects scored an average of 5.20 out of a possible 10
(52%). The reflective subjects scored an average of 5.82, or a little more that 58% correct
answers, which was shown by the analysis of variance to be statistically significant.
These results show that while cognitive style did affect memories for the video, the gist
memory was affected most.
Interaction Between Speed and Style
No significant differences were found to be caused by an interaction between
presentation speed and cognitive style. This means that, at least for the subjects in this
study, cognitive style mediated immediate memory but it was not further affected by a
change in presentation speed. That portion of the subject group that was found to be
impulsive or reflective was affected by the speed in the same way as the remainder of the
sample group.
Implications
The results of this study indicate that students tend to remember gist information
from rapidly presented videos better than those presented at slower speeds. These results
appear to contradict earlier research that seemed to indicate that the viewers would be
able to remember more information from pictures if they are presented at slower speeds.
One of the reasons might have been because the earlier studies measured memory on a
combined basis, concentrating solely on measuring immediate verbatim memory. The
parsing of gist and verbatim scores in the current study has uncovered a potential new
approach to investigate differences in pictorial cognition. The fact that the significant

115
differences in gist memory was masked when the verbatim and gist memory scores were
considered together reveals that researchers may need to identify new paradigms that take
into consideration the goals and intended outcomes of the instructional activities they are
investigating. Not all classroom experiences need to have rote memory as a sole learning
outcome. Stimulating gist memories, like teaching reading comprehension, may have its
own place in overall instructional schemes.
To a degree, the results of the cognitive style instrument used in this study to
categorize subjects indicate that some things about learning styles have changed since the
original instrument was analyzed and developed. When one compares the results of
MFFT-20 cognitive style in the current study to the norms provided by Cairns and
Cammock (1984), not only has the median total number of errors decreased (from 28-30
in the Cairns and Commock studies to eight in the current study), but also so has the
median latency to first response (from 18 to 9.12). These reductions seem to indicate that
latencies to first response for visual activities are growing shorter, but the quicker
responses do not always translate to higher error ratios. Students appear to be developing
a propensity for remembering things from rapid visual presentations.
Another change that took place is the shrinking of the differences in visual
cognition between males and females. With Cairns and Cammock, female responses were
considered to lie so dramatically outside of the norms that they were systematically
eliminated from their studies. In the current study, any differences between males and
females that did exist were found to be non significant. While females still may be found
to be more reflective than their male counter-parts, these differences may be growing
smaller.

116
The results of this study also indicate that using the impulsive-reflective cognitive
style instrumentation may be still a valid measuring tool. While both verbatim and gist
memories were both negatively affected by cognitive style, it was only gist memory for
the rapidly presented videos that was affected significantly. When considering the entire
sample without regard to style, the analysis of variance did not yield any significant
differences. The cognitive style portion of this study uncovered some discrepancies (i.e.
learning difficulties) that might have otherwise gone unnoticed.
Whether the changes found here are the result of differences in casual television
viewing habits or computer usage (or both in combination) was outside the purview of
the current study. However, many of the results indicate that something is different about
the way todays youthful learners perceive visual inputs, creating several interesting
scenarios for future studies.
It is noteworthy that the total correct number of responses for all groups was quite
low (around 50%). The relatively low numbers of correct to total possible responses
serves to reinforce that the purpose for integrating video presentations into teaching and
learning situations has not changed. While a change in presentation style may provide an
essential pre-condition for increasing knowledge, it still needs to be coupled with sound
instructional strategies for any learning to take place.
Recommendations
The results of this study have raised some interesting questions about the nature
of visual perception and it discovered a potential for investigating several new paradigms
for instruction. The following recommendations are made for future studies in this area:
1. When one compares the median latencies and errors and the percentages of

117
fast-accurates between the Caims and Cammock studies and the current
study, there appears to be a general lowering of the latencies and errors and
a general increase in the number of fast-accurates as a percentage of the
total sample. Further investigations could be developed to determine what,
if any, the increased usage of rapidly presented digital media that was
found in the review of the literature may have on cognitive style.
2. Other trends between the Caims and Cammock study and the current one
appear to have developed. For one, females were excluded from the Caims
and Cammock study due a determination that the differences between
males and females were significant enough at that time that including them
would confound the results. In the current study, the differences between
males and females were not significant. An investigation could be made to
determine whether the differences in cognitive styles between males and
females have lessened with the latters increased usage of digital media.
3. One could investigate the optimal duration of rapidly presented frames
before verbatim and gist memories deteriorate to the point where
presentation speed has a significant affect.
4. One could look into the effect of other variations of presentation speeds on
memory. The current study only looked into rapid presentation. There have
been several studies that have looked with limited success into the effects
of slow motion on long-term memory (Corcoran, 1981; Kozma, 1986;
O'Brien, 1968; Olson, 1974; Pearl, 1982). Some of these studies are quite
old and modem digital creation techniques make it much easier to design

118
and evaluate these types of studies than it was previously.
5. Another dimension of learning is long-term memory and forgetting. The
current study only dealt with short-term memory, an essential but
insufficient condition for learning. A study could be developed to
investigate the effects of repetition and rehearsal, and whether presentation
speed has any interaction effect. Future studies could look into using short,
rapidly presented videos as an orientation/leaming set, and expand the
subject base to make the results more generalizable to a larger population.
6. It is possible that the impulsive-reflective comparisons for gist memory
indicated a significant result was due to a relatively low number of test
questions devoted to that portion of the test (10 questions devoted to gist
versus 15 for verbatim). For the current study, a determination had to be
made to balance the need for statistical power with that for limiting the
total number of questions to a reasonable number. While the reliability
analysis performed in the pilot study on the gist questions proved
successful, it is recognized that the low number of questions may have had
a bearing on the results. A future study could be designed to study only the
effect of gist memories where more questions could be asked.
7. Joseph Hill (1981), prior to his untimely death, had begun to look into the
effects of contextualism on recognition and recall. These studies could be
expanded and investigated to determine how contextual memory correlates
with presentation speed.
8. One of the premises of the current study was that the ability of todays

119
media-centric youth to more quickly perceive and assimilate rapidly
presented visual images was due to casual television viewing habits and
video game usage. Studies could be developed into determining whether
increasing perceptual skills can be a trainable activity, especially using
older subjects who might not have had the same viewing and gaming
opportunities previously.
9. The results of this study align to a certain degree with Stephens (1996)
assertions that video montage can convey composite thoughts in as much as
a paragraph does in text-based communications. Based on the results of the
current study, a further investigation could be developed to look further
into the effects of continued used of video montage to develop gist
memory.
10. Studies could be developed to determine the relationship between numbers
of incorrect choices students make prior making correct selections,
latencies, and learning progress. New interactive technologies provide
novel opportunities for educators to easily build test instruments that track
both these test-taking characteristics. While the results were not included,
the computer software used in the current study to track verbatim and gist
memory was also programmed to track latency as well as correct responses
in the exact manner that was used in the automated MFFT-20 cognitive
style test. Tracking latencies as well as number of responses made prior to
correct selection could be incorporated into more formal studies to more
thoroughly track the progress of student learning.

120
Finally, it should be noted that the technology employed allowed the investigator
to conduct this study with a small number of large groups (30+ at the same time). This
was a bane as well as a blessing. There were times when the interactions between
subjects became somewhat difficult to control. Luckily, there were three proctors in the
room to ensure that subjects remained on task and were correctly following directions.
Future studies might fare better if the number of sessions were increased to allow for
smaller group sizes. Large group interactions did cause a certain number of distractions
and may have had some bearing on the results. While these types of large groups might
indeed be more authentic (i.e., they represent more closely an actual classroom setting),
crowd control is an issue and a potential confound to the results of the research.
Summary
In some regard, readers of this study would be correct if they note that it appears
to have raised more questions than it has answered. This was a part of the original
motivation to do the study. The catalyst to do this study was an inclination on the part of
this investigator to apply a research base to the myriad of recent unsubstantiated reports
found in the literature and personal anecdotal observations about the changing nature of
cognition and communications in todays technology rich society.
As a minimum, it appears that ideas concerning the nature of cognitive style are in
need of updating. The review of the literature revealed that little has been done in this
area for almost twenty years and the median latency and errors have been shown to
decrease significantly since original studies took place.
It should be noted that it was never the intent of this study to discover some new
paradigm for learning. Rather, it simply intended to re-open some discussions and

121
reinforce existing theories about learners and knowledge gained from viewing pictures.
The results of this study simply show that todays youths have changed their propensity
for and skills in mediated pictorial cognition. Technological improvements have both
significantly increased the occurrence of rapidly presented montage passages found in
television programs, in movies and movie trailers, and in commercials in particular. They
have also made it possible to more effectively test the effect of the rapid presentation
speeds and cognitive style that was not practicable previously.
The analyses of variance in this study revealed both some significant and non
significant results. They showed that ones cognitive style does mediate memories for
pictorial representations, but, overall, presentation speed does not. In addition, there are
no significant differences between the way males and females remember them.
The lack of significance in certain portions of this study may be significant. First,
the fact that there were no more than random differences between slow medium and fast
presentation speeds is noteworthy in itself, as these results may not be what one might
have expected in light of previous research into the interactions between presentation
speeds and object memory. Further, once the non-significant, combined test scores were
separated into their verbatim and gist portions and were subsequently analyzed, the gist
score variations were found to be significantly affected by presentation speed. Co
mingling the scores for memory together into one combined result had the effect of
masking a significant interaction that was occurring -one that might have otherwise gone
unnoticed. These results further reinforce long-standing, generally accepted instructional
principles that promote audience analysis and the matching of competency assessment
with very specific the instructional goals.

APPENDIX A
SCREEN SHOT FROM MFFT-20 COGNITIVE STYLE TEST
122

APPENDIX B
SCREEN SHOT FROM TEST INSTRUMENT SHOWING SAMPLE VERBATIM
QUESTION
123

REFERENCES
Abelman, R. (1995). Gifted, LD, and gifted/LD children's understanding of temporal
sequencing on television. Journal of Broadcasting and Electronic Media, 39, 297-
312.
Allison, T., Ginter, EL, MacCarthy, G., Nobre, A.C., Puce, A., Luby, ML, et al. (1994).
Face recognition in human extrastriate cortex. Journal of Neurophysiology, 71,
821-825.
Alwitt, L. F., Anderson, D. R., Lorch, E.P., & Levin, S. R. (1980). Preschool children's
visual attention to attributes of television. Human Communication Research, 7,
52-67.
Anderson, D., Alwit, L., Lorch, E., & Levin, S. (1979). Watching children watch
television. In G. M. L. Hale (Ed.), Children's understanding of television (pp.
331-353). New York: Academic Press.
Anderson, D. B., & Bryant, J. (1983). Research of children's television viewing: The state
of the art. In J. A. Bryant, D. (Ed.), Children's understanding of television:
Research on comprehension and attention, (pp. 331-354). New York: Academic
Press.
Anderson, D. R., & Collins, P. A. (1988). The impact of children's education:
Television's influence on cognitive development. Washington, DC: Office of
Educational Research and Improvement.
Anderson, K. K.., & Revelle, W. (1994). Impulsivity and time of day: Is rate of change in
arousal a function of impulsivity? Journal of Personality and Social Psychology,
67(2), 334-344.
Anderson, L. A. (1986). Attention and visual perception: The availability of features.
Paper presented at The Human Factors Society-30th Annual Meeting, Houston,
Texas.
Archer, E. J. (1965). Concept identification as a function of obviousness of relevant and
irrelevant information. In D. P. R. C. Anderson & Ausubel (Ed.), Readings in the
Psychology of Cognition New York: Holt, Rhinehart & Winston.
124

125
Arizmendi, T., Paulsen, K., & Domino, G. (1981, Spring). The Matching Familiar
Figures Test: A primary, secondary, and tertiary evaluation. Journal of
Psychology, 812-818.
Amheim, R. (1974). Virtues and vices of the visual media. In D. R. Olson (Ed.), Media
and Symbols: The forms, expression, communication, and education (pp. 180-
210). Chicago: University of Chicago Press. The Seventy-Third Yearbook of the
National Society for the Study of Education.
Arnold, R. S. (1996). Instructional effectiveness of compressed motion video signals.
(Doctoral Dissertation, University of Central Florida, 1996). Dissertation
Abstracts International, 51 (03), 1103A.
Ault, R. L., Mitchell, C., & Hartmann, D. P. (1967). Some methodological problems in
reflective-impulsivity. Child Development, 47, 227-231.
Ayabe, H. I. (1973). Measuring reflection-impulsivity accurately. Paper presented at the
Annual Meeting of the American Educational Research Association, February 25
- March 1, New Orleans, LA, (ERIC Document Reproduction Service No. ED
079 333).
Bargh, J. A. (1988). Automatic information processing: Implications for communications
and effect. In L. Donohew, Syphder, H. & Higgins, E. T. (Ed.), Communication,
social cognition, and affect (pp. 9-32). Hillsdale, NJ: Lawrence Erlbaum
Associates.
Barnett, B. (2000). The impact of slow motion video on viewer evaluations of television
news stories. News Photographer, 55(7), 4-11.
Bartlett, F. C. (1932). Remembering. Cambridge: Cambridge University Press.
Basil, M. (1994). Multiple resource theory I: Application to television viewing.
Communication Research, 21, 177-207.
Berio, D. K. (1960). The process of communication: An introduction to theory and
practice. Toronto, CA: Holt, Rinehart & Winston.
Berry, L. H. (1991). Visual complexity and pictorial memory: A fifteen-year research
perspective. Paper presented at the Annual Convention of the Association for
Educational Communications and Technology. (ERIC Documentation
Reproduction Service No. 334 975).
Biederman, L, Rabinowitz, J. C., Glass, A. L., & Stacy, E. W. (1974). On the information
extracted from a glance at a scene. Journal of Experimental Psychology, 103,
597-600.

126
Block, J., Block, J. H., & Harrington, D. M. (1974). Some misgivings about the Matching
Familiar Figures Test as a measure of reflection-impulsivity. Journal of
Developmental Psychology, 10, 611-632.
Boyden, J. G., & Gilpin, A. R. (1978). Matching familiar figures test and Stroop test
performance in adults. Perceptual and Motor Skills, 46, 854.
Brainerd, C. J., & Reyna, V. F. (1990). Grist of the gist: Fuzzy-trace theory as the new
intuitionism. Developmental Review, 10, 3-47.
Brainerd, C. J., & Gordon, L. L. (1994). Development of verbatim and gist memory for
numbers. Developmental Psychology, 30(2), 163-177.
Braverman, C. (1969). The world of kinestasis. Media & Methods, 6(3), 61-62.
Bridgeman, B. (1980). Generality of a "Fast" or "Slow" test-taking style across a variety
of cognitive tasks. Journal of Educational Measurement, 17, 211-217.
Brown, R. & Kulick, J. (1977). Flashbulb memories. Cognition, 5, 73-99.
Brumby, M. N. (1982). Consistent differences in cognitive styles shown for qualitative
biological problem-solving. British Journal of Educational Psychology, 52, 244-
257.
Bruning, R. H., Shraw, G. J., & Ronning, R. R. (1999). Cognitive psychology and
instruction. Upper Saddle River, NJ: Prentice-Hall.
Bryant, J., & Rockwell, S. C. (1991). Evolving cognitive models in mass communication
reception processes. In J. Bryant, Zillman, D. (Ed.), Responding to the screen:
Reception and reaction processes. Hillsdale, N.J.: Lawrence Erlbaum Associates.
Bums, B. (1992). Percepts, concepts, and categories: The representation and processing
of information. Amsterdam: Elsevier Science Publishers.
Caims, J., & Cammock, T. (1984). The 20-Item Matching Familiar Figures Test. (ERIC
Document Reproduction Service: No. 015681-4).
Calvert, S. L. & Scott, C. (1989). Sound effects for children's temporal integration of
fast-paced television content. Journal of Broadcasting and Electronic Media,
33(233-246).
Campbell, D., & Fiske, D. (1959). Convergent and discriminate validation by the multi
trait/multi-method matrix. Psychological Bulletin, 56, 81-105.

127
Campbell, D. S., & Davis, R. B. (1982). On the validity of reflection-impulsivity as a
construct in classroom research. Paper presented at the Annual Meeting of the
American Educational Research Association, New York, March, 1982, (ERIC
Document Reproduction Service No. ED 222 502).
Canelos, J. (1986, January). External pacing as an instructional strategy for the design of
micro-computer based instructional program to improve performance on higher
level instructional objectives. Paper presented at the Annual convention of the
Association for Educational Communications and Technology, Las Vegas, N.V.
Carr, T. H. (1982). Words, pictures, and priming: On semantic activation, conscious
identification, and the automaticity of information processing. Journal of
Experimental Psychology Human Perception and Performance, 8, 757-777.
Chu, G. C. & Schramm, W. (1968). Learning from television: What the research says.
Washington, D. C.: National Association of Educational Broadcasters, Stanford,
CA. Institute for Communication Research.
Clark, R. E. (1983). Reconsidering research on learning from media. Review of
Educational Research, 53, 445-459.
Cobb, T. (1997). Cognitive Efficiency: Toward a revised theory of media. Educational
Technology Research and Development, 45(4), 21-35.
Coltheart, V. (1999). Introduction. In V. Coltheart (Ed.), Fleeting Memories: Cognition
of brief visual stimuli (pp. 1-12). Cambridge, MA: The MIT Press.
Comstock, G. A., Chafee, S., Katzman, N., McCombs, & M., & Roberts, D. (1978).
Television and human behavior. New York: Columbia University Press.
Cooper, L. A. (1982). Strategies for visual comparison and representation: Individual
differences. In R. J. Sternberg (Ed.), Advances in the Psychology of Human
Performance Hillsdale, NJ: Lawrence Erlbaum Associates.
Cooper, R. (2000). Visual dominance and the control of action, [Online]. Available:
http://www.psyc.bbk.ac.uk/staff/rc/publications/cogsci98/visual_dominance/
[2001, March 11],
Corcoran, F. (1981). Processing information from screen media: A psycholinguistic
approach. Educational Communications and Technology Journal, 29, 117-128.
Cowan, N. (1998). Children's memories according to fuzzy-trace theory: An endorsement
of the theories purpose and some suggestions to improve its application. Journal
of Experimental Child Psychology, 71(2), 144-54.

128
Cronbach, L. J. & Snow, R. E. (1977). Aptitudes and instructional methods: A handbook
of research on interactions. New York: Irvington.
Crowder, R. G. (1976). Principles of learning and memory. Hillsdale, NJ: Lawrence
Erlbaum Associates.
Dale, E. (1969). Audiovisual methods in teaching. New York: Holt, Rinehart & Winston.
Davis, E. T., Scott, K., Pair, J., Hodges, L.F., & Oliverio, J. (1999). Can audio enhance
visual perception and performance in a virtual environment? Paper presented at
the Human Factors and Ergonomics Society-43rd Annual Meeting, Los Angeles,
CA.
Dick, A. O. (1973). Utilization of visually presented information: Final Report.
Washington, DC: National Center for Educational Research and Development.
(ERIC Document Reproduction Service No. ED 078 365).
Dixon, N. F. (1981). Pre-conscious processing. New York: John Wiley & Sons.
Downs, E. (1989). The effects of cueing strategy, level of information, and motion
condition on children's interpretation of implied motion in pictures. (Doctoral
Dissertation, University of Florida, 1989). Dissertation Abstracts International,
51 (03), 0792A.
Durso, F. T. & Johnson, M. K. (1979). Facilitation in naming and categorizing repeated
pictures and words. Journal of Experimental Psychology: Human Learning and
Memory, 5(5), 449-459.
d'Ydewalle, G., & Vanderbeeken, M. (1990). Perceptual and cognitive processing of
editing rules in film. In R. Groner, d'Ydewalle, G., Parham, R. (Ed.), From Eye to
Mind: Information Acquisition in Perception, Search, and Reading (pp. 129-139).
Amsterdam: Elsevier Science Publishers, B.V.
Ederlyi, M. H. (1985). Psychoanalysis: Freud's cognitive psychology. New York: W. H.
Freeman.
Edgar, G. K. (1997). Visual accommodation and virtual images: Do attentional factors
mediate interacting effects of perceived distance, mental workload, and stimulus
presentation modality? Human Factors, 59(3), 374-382.
Egeland, B., & Weinberg, R. A. (1976). The matching familiar figures test: A look at its
psychometric credibility. Child Development, 47, 483-491.
Ellis, W. D. (1938). A source book of gestalt psychology. New York: Harcourt, Brace &
World.

129
Estes, W. K. (1994). Classification and cognition. New York: Oxford University Press.
Flannagan, P. (1998). Aurally and visually guided search in a virtual environment.
Human Factors, 40 (3), 461-469.
Flowers, J. H. (1995). Musical versus visual graphs: Cross-modal equivalence in
perception of time-series data. Human Factors, 57(3), 553-570.
Gardner, H., Howard, V., & Perkins, D. (1974). Symbol systems: A philosophical and
educational investigation. In D. R. Olson (Ed.), Media and Symbols: The forms of
expression, communication, and education (pp. 27-55). Chicago: University of
Chicago Press. The Seventy-Third Yearbook of the National Society for the Study
of Education.
Gentry, C. G. (1998). Introduction to instructional development. Belmont, CA:
Wadsworth Publishing.
Gibson, E. J. (1969). Principles of perceptual learning and development. New York:
Appleton-Century-Crofts.
Gibson, J. J. (1979). The ecological approach to visual perception. Boston: Houghton
Mifflin.
Goldstein, E. B. (1989). Sensation and perception, (3rd Edition). Pacific Grove, CA:
Wadsworth.
Goodglass, H. (1971). Stimulus duration and visual processing time. Perceptual and
Motor Skills, 33, 179-182.
Green, K. E. (1985). Cognitive style: A review of the literature Chicago, IL: Johnson
O'Connor Research Foundation, Human Engineering Lab. (ERIC Document
Reproduction Service No. ED 289 902).
Gross, L. (1974). Modes of communication and the acquisition of symbolic competence.
In D. R. Olson (Ed.), Media and symbols: The forms, expression, communication,
and education (pp. 56-80). Chicago: University of Chicago Press. The Seventy-
Third Yearbook of the National Society for the Study of Education.
Gummerman, K., & Gray, C. R. (1972). Age, iconic storage, and visual information
processing. Journal of Experimental Psychology, 13, 165-170.
Haber, R. N. (1970, May). How we remember what we see. Scientific American, 221,
104-112.
Hawkins, R. P., Pingree, S., Bruce, L., & Tapper, J. (1997). Strategy and style in attention
to television. Journal of Broadcasting & Electronic Media, 41, 245-264.

130
Hayes-Roth, B., & Hayes-Roth, F. (1977). Concept learning and the recognition and
classification of exemplars. Journal of verbal learning and verbal behavior, 18
(5), 321-338.
Hedberg, J. G., & McNamara, S. E. (1985). Matching feedback and cognitive style in
visual CA1 tasks. Paper presented at the Annual Meeting of the American
Educational Research Association, Chicago, IL, March 31 April 4, 1985. (ERIC
Document Reproduction Service No. ED 260 105).
Hill, J. E. (1981). The educational sciences: A conceptual framework. West Bloomfield,
MI: Hill Educational Sciences Research Foundation.
Hill, S., & Lang, A. (1993). The effects of redundancy, pacing, and visual complexity on
memory and recognition of in-stadium advertisements. Paper presented at the
International Communication Association, Miami, FL.
Hill, W. F. (1985). Learning: A survey of psychological interpretations (4th ed.). New
York: Harper & Row.
Hitchon, J., Druckler, P., & Thorson, E. (1994). Effects of ambiguity and complexity on
consumer response to music video commercials. Journal of Broadcasting and
Electronic Media, 38, 294-306.
Hokanson, B. (2000). Accelerated thought: Electronic cognition, digital image creation
and analysis as a means to examine learning and cognition. Dissertation
Abstracts International, 61 (03), 953A. (UMI No. 9963005).
Huba, M. E., & Vellutino, F. R. (1980, August). The development of visual encoding and
retention skills. Journal of Experimental Child Psychology, 30, 88-97.
Ide, T. R. (1974). The potentials and limitations of television as an educational medium.
In D. R. Olson (Ed.), Media and Symbols: The forms of expression,
communication, and education (pp. 330-356). Chicago: University of Chicago
Press. The Seventy-Third Yearbook of the National Society for the Study of
Education.
Ikegulu, P. R., & Ikegulu, T. N. (1999). The effectiveness of window presentation
strategy and cognitive style of field dependence status on learning from mediated
instructions. (ERIC Document Reproduction Service ED 428 758).
Intraub, H. (1999). Understanding and remembering briefly glimpsed pictures:
Implications for visual scanning and memory. In V. Coltheart (Ed.), Fleeting
memories: Cognition of brief visual stimuli (pp. 47-94). Cambridge, MA: The
MIT Press.

131
Jankowski, G., & Fuchs, D. (1995). Television today and tomorrow: It won't be what you
think. New York: Oxford University Press.
Jonassen, D. H. (1996). Computers in the classroom: Mindtools for critical thinking.
Englewood Cliffs, NJ: Prentice-Hall.
Joyce, B., Weil, M., & Calhoun, E. (2000). Models of teaching (6th ed.). Boston: Allyn
and Bacon.
Kagan, J. (1965). Impulsive and reflective children: Significance of cognitive tempo. In J.
D. Krumboltz (Ed.), Learning and the Educational Process. Chicago: Rand
McNally.
Kagan, J. (1966). Reflection-impulsivity: The generality and dynamics of conceptual
tempo. Journal of Abnormal Psychology, 71, 17-24.
Kahneman, D. (1973). Attention and effort. Englewood Cliffs, NJ: Prentice Hall, Inc.
Kay, A. (1999). User interface: A personal view. In B. Laurel (Ed.), The Art of Human-
Computer Interface Design (13th ed., pp. 191-207). Reading, MA: Addison-
Wesley.
Keller, J. M. (1983). Motivational design of instruction. In C. M. Reigeluth (Ed.),
Instructional Theories and Models: A Cherview of Their Current Status New
York: Lawrence Erlbaum Associates.
Keller, P. F. G. (1976). Pictorial memory processes under conditions of the kinestasis
fdm. (Doctoral Dissertation, Southern Illinois University, 1976). Dissertation
Abstracts International, 38 (05), 2498.
Kini, A. S. (1994). Effects of cognitive style and verbal and visual presentation modes on
concept learning in CBI. Paper presented at the Annual Meeting of the American
Educational Research Association. (ERIC Document Reproduction Service No.
ED 371 032), New Orleans.
Koffka, K. (1935). Principles of gestalt psychology. Reprint 1963. New York: Harcourt,
Brace & World.
Kogan, N. (1971). Educational implication of cognitive styles. In G. S. Lesser (Ed.),
Psychology and Educational Practice Glenview, IL: Scott Foresman.
Kozma, R. B. (1986, Spring). Implications of instructional psychology for the design of
educational television. Educational Communications and Technology Journal. 34,
11-19.

132
Krendl, K. A., & Watkins, B. (1983). Understanding television: An exploratory inquiry
into the reconstruction of narrative content. Educational Communication and
Technology Journal, 31, 201-212.
Ksobiech, K. J. (1976). The importance of perceived task and type of presentation in
student response to instructional television. A V Communication Review, 24, 401 -
412.
Lang, A. (Ed.). (1994). Measuring psychological responses to television. Hillsdale, N. J.:
Lawrence Erlbaum Associates.
Lang, A. (1996). Negative video as structure: Emotion, attention, capacity, and memory.
Journal of Broadcasting and Electronic Media, 40(4), 460-477.
Lang, A., & Basil, M. (1998). Attention, resource allocation, and communication
research: What do secondary task reaction times measure anyway? In M. Roloff
(Ed.), Mass Communication Yearbook (Vol. 21, pp. 443-474). Beverly Hills, CA:
Sage.
Lang, A., Bolls, P., Potter, R., & Kawahara, K. (1999, Fall). The effects of production
pacing and arousing content on the information processing of television messages.
Journal of Broadcasting & Electronic Media, 43(4), 451-468.
Lang, A., Zhou, S., Schwartz, N. Bolis, P., & Potter, R. (2000). The effects of edits on
arousal, attention, and memory for television messages: When and edit is an edit,
can an edit be too much. Journal of Broadcasting & Electronic Media, 44(1), 94-
109.
Langer, E. J., Blank, A., & Chanowitz, B. (1978). The mindlessness of ostensibly
thoughtful action: The role of "placebic" information in interpersonal interaction.
Journal of Personality and Social Psychology, 36, 635-642.
Langer, E. J., & Imber, L. G. (1979). When practice makes imperfect: Debilitating effects
of over-learning. Journal of Personality and Social Psychology, 37, 2014-2024.
LeDoux, J. (1991). Emotion and the limbic system concept. Concepts in Neuroscience,
2(2), 169-199.
Leino, J. (1981). Psychometric test theory and cognitive processes: A theoretical scrutiny
and empirical research. Research Bulletin No. 57, Helsinki University, Institute of
Education, Finland. (ERIC Document Reproduction Service. No. ED 223 672).
Loftus, G. R., & Kallman, H. J. (1979). Encoding and use of detail information in picture
recognition. Journal of Experimental Psychology: Human Learning and Memory,
5(3), 197-211.

133
Long, G. M. (1982). Interaction of arousal and task difficulty upon perceptual restriction.
Perceptual and Motor Skills, 32(1), 259-264.
Luckett, J. D. (1996). Music preparation and the good language learner. (Doctoral
Dissertation, University of Central Florida, 1996). Dissertation Abstracts
International, 57 (10), 4296A.
Marr, D. (1982). Vision. New York: Freeman Press.
Mayer, R. E., Bove, W Bryman, A., Mars, R., & Tapangco, L. (1996). When less is
more: Meaningful learning from visual and verbal summaries of science textbook
lessons. Journal of Educational Psychology, <3<3(1), 64-73.
McCollum, J. F., & Bryant, J. (1999). Pacing in children's television programming. Mass
Communication & Soc/e(y(March), 2-40.
McLuhan, M. (1964). Understanding media: The extensions of man. New York.
Merrinboer, J. J. G. (1990). Instructional strategies for teaching computer programming:
Interactions with the cognitive style reflection-impulsivity. Journal of Research
on Computing in Education, 23(1), 45-53.
Messer, S. B. (1970). Reflection-impulsivity: Stability and school failure. Journal of
Educational Psychology, 61, 467-490.
Messick, S. (1970). The criterion problem in the evaluation of instruction: Assessing
possible, not just intended outcomes. In D. E. W. M. C. Wittrock (Ed.), The
evaluation of instruction: Issues and problems New York: Holt Rinehart.
Meyrowitz, J. (1985). No sense ofplace: The impact of electronic media on social
behavior. New York: Oxford University Press.
Miller, P. H & Bjorkland, D. F. (1998). Contemplating fuzzy-trace theory: The gist of it.
Journal of Experimental Child Psychology, 71, 184-193.
Moreno, R., & Mayer, R. E. (2000). Coherence effect in multimedia learning: The case
for minimizing irrelevant sounds in the design of multimedia instructional
messages. Journal of Educational Psychology, 92 (l), 117-125.
Murch, G. M. (1973). Visual and Auditory Perception. Indianapolis, IN: Bobbs-Merrill
Company.
Murdock, B. B. (1982). Recognition memory. In C. R. Puff (Ed.), Handbook of Research
Methods in Human Memory and Cognition, (pp. 2-27). New York: Academic
Press.

134
Nazarro, J. R., & Nazarro, J. N. (1970). Auditory versus visual learning of temporal
patterns. Journal of Experimental Psychology, 84 (3), 472-478.
Nelson, D. D. (1990). Personal tempo as a consideration in the rhythmic training of first
grade students. (Doctoral Dissertation, University of Florida, 1990). Dissertation
Abstracts International, 51 (10), 3356A.
Neuman, O., & Prinz, W. (1990). Relationships between perception and action. New
York: Springer-Verlag.
Neuman, W. R. (1976). Patterns of recall among television news viewers. Public Opinion
Quarterly, 40, 115-123.
Nugent, G. C. (1982). Pictures, audio, and print: Symbolic representation and effect on
learning. Education Communications and Technology Journal, 30, 163-174.
O'Brien, C. R. (1968). Variables of perception in selected individuals viewing
progressively accurate visual stimuli. (Doctoral Dissertation, University of
Florida, 1990). Dissertation Abstracts International, 30 (01), 0173A.
Okun, M. A., Callistus, W. M., & Knoblock, L. B. (1979). Adult age differences in
cognitive tempo. Paper presented at the Annual Meeting of the American
Educational Association, San Francisco.
Olson, D. R. (1974). Introduction. In D. R. Olson (Ed.), Media and symbols: The forms of
expression, communication, and education. The Seventy-Third Yearbook of the
National Society for the Study of Education. Chicago: University of Chicago
Press.
Ong, W. J. (1982). Orality and literacy: The technology of the word. London: Routledge.
Paivio, A. (1986). Mental representation: A dual coding approach. Oxford, England:
Oxford University Press.
Patel, A. D., Peretz, I, Tramo, M., & Labreque, R. (1998, January). Processing prosodic
and musical patterns: A neuro-psychological investigation. Brain and language,
(5/(1), 123-144.
Pearl, D. (1982). Television and behavior: Ten Years of scientific progress and
implications for the eighties. Volume 1: Summary Report. Rockville, MD: U.S.
Department of Health and Human Services.
Peck, K. L. (1987). Adaptive external pacing as an alternative to self-pacing in
computerized instruction. (Doctoral Dissertation, University of Colorado, 1987).
Dissertation Abstracts International, 48 (05), 1105A.

135
Pezdek, K. (1977). The cross-modality semantic integration of sentence and picture
memory. Experimental Psychology: Human Learning and Memory, 3, 515-524.
Plass, J. L., Chun, D. M, & Mayer, R. E. (1998). Supporting visual and verbal learning
preferences in a second-language multimedia learning environment. Journal of
Educational Psychology, 90(1), 25-36.
Posner, M. I. (1982). Cummulative development of attentional theory. American
Psychologist, 37, 168-179.
Postman, N. (1986). Amusing ourselves to death: Public discourse in the age of show
business. London: Heineman.
Ramachandran, V. S. (1994). 2-D or not 2-D: That is the question. In H. R.I. Gregory, J.
(Ed.), The Artful Eye (pp. 249-267). Oxford, England: Oxford University Press.
Reeves, B., Thorson, E. & Schleuder, J. (1986). Attention to television: Psychological
theories and chronometric measures. In J. Bryant, & D. Zillman (Ed.),
Perspectives on media effects (pp. 251-279). Hillsdale, NJ: Lawrence Erlbaum
Associates.
Reeves, B., & Geiger, S. (1994). Designing experiments that assess psychological
responses to media messages. In A. Lang (Ed.), Measuring Psychological
Responses to Media (pp. 243). Hillsdale, NJ: Lawrence Erlbaum Associates.
Reeves, B., & Nass, C. (1996). The media equation: How people treat computers,
television, and new media like real people and places. New York: CSLI
Publications.
Rettenbach, R. (1999). Do deaf people see better? Texture augmentation and visual
search compensate in adult but not in juvenile subjects. Journal of Cognitive
Neuroscience, 11(5), 560-580.
Reyna, V. F., & Kieman, B. (1994). Development of gist versus verbatim memory in
sentence recognition: Effects of lexical familiarity, semantic content, encoding
instructions, and retention interval. Developmental Psychology, 20(2), 178-191.
Ridberg, E. H., Parke, R. D., & Hetherington, E. M. (1970). Modification of impulsive
and reflective cognitive styles through observation of film-mediated models.
Journal of Experimental Psychology, 5(3), 369-377.
Salcedo, A. M. (1985). A rhetorical and structural analysis of instructional television .
Anaheim, CA: Association for Educational Communications and Technology.
(ERIC Document Reproduction Service No. 256 332).

136
Salkind, N. J., & Wright, J. C. (1977). The development of reflection-impulsivity and
cognitive efficiency. Human Development, 20, 377-387.
Salomon, G. (1979). Interaction of media, cognition, and learning. San Francisco: Josey-
Boss.
Salomon, G. (1984). Television is "easy" and print is "tough": The differential investment
of mental effort in learning as a function of perceptions and attributions. Journal
of Educational Psychology, 76, 647-658.
Salomon, G. (1994). Interaction of media, cognition, and learning. Hillsdale, NJ:
Lawrence Erlbaum Associates.
Salomon, G., & Cohen, A. (1977). Television formats: Mastery of mental skills and the
acquisition of knowledge. Journal of Educational Psychology, 69, 612-619.
Schale, F. (1971, December). Measuring degree and rate of visual awareness in rapid
reading on television. Paper presented at the National Reading Conference,
Tampa, FL.
Schank, R. C., & Abelson, R. P. (1977). Scripts, plans, goals, and understanding.
Hillsdale, NJ: Lawrence Erlbaum Associates.
Seidman, S. A. (1981, Spring). On the contributions of music to media productions.
Educational Communications and Technology Journal, 29.
Seward-Barry, A. M. (1997). Visual intelligence. Albany: State University of New York
Press.
Shaffer, R. J., Greenspan, S. I., Tuchman, R. F., Cassily, J. F., Jacokes, L. E. & Stemmer,
P. J. (2000, August 19). Interactive metronome: Effects on motor control,
concentration, control of aggression, and learning in children with attention-
deficit/hyperactivity disorder. Paper presented at the Progress in Motor Control II
Congress, Penn State University, State College.
Shalev, L., & Algom, D. (2000). Stroop and Gamer effects in and out of Posner's beam:
Reconciling two concepts in selective attention. Journal of Experimental
Psychology, 26(1), 997-1017.
Sheingold, K. (1973, August). Developmental differences in intake and storage of visual
information. Journal of Experimental Child Psychology, 16, 1-11.
Sheppard, R. N. (1967). Recognition memory for words, sentences and pictures. Journal
of Verbal Learning & Verbal Behavior, 6, 156-163.

137
Singer, J. L. (1980). The power and limitations of television: A cognitive-affective
analysis. In P. L. Tannenbaum (Ed.), The entertainment functions of television .
Hillsdale, NJ: Lawrence Erlbaum Associates.
Smith, E. E., & Haviland, S. E. (1972). Why words are perceived more accurately than
non-words: Inference versus unitization. Journal of Experimental Psychology, 92,
59-64.
Sperling, G. (1963). A model for visual memory. Human Factors, 5, 5-31.
Stephens, M. (1996). The rise of the image the fall of the word. New York: Oxford
University Press.
Sterrit, G. M., Camp, B., & Lipman, B. S. (1996). Effects of early auditory deprivation
upon auditory and visual information processing. Perceptual and Motor Skills, 23,
123-130.
Thompson, A. D., Simonson, M. R., & Hargrave, C. P. (1996). Educational technology:
A review of the research. Washington, DC: Association for Educational
Communications and Technology.
Thorson, E., & Lang, A. (1992). The effects of television videographics and lecture
familiarity on adult cardiac orienting responses and memory. Communication
Research, 19(3), 346-349.
Tovee, M. J. (1998). The speed of thought: How fast can you see? In M. J. Tovee (Ed.),
The speed of thought: Information processing in the cerebral cortex (pp. 143-
152). New York: Springer-Verlag.
Tse, T., Vegh, S., Marchionini, S., & Shneiderman, B. (1999). An exploratory study of
video browsing user interface designs and research methodologies: Effectiveness
in information seeking tasks. Paper presented at the 62nd ASIS Annual Meeting,
Washington, DC.
Tulving, E. (1972). Episodic and semantic memory. In W. E. Tulving & Donaldson
(Ed.), Organization of Memory (pp. 381-403). New York: Academic Press.
Tyner, K. (1998). Literacy in a digital world: Teaching and learning in the age of
information. Mahwah, NJ: Lawrence Erlbaum Associates.
Verhagen, P. W. (1992). Preferred length of video segments in interactive video
programs. Proceedings of Selected Research and Development Presentations at
the Convention of the Association for Educational Communications and
Technology. (ERIC Document Reproduction Service No. ED 348 036).

138
Voss, J. F., Tyler, S. W & Bisanz, G. L. (1982). Prose comprehension and memory. In
C. R. Puff (Ed.), Handbook of research methods in human memory and cognition
(pp. 349-392). New York: Academic Press.
Wagely, M. (1978). The effect of music on affective and cognitive development of sound-
symbol recognition among preschool children. Unpublished Doctoral
Dissertation, Texas Woman's University, Denton.
Walma van der Molen, J. H., & Van der Voort, T. H. A. (2000). The impact of television,
print, and audio on children's recall of the news: A study of three alternative
explanations of the dual coding hypothesis. Human Communication Research, 26,
3-26.
Walters, D. L. (1983). The relationship between personal tempo in primary-aged
children and their ability to synchronize movement with music. Dissertation
Abstracts International, 44 (02), 423A.
Watkins, J. M., Lee, H. B., & Erlich, O. (1978). The generalizability of the matching
familiar figures test. Paper presented at the Annual Meeting of the American
Educational Research Association (ERIC Document Reproduction Service No.
ED 175 882), Toronto, CA.
Watt, J. H., & Krull, R. (1977). An examination of three models of television viewing
and aggression. Human Communication Research, 3, 99-112.
Wertheimer, M. (1959). Productive Thinking (Enlarged Ed.). New York: Harper &
Rowe.
Wetzel, C. D., Radtke, P. H., & Stem, H. W. (1994). Instructional effectiveness of video
media. Hillsdale, NJ: Lawrence Erlbaum Associates.
Winn, W. (1982). Visualization in learning and instruction: A cognitive approach.
Educational Communications and Technology Journal, 30, 3-25.
Witkin, H. A., & Goodenough, D. R. (1981). Cognitive styles: Essence and origins. Field
dependence and field independence. New York: International Universities Press.
Wittrock, M. (1979). The cognitive movement in instruction. Educational Researcher, 8,
5-11.
Woodley, J. W. (1984). The role of non-visual information in visual perception: An
integration of research perspectives Austin, TX: (ERIC Document Reproduction
Service No. ED 283 134).

139
Worth, S., & Adair, J. (1997). Through Navajo eyes: An exploration in film
communication and anthropology (2nd Edition). Albuquerque: University of New
Mexico Press.
Wright, J. C. (1979). Changes in conceptual tempo over time: Developmental versus
logical familiarization. Paper presented at the Bicentennial Meeting of the Society
for Research in Child Development. March 15-18, San Francisco.
Yerkes, R. M., & Dodson, J. D. (1908). The relationship of strength of stimulus to
rapidity of habit formation. Journal of Neurological Psychology, 18, 459-482.
Zillman, D. (1991). Television viewing and physiological arousal. In J. Z. Bryant, D.
(Ed.), Responding to the screen: Reception and reaction processes (pp. 103-133).
Hillsdale, NJ: Lawrence Erlbaum Associates.
Zillman, D. & Brosius, H. (2000). Exemplification in communication. Mahwah, NJ:
Lawrence Erlbaum Associates.

BIOGRAPHICAL SKETCH
Mr. Kenny holds a bachelors degree in Education from Niagara University in
New York and a masters degree in instructional technology/media: educational
technology from the University of Central Florida. Prior to entering into the doctoral
program at the University of Florida, he spent more than 20 years in the computer
services industry and in private business. He has spent the last six years teaching media
studies and technology on the secondary level in Orlando, Florida.
Mr. Kenny is an active member in the Florida Association of Media Educators
(FAME), and the Association of Educational Communications and Technology (AECT)
and has presented several papers at conferences sponsored by both of these organizations.
Mr. Kenny also published a textbook in 2001, entitled Teaching Television
Production in the Digital Age, a book dedicated to showing new teachers how to establish
television and media studies programs in their schools.
Mr. Kenny has been married to his wife, Sandra, for thirty years. He has two
married children.
140

I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy
¡O/. Yidm*
Lee J. Mlally, Chair /
Associate Professor of Teaching and
Learning
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy
Jef
Associqre Professor of Teaching and
Learning
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of Philosophy
M. David Miller,
Professor of Educational Psychology
I certify that I have read this study and that in my opinion it conforms to
acceptable standards of scholarly presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor (/fPhilosophy
David Ostroff
Professor of Journalism and
Communications
This dissertation was submitted to the Graduate Faculty of the College of
Education and to the Graduate School and was accepted as partial fulfillment of the
requirements for the degree of Doctor of Philosophy.
May, 2002
Dean, Graduate School

LD
1780
20 CjL
X\^
UNIVERSITY OF FLORIDA
3 1262 08557 2450