• TABLE OF CONTENTS
HIDE
 Title Page
 Acknowledgement
 List of Tables
 Abstract
 Introduction
 Review of the literature
 Conclusions from the literature...
 Methodology
 Analysis of the data
 Summary and conclusions
 Appendices
 References
 Biographical sketch














Group Title: effect of racial prejudice on visual depth perception
Title: The effect of racial prejudice on visual depth perception
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00097503/00001
 Material Information
Title: The effect of racial prejudice on visual depth perception an examination of perceptual defensesensitization
Physical Description: ix, 121 leaves : ill. ; 28cm.
Language: English
Creator: Goldstein, Mark Lloyd, 1948-
Publication Date: 1976
Copyright Date: 1976
 Subjects
Subject: Race discrimination -- Psychological aspects   ( lcsh )
Visual perception   ( lcsh )
Foundations of Education thesis Ph. D   ( lcsh )
Dissertations, Academic -- Foundations of Education -- UF   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Thesis: Thesis--University of Florida.
Bibliography: Bibliography: leaves 111-119.
Additional Physical Form: Also available on World Wide Web
General Note: Typescript.
General Note: Vita.
Statement of Responsibility: by Mark Lloyd Goldstein.
 Record Information
Bibliographic ID: UF00097503
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000170137
oclc - 02917994
notis - AAT6551

Downloads

This item has the following downloads:

PDF ( 4 MBs ) ( PDF )


Table of Contents
    Title Page
        Page i
        Page i-a
    Acknowledgement
        Page ii
        Page iii
        Page iv
    List of Tables
        Page v
        Page vi
    Abstract
        Page vii
        Page viii
        Page ix
    Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
    Review of the literature
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
    Conclusions from the literature review and hypotheses generated
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
    Methodology
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
    Analysis of the data
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
    Summary and conclusions
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
    Appendices
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
    References
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
    Biographical sketch
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
Full Text








THE EFFECT OF RACIAL PREJUDICE ON VISUAL DEPTH PERCEPTION:
AN EXAMINATION OF PERCEPTUAL DEFENSE/SENSITIZATION













By

MARK LLOYD GOLDSTEIN


ROTATIONN PRESENTED TO THE GRADUATE COUNCIL OF THE
'RSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
:REMENTS FOR THE DEGREE OF DOCTOR OF- PHILOSOPHY








UNIVERSITY OF FLORIDA

1976












ACKNOWLEDGrENTS

The author wishes to gratefully acknowledge the

participation and cooperation of all the people who in

various ways helped in the development and preparation of

this document. He is indebted to the entire faculty who,

as teachers, provided the knowledge and skills necessary

to undort!ke and complete this task. In particular, he

is beholden to his supervisory committee members.

Dr. Barry Guinagh and Dr. Hannelore Wass, in their

roles as co-chairmen of the supervisory committee, provided

invaluable assistance in guiding the direction of this

study as well as critical comments and suggestions in

making the final product a worthy one.

Dr. Hattie Bessent and Dr. Robert Ziller not only

provided technical assistance and comments but also provided

warm, supportive counseling and friendship throughout the

dissertation period.

Finally, and most importantly, the author wishes to

thank his wife, Janis. for her constant love, devotion,

understanding patience, as well as assistance throughout

the years of his doctoral studies.






TABLE OF CONTENTS

ACKNOWLEDGEMENTS . . . . . . . . .

LIST OF TABLES . . . . . . . . .

ABSTRACT . . . . . . . . . . .

Chapter
I. INTRODUCTION . . . . . . . . .

Definitions . . . . . . . . .

Purpose of the Study . . . . . . .

II. REVIEW OF THE LITERATURE . . . . . .

III. CONCLUSIONS FROM THE LITERATURE REVIEW
AND HYPOTHESES GENERATED . . . . .

IV. METHODOLOGY . . . . . . . . .

Subjects . . . . . . . . . .

Instrument Selection . . . . . . .

Experimental Procedure . . . . . .

Stimuli . . . . . . . . . .

Starting Position . . . . . . .

Recording Device . . . . . . . .

Statistical Procedure . . . . . .

V. ANALYSIS OF THE DATA . . . . . . .

VI. SUM1,MARY AND CONCLUSIONS . . . . . .

The Sample . . . . . . . . .

Treatment of the Data . . . . . .

Results . . . . . . . . . .

Conclusions . . . . . . . . .

Recommendations for Further Study . . .


ii

v

vii


IIL






TABLE OF CONTENTS continued


Appendix

1. STIMULI EMPLOYED . . . . . . . .

2. SELECTION OF RACIALLY-CONNOTATIVE STIMULI . .

3. ORAL STATEMENT TO SUBJECTS COMPLETING
THE SOCIAL STIUATIONS QUESTIONNAIRE . . .

4. MODIFIED SOCIAL SITUATIONS QUESTIONNAIRE .

5. IDENTIFICATION SHEET . . . . . . .

6. EXPERIMENT RECORDING SHEET AND EXAMPLES OF
INDIVIDUAL RESPONSES OF HIGH AND LOW
PREJUDICED SUBJECTS . .. . . . .


LIST OF REFERENCES . . . . . . .


BIOGRAPHICAL SKETCH . . . . . . . . .


96


98

100

107



109


111


120











LIST OF TABLES


Table
1. CLASSIFICATION OF THEORIES IN PERCEPTUAL
DEFENSE/SENSITIZATION . . . . .. 44

2. MAJOR METHODOLOGICAL INADEQUACIES OF
PERCEPTUAL DEFENSE/SENSITIZATION STUDIES. 49

3. FREQUENCY DISTRIBUTION FOR THE RACIAL ATTITUDE
SURVEY . . . . . . . . . 66

4. DISTANCE MEANS OF THE RACIALLY CONNOTATIVE
WORDS AND THEIR CONTROL WORDS FOR THE
HIGH PREJUDICED, AVERAGE PREJUDICED, AND
LOW PREJUDICED GROUPS . . . . .. 67

5. DISTANCE MEANS OF THE EGO ALIEN WORDS AND
THEIR CONTROL WORDS FOR THE HIGH PREJUDICED,
AVERAGE PREJUDICED, AND LOW PREJUDICED GROUPS.. 68

6. ANALYSIS OF VARIANCE OF RACIALLY CONNOTATIVE
WORDS BETWEEN HIGH PREJUDICED, AVERAGE
PREJUDICED AND LOW PREJUDICED GROUPS . . 69

7. ANALYSIS OF VARIANCE OF RACIALLY CONNOTATIVE
CONTROL WORDS BETWEEN HIGH PREJUDICED,
AVERAGE PREJUDICED AND LOW PREJUDICED
GROUPS . . . . . . . . . 70

8. ANALYSIS OF VARIANCE OF EGO ALIEN WORDS BETWEEN
HIGH PREJUDICED, AVERAGE PREJUDICED AND
LOW PREJUDICED GROUPS . . . . .. 70

9. ANALYSIS OF VARIANCE OF EGO ALIEN CONTROL
WORDS BETWEEN HIGH PREJUDICED, AVERAGE
PREJUDICED AND LOW PREJUDICED GROUPS . . 71

10. DISTANCE MEANS OF RACIALLY CONNOTATIVE WORDS
AND THEIR CONTROL WORDS BETWEEN HIGH
PREJUDICED, AVERAGE PREJUDICED, AND LOW
PREJUDICED GROUPS FOR HALES AND FEMALES . 72







LIST OF TABLES continued


Table
11. DISTANCE MEANS OF EGO ALIEN WORDS AND THEIR
CONTROL WORDS BETWEEN HIGH PREJUDICED,
AVERAGE PREJUDICED AND LOW PREJUDICED
GROUPS FOR MALES AND FEMALES . . . .. 73

12. DISTANCE MEANS OF RACIALLY CONNOTATIVE
WORDS AND THEIR CONTROL WORDS AND
EGO ALIEN WORDS AND THEIR CONTROL WORDS
BETWEEN NEAR AND FAR STARTING POSITIONS
FOR THE HIGH PREJUDICED GROUP ...... .75

13. DISTANCE MEANS OF RACIALLY CONNOTATIVE
WORDS AND THEIR CONTROL WORDS AND
EGO ALIEN WORDS AND THEIR CONTROL WORDS
BETWEEN NEAR AND FAR STARTING POSITIONS
FOR THE AVERAGE PREJUDICED GROUP . . .. 76

14. DISTANCE MEANS OF RACIALLY CONNOTATIVE
WORDS AND THEIR CONTROL WORDS AND
EGO ALIEN WORDS AND THEIR CONTROL WORDS
BETWEEN NEAR AND FAR STARTING POSITIONS
FOR THE LOW PREJUDICED GROUP . . . .. 77











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


THE EFFECT OF RACIAL ATTITUDE ON VISUAL DEPTH PERCEPTION:
AN EXAMINATION OF PERCEPTUAL DEFENSE/SENSITIZATION

By

Mark Lloyd Goldstein

June 1976

Chairman: Dr. Barry Guinagh
Major Department: Foundations of Education

The problem to which this dissertation is addressed,

and which has concerned Bruner and Postman, Combs and

Snygg and numerous others, is whether or not personal

values and attitudes influence perceptual processes.

Specifically, this paper examines the relationship between

racial attitude and visual depth perception.

Eighty-four Caucasian high school students were

administered a modified Social Situations Questionaire.

From the results of this survey, students were chosen

for the experimental facet of the study. Two experimental

groups -- high prejudiced (one standard deviation or

more above the mean on the attitude survey) and low


vii






prejudiced (one standard deviation or more below the mean) --

along with a control group (clustered around the mean) were

employed.

Each subject, after being dark adapted for approximately

five minutes, was seated in front of the experimental

apparatus. For five seconds, the subject observed a word

painted in luminescent paint. After the stimulus was re-

moved, a luminescent marker placed on a track below the

word was shown to the subject. The subject was required

to tell the examiner whether to move the marker closer to

or further away from himself so that the marker was "directly

below the spot where the word appeared."

Unknown to the subject, the word was always positioned

at a distance of four feet in the median plane from the

slot through which the subject looked.

Two words were presented to the subject which were thought

to be ego alien to each subject (e.g. PIG). For each ego

alien word, there was a neutral control word presented which

was perceptually similar except for a change in one letter

(e.g. PIN). Three words were also presented which were

thought to be racially connotative to each subject (e.g.

NIGGER). Again, perceptually similar control words were

presented (e.g. BIGGER). The starting position of the

luminescent marker (closer to the subject than the word or


viii





farther away from the subject than the word) was randomly

varied on two trials. Distance was measured from this

zero point.

There are three major conclusions generated from this

study. First, it was found that racial attitude does, in

fact, influence visual depth perception. Secondly, this

disseratation provides support for the perceptual defense

position and questions previous research which demonstrated

a sensitization or duality effect. Finally, the present

research provides credence for Snygg and Comb's phenomenono-

logical theory and questions the learning theory position.












CHAPTER I
INTRODUCTION

Over 25 years ago, a paper by Bruner and Postman

(1947a) initiated one of the most significant controversies

in the study of perception. The paper, which suggested

that the perception of external stimuli is not independent

of the influences of internal events, began what has come

to be known as the "new look" view in perception. With

the advent of the "new look" in perception, also came

two major problems. The first problem centered around

three major theoretical positions and the concommitant

conflict even within each theoretical camp. The second

problem focused on methodology and the lack of controls

employed in perceptual research. Numerous publications

since the Bruner and Postman article have attempted to

explore if and how attitudes, values, expectancies,

needs and psychodynamic defenses impinge upon perception.

The problem to which this dissertation is addressed is

whether or not personal attitudes influence perceptual

processes. In particular, the purpose of this paper is

to examine the effect of racial attitude among Caucasian

high school students on visual depth perception.







Definitions

Historically, psychologists have used the word "perception"

to refer only to "a single unified meaning obtained from

sensory processes while a stimulus is present" (Bartley, 1969,

p. 15). Early psychologists such as E. B. Tichener and

W. Wundt viewed "perception" as the result of learning added

to raw sensations, while Ames' transactional approach stated

that perception is developed by each individual through his

own unique transactions with the environment. Perception

becomes a learned act of constructing reality to fit one's

assumptions about it. Cognitive field psychologists think

that perception not only involves what an individual senses

and feels, but also the behavior associated with the senses

and feelings. Although the various definitions differ on

the basis of their respective theoretical frameworks, the

definition of perception in Benjamin Wolman's Dictionary of

Behavioral Science seems to be general enough to encompass

their commonality. The working definition for perception in

this dissertation will be: "the process of obtaining informa-

tion about the world through the senses" (Wolman, 1972, p. 386).

Therefore "visual perception" as employed in this study

will refer to the process of obtaining information about the

world through vision. More specifically, visual depth perception

as defined, for this dissertation is defined as the perception of




3

distance between the stimulus and the subject through

the visual sense.

Because the perceptual system is made up of all the

sensory mechanisms, it is conceivable that several sense

modalities may be involved as they act together to meet

environmental demands and goals of the organism. For

example, during the act of eating an individual may employ

four of his senses which contribute to the totality of his

perception. Attempts to dissect the weight of each sense

in the perception are doomed to failure because of the lack

of scientific precision at this time.

Just as it is difficult to determine the relative

weight of one sense within a perception, it is just as

difficult to separate and distinguish between sensory,

cognitive and social perception.

Social perception is viewed as "something crucially

determined by the specific character or state of the indi-

vidual perceiver rather than being totally stimulus bound.

However, sensory mechanisms are still integrally involved in

social perception" (Bartley, 1969, p. 74). In cognitive

perception, defined as the process by which information is

acquired through experience and becomes part of the organism's

storage of facts (Fergus, 1966), both sensory and social per-

ception play integral roles. In analyzing a learning task for







example, the sensory channels employed (sensory perception)

as well as the state of the individual perceiver (social

perception) both have roles in determining the total

perception.

Although the terms "cognitive perception, social per-

ception, and sensory perception" connote three different

forms of perception, in reality one does not exist without

the others. As in attempts to delineate between perception,

learning and thinking, attempts to separate forms of per-

ception becomes a matter of semantics. Therefore, for the

purposes of this study, perception will encompass cognitive,

social, and sensory perception.

Just as authorities differ as to what constitutes

"perception" there also appears to be a lack of semantical

agreement on "perceptual defense." Perceptual defense is

one manifestation of the general idea derived from Freudian

theory that a defensive process should produce observable

effects in any cognitive situation when the individual's

security or self esteem is threatened and there is an

opportunity for him to reduce the affective disturbance. In

the personality theory of Snygg and Combs (1959) as well as

that of Rogers (1951), defensiveness is described as primarily

a perceptual phenomonenon which follows as a consequence of

threat to the individual's self. Defense, in essence, is




5


inaccurate perception which prevents that which is threat-

ening fro:n reaching conscious awareness. Therefore, aspects

of the environment may be denied awareness or may be mis-

perceived by the individual. Furthermore, the adequacy of

the individual's personal adjustment is considered to be

inversely related to the degree to which experiences are

denied awareness.

Although "perceptual defense" has been adopted as the

nomenclature to describe inaccurate perception, in reality

two phenomena are encompassed by the term -- pcrceptual

defense and perceptual sensitization or vigilance. Perceptual

defense occurs when the subject's recognition threshold for

an emotional stimulus is greater than for neutral stiiTali,

whereas perceptual sensitization or vigilance occurs \whe:n the:

subject's recognition threshold for an emotional stjimul3us is

less than for neutral stimuli. In other words, whe.n presente.iJ

with an emotionally laden stimulus and a neutral stimulus,

defense occurs when the subject requires, for example, longer

exposure time for the emotionally laden stimulus in order to

recognize the stimulus; vigilance occurs when a shorter exposu'ic

time for the emotionally laden word is required for recognition.

For this dissertation, perceptual defense iill hereafter

refer only to the one proce::s and not encompass both defense

and sensitization. When referring to the general pe.ncmerno',

"perceptual defense/sensitizacton" will. be utilized.







Purpose of the Study

Within this general paradigm of perceptual defense/

sensitization much of the controversy has centered around

two issues -- experimental methodology and theoretical

formulation. With regards to methodology, experimental

rigidity has impeded progress and knowledge. Rigidity with

reference to perceptual defense/sensitization can be demon-

strated by the adherence to the psychophysical method of the

ascending Method of Limits in almost every study. Defining

perceptual defense/sensitization in relation to this methodology

places too much emphasis on negative factors, that is, the

not seeing of a stimulus. If the subject reports seeing the

neutral word either before or instead of the anxiety-provoking

stimulus, it is thought that he must be defending against

seeing the anxiety-provoking stimulus. In addition, negative

indicators are vulnerable because they yield little if any

information about the process involved, and there is no reason

to conclude that the failure to report seeing a word was indeed

caused by the word itself rather than by some other possibility.

In most of the perceptual defense/sensitization studies, the

only response available to the subject is to see or not see

the stimulus. Time is the major variable with the subject

seeing either the conflict stimulus or the neutral stimulus

first. One objective of the present study is to provide the

subject with a broader range of response alternatives.







The conglomeration of theoretical positions revolving

around perceptual defense/sensitization has been a second major

area of controversy. The perceptual defense/perceptual

sensitization dynamic will be explored in the review of the

literature in Chapter Two as it pertains to a number of theories

including Howe's word probability theory, Coldiamond's response

bias theory, Bruner and Postman's hypothesis theory, learning

theories, and Neisser's information processing theory. In

addition, two conflicting theories relating racial attitude

to perceptual defense will be delineated. A second objective

of this dissertation will be to shed additional light upon

the present theoretical chaos and add support to one or more

positions.

The confusion surrounding theoretical postulates has become

increasingly complex as new theories are added to the ever-

growing number. There appear to be three major positions -- those

theories which espouse only perceptual defense, those theories

which support only perceptual sensitization, and those theories

which adhere to a dual process of defense and sensitization.

To muddle the picture even further, there are serious disagree-

ments within each theoretical camp. Learning theorists and

phenomenologists, for example, both posit a perceptual defense

approach to the problem, yet view the defense process in very

different perspectives, as might be expected.







Compounding the present picture is the fact that almost

all of the theories and studies have pertained almost exclu-

sively to the use of taboo or other ego alien stimuli and

have failed to delineate any personality characteristics of

the subjects which might influence the subject's response

patterns.

It seems logical that before this theoretical schizo-

phrenia can be remedied, research needs to be conducted

which 1) refines the methodogical inadequacies of previous

research and 2) initiates an investigation of the effect of

a personality variable upon the subject's response patterns.

Keeping these two guidelines in mind, the present research

has attempted to attack these two weaknesses and thereby

provide more insight into the present disagreement.












CHAPTER II
REVIEW OF THE LITERATURE

The "new-look" approach to perception made its appear-

ance on the psychological scene in articles by Bruner and

Postman (1947a, 1947b), McGinnies (1949), and Postman,

Bruner, and McGinnies (1948). In their initial study,

Bruner and Postman (1947a) observed differences in the

tachistoscopic duration necessary for the recognition of

threatening or emotional stimuli as opposed to neutral

stimuli. Subjects reported seeing neutral words prior to

reporting the emotionally charged words. Bruner and

Postman believed that the higher threshold for the emotion-

ally charged words represented a way in which the subjects

perceptually defended themselves against seeing these words,

and they labeled this process perceptual defense. In the

same research, the authors found a second perceptual process

which they called perceptual sensitization. In this phenom-

enon, the subjects reported seeing the emotionally laden words

prior to recognizing the neutral words.

In another study, Bruner and Postman (1947b) administered

a word association cest and subsequently studied the tachisto-

scopic duration necessary for the recognition of words with







long, medium, and short association times. For some subjects,

words with long association times (words indicating emotional

disturbances) required much longer durations for recognition

than words with medium or short association times (neutral

words). The authors termed this heightened recognition

threshold perceptual defense and likened it to the process

of repression whereby anxiety-provoking stimuli were defended

against in perception or prevented from attaining conscious

awareness is order to minimize anxiety. They also found,

however, that in certain subjects long association time

words had lower thresholds for recognition. The principle

of perceptual sensitization was invoked to account for this

lowering of thresholds for affect-laden words.

The concept of perceptual defense/sensitization also

appeared in subsequent articles by the same authors (Bruner

and Postman, 1948; Bruner and Postman, 1949; Postman and

Bruner, 1948) where the tachistoscopic durations necessary

for recognition of emotional words (words representing

low values as assessed by the Allport-Vernon Study of Values)

were found to have higher recognition thresholds than neutral

words (words representing high values as assessed by the

Allport-Vernon Study). Support for perceptual defense was

again invoked, although little evidence in this series of

studies was found for the perceptual sensitization position.







The entire problem area gained a great deal of momentum

following an article by McGinnies (1949). The perceptual

duration thresholds for recognition of a group of neutral

words and a group of taboo words such as whore, bitch, and

kotex were studied. McGinnies obtained the recognition

thresholds and concurrently measured the galvanic skin response

(CSR) during prerecognition and recognition trails. Not only

did he find that the sexually taboo words tended to require

higher durations for recognition but also that the subjects

had created GSR's on the taboo words than they did on the

neutral words. The higher recognition threshold for the taboo

words was considered to be a manifestation of perceptual

defense, and the greater GSR to the taboo stimuli was believed

to be not only an indication of the active nature of the

inhibition occurring in perception but also an unconscious

detection or manifestation of anxiety elicited by these

emotional words. Other investigators became interested in

this problem area and started the "dirty word" studies (Byrne

and Sheffield, 1964; Eriksen, 1951; Howes and Solomon, 1950;

Postman and Leytham, 1951; and Weiner, 1955), which have been

severely attacked for both methodological and theoretical

reasons. However, a more recent study attempting to control

for these weaknesses clearly demonstrated the perceptual

pehnomenon (Sauber, 1971). The taboo/neutral word difference







score attained significance beyond the .001 level, indicating

higher recognition thresholds for disagreeable words.

In apparent conflict with the Bruner and Postman

results (showing two perceptual processes), the dirty word

studies only found evidence of perceptual defense. Similar

results were found by Beier and Cowen (1953), Cowen and

Obrist (1958), and Zuckerman (1955). In his 1955 study,

Zuckerman measured recognition thresholds for neutral words

and for words with aggressive meanings and found that the

aggressive words had significantly higher recognition thresholds

than neutral words.

There have also been reports in the literature indicating

that emotional stimuli have had lower recognition thresholds

than neutral stimuli. Datson (1956) used words with nonsexual,

heterosexual and homosexual connotations and found lower

recognition thresholds for the homosexual stimuli. Eriksen

and Browne (1956) gave their subjects an experience of failure

on an anagram solution test. They found that the words for

which the subjects had been asked to find anagrams (words

assumed to be associated with failure) had lower recognition

thresholds than neutral words which had not been in the anagram

test.

Pustell (1957) and Chapman and Feather (1972) conditioned

an electric shock with certain stimuli and found a lower







recognition threshold for these stimuli than for stimuli

which did not have an electric shock attached.

Two different approaches to resolving the question

whether the recognition threshold is raised, lowered or

both raised and lowered when emotional stimuli are employed

have been delineated in the literature. One approach stresses

the variability in technique or methodology. Proponents of

this position have attempted to resolve the problem by searching

for variables which have led to defense, sensitization, or

both. A second approach, which has generated most of the

support in the literature stresses the variability within

the individual. Proponents of this approach have attempted

to find other variables within the personality which might

be correlated with one or another type of response. Early

studies by Carpenter, Wiener, and Carpenter (1956), Eriksen

(1951, 1952), Kates and Klein (1954), Lazarus, Eriksen, and

Fonda (1951), and Nelson (1955) found that personality and/or

intellectual differences in subjects had a significant effect

upon responses to emotional stimuli.

Three experimenters used the Minnesota Multiphasic

Personality Inventory (MMPI) in order to differentiate groups

for the study of personality correlates of perceptual defense/

sensitization. Greenbaum (1956) reported that high scorers

on the MMPI found photographs of faces with hostile expressions







easier to recognize than low scorers did. No differences

between the groups were found to photographs of friendly

faces.

Mathews and Wertheimer (1958) used two MMPI scales to

select their subjects. Their groups were high scorers on

the Hysteria (Hy) scale and high scorers on the Psychasthenia

(Pt) scale. Using associative reaction time as a criterion,

they selected neutral and emotional stimuli for each subject

individually and found a significant tendency for emotional

words to have higher thresholds than neutral words (perceptual

defense) in the Hy group. No such difference was found among

the high Pt scorers.

Porzemsky (1969) separated obese women into three groups

by means of the MMPI and subdivided each group into two sub-

groups based upon the results of the Repressor-Sensitizer (R-S)

subscale of the I.IPI. Employing five words which were regarded

as ego-alien to each of the three groups as well as neutral

words. Porzemsky discovered that the ego-alien words had

significantly higher recognition thresholds than the neutral

words for all three groups. No differences were found for

the Repressors versus Sensitizers on the R-S scale.

Although a number of studies have failed to support

Bruner and Postman's contention that two dichotomous phenomena --

perceptual defense and perceptual sensitization -- were present,







this postulate of two mechanisms gained wide acceptance in

the literature. Even a recent review of the literature in

the area by Erdelyi (1974) took this point of view.

Although the duality of the perceptual defense-
perceptual vigilence phenomenon is an interesting
problem on its own right, it is not of major concern
here since this duality never became a serious issue
in the critical literature. This relative lack of
concern can probably be attributed to the rather clear
and certainly massive demonstration that the specific
mode of reaction could be independently predicted... (p.7).

Whereas the perceptual defense/sensitization differentiation

has received little attention in the literature, other

problems in this area have been debated frequently.

One debate has centered around the methodology employed.

Perceptual defense/sensitization has been primarily and

almost exclusively studied by using the traditional psycho-

physical methodology of the ascending Method of Limits,

where early presentations are made under conditions which

prevent the subjects from adequately viewing the stimulus.

Gradually, some of the hinderances are removed until the

subject accurately reports the perceived stimulus. When

this point is reached, the recognition threshold is determined.

The apparatus most frequently employed in these experi-

ments has been the tachistoscope which presents stimuli

visually at the exposure speed designated by the experimenter,

who can also vary the amount of illumination. Thus, the







experimenter chooses either of two possible tachistoscopic

methodologies: varying the amount of illumination while

holding the exposure speed constant or varying the exposure

speed while holding the illumination constant.

Varying the exposure time is the most common method

used in perceptual defense/sensitization experiments (e.g.,
/
Freeman, 1954, 1955). With this method, both the critical

and neutral stimuli are initially presented at such rapid

speeds that the stimuli cannot be seen; slowly the speed of

presentation is reduced until the recognition threshold is

obtained. If the neutral words are recognized at a more

rapid exposure time than the critical words, the phenomenon

is entitled perceptual defense. If the opposite is true,

it is called perceptual sensitization or vigilence.

Other studies have varied the illumination during the

experiment (e.g., Spence, 1957). Sucessive presentations

take place under an illumination which is initially dim and

gradually increases in intensity until recognition occurs.

Each presentation has the same exposure speed, a period of

time too short to allow recognition at the lower illumination,

but long enough to permit recognition at higher illumination

levels. The level of illumination at which recognition occurs

provides a measure of the recognition threshold.

Berger (1956) varied the distance between the subject







and the stimulus, placing the stimulus too far from the

subject to be recognized and gradually moving it closer

until recognition occurred. The distance remaining between

the subject and stimulus served as a measure of recognition

threshold.

Instruments other than the tachistoscope have been

limited. One investigator exposed subjects to carbon

copies of typed words which ranged from the most smudged

and illegible to completely legible (Beier and Cowen, 1953).

The recognition threshold was determined by the number of

unrecognized words before the subject correctly recognized

the word.

Other studies have used the auditory modality. In

these either the acoustical equipment has been varied in

voltage (Vanderplas and Blake, 1949) or the tape recording

has been varied in intensity (Kleinman, 1957; Kurland, 1954).

All of the studies cited thus far have used the ascending

Method of Limits methodology to determine the recognition

threshold. A great difficulty with this method is that the

threshold is a statistical concept which requires the average

of a number of measurements since the threshold oscillates

constantly (depending on factors such as practice, fatigue,

etc.) However, in many of the experiments previously cited,

the threshold was determined on the basis of only one trial.







An additional problem with the methodology is the

restriction of the range of possible responses. The response

category is limited to whether the subject did or did not

see or hear the critical stimulus. As Garner, Hake, and

Eriksen (1956) state, "If the number of response categories

is two small to demonstrate the perceptual discrimination

capacity of a subject, then the outcome of the experiment

will be limited by a property of the response system rather

than by a property of the perceptual system" (p. 321).

Few studies have utilized a methodology other than

the ascending Method of Limits. Gordon (1957) presented

nonsense syllables in a light box in a dark-room situation

and asked the subjects to guess how far away the distant

stimuli were. Davis (1959) and Van de Castle (1960) in

separate studies utilized a stereoscopic methodology in which

critical words were presented to one eye while neutral words

were presented to the other eye. The subject's task was to

report which words he saw.

Another area of concern in the literature has centered

around the stimuli employed. Individual words of various

types have usually served as stimuli. Other than the sexually

taboo words previously discussed, emotionally laden words have

generally been used. Postman and Solomon (1950) used failure-

related words after exposing the subjects to a failure experience.







Eriksen (1951, 1952) employed words which showed disturbances

on word association tests. Another methodology utilized

nonsense syllables which were previously paired with an elec-

tric shock (Lazarus and McCleary, 1953). Spielberger (1956)

exposed stutterers to stutter-arousing words. Wiener (1955)

used words with a double entendre (fairy, pussy, etc.),

establishing the connotation of the word through the context.

Undesirable traits as opposed to desirable traits were used

by Postman and Leytham (1951). Derogatory adjectives were

used by Whittaker, Gilchrest, and Fisher (1952), while

Eriksen (1951) utilized aggressive, homosexual, and succorance-

need words as stimuli.

Other than verbal stimuli, most experimenters have

employed pictures. Dulaney (1957) used geometrical figures

while Blum (1954) exposed subjects to Blacky pictures.

Eriksen (1951) displayed pictures in which aggressive,

homosexual and succorance themes were portrayed.

In almost all of the experiments, subjects have been

free to respond with whatever word they choose. However,

Minard (1965) provided the subjects with a list of alterna-

tives from which they must choose the correct stimulus. A

group of studies have adopted a response methodology first

employed by Blackwell (1952, 1953), where subjects were re-

quired to choose from four stimulus configurations (forced

choice).







Another issue in perceptual defense/sensitization has

revolved around the variables affecting an individual's

response--those factors which affect all responses and those

which affect specific responses (i.e. to the critical stimuli).

Mishkin and Forgays (1952) and Orbach (1952) in

separate studies produced evidence that the ease of

recognition of printed words depends on the retinal area

that is stimulated. Melville (1967) reported an interaction

between retinal focus and word length in the determination

of word-recognition thresholds. Howes and Solomon (1951)

reported that visual recognition of words improves with

practice, while Postman and Bruner (1948) found an improve-

ment in the recognition of three word sentences after a

practice experience of examining drawings exposed tachisto-

scopically.

On a practical level, these studies seem to have two

implications for perceptual defense/sensitization studies.

First, the need to continually control for practice effects

must be emphasized. Therefore, findings from experiments

which determine the recognition threshold on the basis of a

few trials at the early stages of the stimulus series must

be questioned. Secondly, the word length for critical and

neutral words employed in the studies must be equal, if

Melville's research is accurate.

Although no discussion of the influence of practice







efforts on neutral versus critical stimuli could be found

in the literature, theoretically, practice effects should

equally affect neutral as well as critical stimuli.

Because the critical stimuli most often represent one

class of stimuli (e.g., sexually taboo words), whereas the

neutral stimuli may come from many classes, a factor which

selectively affects one class of stimuli would more than

likely affect the critical rather than the neutral stimuli.

For example, Eriksen and Browne (1956) reported that the

recency of prior experience with the stimuli is a crucial

variable in response latency. Sexually taboo words would

be more likely to be affected by this variable than would

neutral words.

The criticism that taboo and neutral words differed

markedly in familiarity or frequency of past occurrence in

the subject's experience was first exposed by Howes and

Solomon (1950). The authors showed that the frequency with

which the taboo words occurred in the Thorndike-Lorge word

tables (1944) was appreciably lower than those for neutral

words. They advanced the hypothesis that differential

recognition thresholds for words were a function of the

frequency with which these words had been experienced in the

past. Stimuli which had a high frequency of prior occurrence

would have low visual duration thresholds while infrequent







or rare stimuli would be expected to have high thresholds.

In further studies (Howes and Solomon, 1951; Solomon and

Howes, 1951), these authors demonstrated that the duration

thresholds could be predicted by the Thorndike-Lorge tables

of word frequency.

The Thorndike-Lorge tables are based upon the frequency

of the word occurrence in children's books or popular adult

magazines and as such were outdated in 1950. In addition,

taboo words probably appear less frequently in written

English than in conversation, and hence the frequency tables

may systematically underestimate the frequency of usage.

If the familiarity of taboo words is indeed underestimated,

Howes and Solomon's hypothesis stands open to debate. More

recently, Eriksen (1963) published data that focuses upon

this question of frequency. His research indicates that

the Thorndike-Lorge tables are misleading with regard to

taboo words. In addition, it raises serious doubt regarding

the frequency criticism leveled by Howes and Solomon. Erdelyi

(1974) points out that the significant threshold difference

in the Howes and Solomon study was due to differences

between the "extremely infrequent and very frequent words."

Still, Broadbent (1967) obtained more powerful frequency

effects than Howes and Solomon, indicating a need for further

exploration into frequency effects before the issue can be

dismissed.







Nevertheless, the criticism by Howes and Solomon did

have a marked effect upon later studies. Experiments

controlling for or neutralizing frequency effects still

reported perceptual defense and/or perceptual vigilence

outcomes (Chapman and Feather, 1972; Dulaney, 1957; Levy,

1958; Pustell, 1957; Sales and Haber, 1968; and Wiener,

1955).

Closely related to the frequency argument has been the

expectancy set argument. The proponents of this position

argue that perceptual defense data are largely a consequence

of the subject's expectations. Since unexpected stimuli

tend to have higher thresholds (Cable, 1969; Forrest, Gordon,

and Taylor, 1965; Freeman, 1954; Howie, 1952; Lacy, Lauringer,

and Adamson, 1953; Luchins, 1950; and Postman, Bronson, and

Gropper, 1953), use of the expectancy set argument to explain

perceptual defense seems plausible. However, a number of studies

documenting perceptual vigilence challenges the expectancy

set hypothesis (Bootzin and Natsoulas, 1965; Bootzin and

Stephens, 1967; Chapman and Feather, 1972; Chapman, 1974;

Dorfman, 1967; Dulany, 1957; Pustell, 1957; and Wiener, 1955).

Perhaps the most heated argument regarding the perceptual

defense/sensitization concept revolves around the response

bias position. The response bias issue is related to the

reporting of the perceptual experience rather than to the

perceptual phenomenon itself, in that not all responses have







an equal probability of appearing in the response sample.

The response bias argument was initiated by Howes and

Solomon's (1950) criticism of the McGinnies study. In

their article, they attacked McGinnies findings on the basis

of the frequency argument (previously discussed) as well

as on the basis of response bias. Howes and Solomon argued

that the subjects' responses may have been skewed not because

of what they saw or did not see, but because of what they

were willing to report. The response suppression occurs

when a subject recognizes a stimulus but fails to tell the

experimenter because of the social implications involved.

For example, the subject may wait for a few more exposures

in order to be absolutely certain before he responds. Howes

and Solomon suggest that response suppression will probably

occur whenever taboo stimuli are used. If their hypothesis

is true, then the effect of response suppression would be to

raise the recognition threshold measures artificially for

the taboo stimuli.

The criticism leveled by Howes and Solomon engendered

significant changes in the experimental methodology used in

the studies which followed. Experimenters began controlling

for response bias through a combination of mathematical and

experimental controls.

One popular experimental control was to forewarn subjects







that taboo words might appear. Lacy, Lewinger, and Adamson

(1953) and Freeman (1954, 1955) reported that forewarning

lowered the thresholds of taboo words to a level below that

of neutral words. According to Lacy et al., the more

specific the forewarning, the greater the fall in recog-

nition threshold. Fulkerson (1957) also investigated the

effect of forewarning on taboo-word thresholds, but found

it not to be significant except in the case of the first

taboo stimulus presented. Cowen and Beier (1950) found

that forewarning significantly lowered thresholds for taboo

stimuli, although they remained higher than those for neutral

stimuli.

Mathews and Wertheimer (1958) indirectly attempted

to secure a score which would represent the subjects' tendency

not to call the emotional words. The subjects were given

a list of eight words (four neutral and four emotional) and

were forewarned that every word on the list would be flashed.

In reality, the experimenters only flashed four words (two

neutral and two emotional). They were able to attain measures

of both response and perceptual suppression and still found

a significant perceptual defense phenomenon after subtracting

the measures of response suppression from those of perceptual

suppression.

A second methodology designed to control for response







bias has been that employed by Coldiamond and Hawkins (1958)

as well as Goldstein (1962). Utilizing a non-stimulus con-

trol, subjects were given a closed set of possibilities

from which to choose in order to eliminate the variability

contributed by guessing. If no difference was evident in

the distribution of responses when a stimulus was present

versus a control condition when it was absent, then one

could conclude that the response suppression accounted for

the effect. Although Goldiamond and Hawkins (1958) and

Goldstein (1962) found support for response bias utilizing

the above described methodology, numerous other studies

(Brown and Rubenstein, 1961; Zajonc and Nieuwenhuyse, 1964)

offered evidence to the contrary.

Another experimental methodology which attempts to

avoid response bias is the spatial forced choice technique,

where more than one stimulus word, design, picture or object

is simultaneously presented. The subjects locate one of

the stimuli by determining its spatial position in the

group of stimuli (upper left, lower left, upper right, lower

right). The use of a spatial forced-choice indicator

markedly changes the psychological nature of the recognition

situation, since it is no longer necessary for the subject

to determine the presence or absence of one of a class of

stimuli (emotional or neutral words). Instead, the subject







is told a stimulus will be presented, and that he must merely

discriminate it from the others which will also be presented.

Blum (1954), employing this spatial forced-choice

technique, presented Blacky pictures tachistoscopically

before and after a situation in which feelings of psycho-

sexual conflict were aroused in the subjects. In another

experiment, Blum (1955) exposed four Blacky pictures simul-

taneously on the screen to subjects who were familiar with

the pictures. The subjects were told that all eleven pictures

would appear equally, and that they must identify each picture

presented. Responses were classified as to whether the

pictures mentioned were present or absent and to the presence

or absence of conflict plus repression in the subject. The

author compared the frequency with which the names of different

Blacky pictures (assumed to be anxiety provoking) were given

as responses to tachistoscopic exposures where the Blacky

pictures, unknown to the subject, were not exposed. The

responses were then compared with the frequencies obtained

when the anxiety-provoking pictures were actually presented.

Blum found that subjects avoided calling the names of

pictures relating to their own conflicts and repressions,

but only when these pictures were actually present. No

such avoidance behavior (relative to the pictures which were

neutral for the subjects) was exhibited toward the pictures

which were not presented.







More recent studies have employed still more refined

techniques to control for response bias. Bootzin and

Natsoulas (1965) and Bootzin and Stephens (1967) controlled

response bias by forcing the subject to choose one of two

neutral words or one of two "rude" words as a response.

Rude and neutral words were never presented together and

no systematic preference within pairs could be found.

Although response bias as a source of variation was con-

trolled, the studies still were able to document perceptual

defense. Ruiz and Krauss (1968) and Zigler and Yopse (1960)

required the subject to respond with a taboo word to indicate

perception of a neutral word, and to respond with a neutral

word to indicate perception of a taboo word. The report mode

was calculated to eliminate report suppression. Again,

perceptual defense outcomes were found.

The response bias criticism of Howes and Solomon (1950)

led to various methodological changes to avoid response bias.

Nonetheless, whenever a verbal response is required from the

subject, the verbal response is subject to various motiva-

tional factors. To avoid this problem, a methodology would

need to require the subject to make a non-verbal response or

to limit the response to a "yes" or "no" answer. One glaring

weakness of all the studies reviewed is their failure to

control this problem and implement a methodology which will

satisfy one of the criteria discussed above.







More recently a new question has arisen in perceptual

defense literature. According to Erdelyi (1974), "the

major criticisms of the perceptual defense and vigilence

hypothesis are reducible ... to the question of the locus

of selectivity (bias)" (p. 11). In more simplistic terms,

the issue of selective attention (of which the frequency,

expectancy set, and response bias arguments are considered

parts) has become relevant in recent perceptual defense/

vigilence literature. At the forefront of controversy is the

information processing approach, particularly the multiprocess

theories of Dixon (1970) and Neisser (1967). The basic

premise is that between the stimulus and the response, a

whole complex of actively interacting systems intervenes.

Inputs are subjected to different kinds of transformations

and storage, in such a manner that different selection

processes are likely to be operating at different levels

of processing.

Although the information-processing approach to selective

attention has received considerable attention in recent

literature, the role of selective attention in perceptual

defense/vigilence theory is as yet unclear. Certainly,

more research is necessary in order to explore new hypotheses.

For example, Erdelyi (1974) proposes that "selectivity is

pervasive throughout the cognitive continue, from input







to output, and thus is likely to be found in most places

an investigator searches" (p. 4). Erdelyi's hypothesis implies

that the motivational basis for selectivity may vary, re-

sulting in biased or selective processing of information

in one of many ways. If future research supports Erdelyi's

contention, then the perceptual defense/vigilence argument

may have an added dimension to consider. For a more complete

discussion of selective attention and/or information pro-

cessing, the reader is encouraged to review the work of

Broadbent (1958), Erdelyi (1974), and Neisser (1967).

Just as the criticism by Howes and Solomon (1950) led

to various methodological changes to control for problems

of response, a criticism by Eriksen (1954) engendered

methodological changes in the stimulus. Eriksen criticized

those perceptual defense/sensitization studies which lacked

independent verification of the anxiety arousing nature of

the stimulus.

...The implicit assumption that the taboo words are
anxiety-arousing for all or even a majority of the sub-
jects is extremely gratuitous. Even if this assumption
were substantially correct, the studies using this
procedure make no provision for individual differences
among subjects in terms of how they respond to or
handle this anxiety. If one wishes to determine whether
psychological defenses can affect recognition thresholds,
it would seem obvious that a first requisite is to
show that the particular stimuli give rise to defen-
sive behavior as determined by other independent
criteria (p. 219).







Three techniques have been employed in order to insure

that the stimuli are, in fact, anxiety provoking. Presenting

the stimuli on two different occasions is one methodology.

A word association test, to determine an objective measure

of the stimuli emotionality for the subject, might be

utilized for example. The second confrontation with the

stimuli would be during the experiment itself.

Another method to guarantee stimulus emotionality

is to build it into the experimental design. That is,

instead of testing for the emotional significance of the

stimulus, the experimenter creates conditions which insure

the attachment of emotional significance to the stimulus.

For example, conditioning the stimuli to shock or using

failure related words as the stimuli (after the subjects

were exposed to frustrating failure experiences) have

been used to accomplish this condition.

Finally a third strategy to insure the emotional

significance of stimuli requires the combination of a

priori and empirical criteria. In this method, a class

of stimuli which are likely to be emotional for certain

subjects is chosen. Subjects, for whom it is believed

the stimuli will be anxiety-provoking, are then selected.

Thus, if the stimuli are racially connotative words, subjects

would be individuals who are highly prejudiced. The criteria







for selecting subjects has usually relied upon nontesting

procedures such as psychiatric diagnosis. Carpenter,

Wiener, and Carpenter (1956), for example, selected

subjects who used repressive or sensitizing mechanisms

in particular conflict areas and submitted words designed

to evoke these conflicts.

Motivation, interests, values and their affects upon

physical perception have also been studied. Wispe and

Drambarean (1953) investigated the effect of food and water

deprivation on the recognition of hunger-relevant and thirst-

relevant words and found that after periods of ten and twenty-

five hours of deprivation, need-relevant stimulus words

had lower recognition thresholds that neutral words.

McClelland and Atkinson (1948) also noted an increase in

food responses as a function of deprivation.

Considering ego-involvement as a motivational state,

Freeman (1954) used "ego-involving instructions" for the

experimental group which led the subjects to believe that

the perceptual task was related to academic success and

aptitude. Ego involvement had the effect of reducing thres-

holds for all words (neutral and critical). Another experi-

mental technique for inducing motivation has been to condition

anxiety towards specific words by means of electric shock.

Pustell (1957) produced perceptual sensitization experi-








mentally by associating an electric shock with a number

of neutral geometric figures. Hochberg, Haber, and Ryan

(1955) sounded a buzzer simultaneously with, as well as

after, presentation of a nonsense syllable. The buzzer

was followed by an electric shock in both cases and resulted

in higher thresholds for the critical as opposed to neutral

nonsense syllables.

If motivation, recency of prior experience with the

stimuli, as well as other variables affect the individual's

response with this methodology, the practicability of this

methodology and the need for a methodology to control for

these variables should be considered.

Although there is considerable research on perceptual

defense/sensitization, research specifically investigating

prejudice and perceptual defense/sensitization is limited.

Steelman (1940) showed Caucasian subjects fifteen pictures

showing white people and another fifteen pictures showing

black people; then the experimenter mixed in an additional

eight pictures depicting white people and eight more pictures

depicting black people. Subjects were asked to pick out

the pictures in the initial presentation series from all

the pictures. The results showed that low prejudiced

subjects (as measured by an interview) recognized signifi-

cantly more pictures with black people in them than did

the high prejudiced group.







Malpass and Kravitz (1969) found evidence that there

is better recognition and memory of faces belonging to

one's own race than there is for faces of another race.

The findings were not related to the level of prejudice

of the subjects though.

Under the guise of a yearbook evaluation study,

Sensening, Jones, and Varney (1973) had twenty-two prejudiced

and non-prejudiced white male undergraduate students inspect

twenty-five photographs of whites and twenty-five photographs

of blacks. With inspection time as the dependent measure,

the authors found that a significant interaction occurred be-

tween race of the person depicted in the photograph and the

subject's prejudice level (as measured by a racial inven-

tory). Non-prejudiced subjects spent equal amounts of time

looking at the photographs of whites and of blacks, whereas

prejudiced subjects spent significantly less time looking

at photographs of blacks than at photographs of whites

(significant at .01 level).

Based on the studies by Steelman (1940) and Sensening,

Jones, and Varney (1973) one would expect that if an indivi-

dual has a strong negative orientation toward blacks, then

exposure to blacks would be avoided. Such avoidance is

consistent with the idea of perceptual defense.

However, if one assumes that the prejudiced white








feels threatened by blacks, and there appears to be some

evidence for this (Campbell, 1965), then avoidance may be

somewhat surprising from the point of view of perceptual

vigilence. Jones and Gerhard (1967) define perceptual

vigilence (sensitization) as "an enhanced readiness to

perceive certain stimuli that have information value for

the person, including those that alert him to impending

danger" (p. 718). From this theoretical viewpoint, it is

predicted that a highly prejudiced person might not avoid

exposure to blacks, but would, rather, attend to blacks

very carefully.

In consideration of this data, it would be expected

that prejudiced individuals would focus less on the highly

connotative words, perceiving them at a farther distance

than the control words or that prejudiced individuals would

focus more on the highly connotative words, perceiving them

at a closer distance than the control words.

Although there are only two theoretical positions

which confront the relationship of racial attitude and

perceptual defense/sensitization, a number of explanations

are addressed to the perceptual defense/sensitization

phenomenon.

Howes (1954) offered a theory to account for the

perceptual defense findings in terms of the probability







of any specific word being evoked as a response. The prob-

ability of a word is the strength of the subject's tendency

to emit the word, and this summates with the stimulus factor

to produce a momentary probability value. If the stimulus

word is of a lower probability than a neutral word according

to the Thorndike-Lorge tables, then the momentary probability

for the subject to emit this word would be smaller than for

the neutral word (which has a higher probability). Evidence

which challenges this interpretation has already been

considered.

Goldiamond (1958) asserted his response bias theory

to account for perceptual defense and supplied experimental

substantiation for the response bias position (Goldiamond

and Hawkins, 1958). The authors gave their subjects prior

experiences with a number of paralogs (pairs of nonsense

syllables), which had varied frequencies. The subjects

were then tested in what seemed to be a standard recognition

experiment with the paralogs as stimuli. However, the para-

logs did not, in fact, serve as stimuli, since the correct

response on each trial was arbitrarily predetermined by the

experimenters. The subjects tended to guess high frequency

paralogs rather than low frequency paralogs though, so that

more correct responses were made when high frequency paralogs

were really correct. In other words, the high frequency







paralogs obtained lower recognition thresholds than the

low frequency paralogs, because of the existence of a

response bias in their favor. Goldiamond (1958) applied

this finding to perceptual defense, arguing that a bias

against using a word would give it the appearance of being

difficult to recognize.

Goldiamond's theory does an adequate job of accounting

for many findings in the literature which employ the

ascending Method of Limits procedure. If evidence for

perceptual defense can be found with a completely different

technique than one in which recognition thresholds are

critical, then Goldiamond's hypothesis may be tarnished.

Several writers have explained the phenomenon by

invoking biological adaptation and survival principles.

In advocating the presence of both defense and sensitization

Pustell (1957) stated:

If perceptual vigilence occurs, he is sensitized to
the conflict stimulus and thereby is prepared for what-
ever behavior is needed in order to cope with it. On
the other hand, if perceptual defense occurs, he is
relatively insensitive to the conflict stimulus, and thus
perhaps successfully avoids the anxiety which comes with
conflict situations (p. 88).

This method of accounting for the findings has failed to

gain support, because of the failure of data to support this

contention. Not only is this explanation teleological, but

it also fails to predict what stimuli will lead to

sensitization and which will lead to defense. The biological








adaption theory serves only as a possible explanation of

the event after it occurred.

Postman (1951, 1953) advanced a general theory of

perception, termed hypothesis theory, to account for the

duality of the defense/sensitization phenomenon. His

basic assumption is that an individual is always prepared

to perceive something, and forms hypotheses as to what will

be perceived. Postman attempts to explain the perceptual

defense phenomena as attributable to the "dominance of

strong alternative hypotheses" which interfere and delay

the recognition of emotional stimuli. While those who

adopt the "defense" mechanism have strong dominant (non-

emotional) hypotheses which require large amounts of

appropriate information before they are rejected, those

who adopt the "sensitizing" mechanism are considered to

have stronger negative (emotional) hypotheses.

Postman lists four main ingredients of hypothesis

strength: 1) frequency of past confirmation of the

hypothesis; 2) motivational support for the hypothesis;

3) the number of alternative hypotheses; 4) cognitive

support for the hypothesis.

Little evidence to support Postman's theory has

appeared in the literature. However, a study by Levy

(1958), which found that emotional stimulus words evoked







fewer chained associates than did neutral stimulus words,

offers evidence to the contrary.

Another theoretical position, introduced by Lysak

(1954) and later Datson (1956), argues that individuals

tend to focus more closely upon anxiety-provoking stimuli,

so that lower recognition thresholds for these stimuli

result. Little support has been generated for this

hypothesis, since the majority of the research has shown

either perceptual defense or perceptual defense and

sensitization.

The learning theory account of perceptual defense/

sensitization removes the phenomenon from the field

of perception and places it back with response variables

which are subject to the laws of learning. Because all

perceptions have to be reported in terms of some response,

the question of whether the variance exists in the per-

ception itself or in the response is a difficult one.

Behavior theorists believe that each individual has

at his disposal a hierarchy of responses to any given

stimulus. If a response is rewarded, it is more likely to

recur the next time the stimulus appears; if the response

is punished, it is less likely to appear during future

presentations of the stimulus and is replaced by another

response. Anxiety is considered to be a punisher, so







a response which elicits anxiety will be less likely

to occur again in the future. In the tachistoscopic

situation the stimulus is the word or picture presented,

and the response is the verbal statement, galvanic skin

response, etc. of the subject.

The concept of generalization is also brought forth

in the behavioral theorists attempts to account for per-

ceptual defense. In stimulus generalization, the response

learned to one stimulus may also be elicited by other stimuli

which resemble the original stimulus. If anxiety is

thought of as a learned response to a conflict-stimulus

word presented tachistoscopically, the response may genera-

lize and be elicited by another stimulus, such as a neutral

word which physically resembles the conflict stimulus.

Some evidence for this position exists. For example

McGinnies and Sherman (1952) found a perceptual defense

effect for neutral words which immediately followed the

anxiety-provoking stimulus. If perceptual defense is

learned, then the findings of McGinnies and Sherman support

the concept of stimulus generalization.

Some researchers have gone beyond the hypothesis

theory advanced by Postman and postulated that a curvilinear

relationship exists between sensitivity to input and extent

of input emotionality (Brown, 1961). More recently,







Neisser (1967) has expanded and further developed Brown's

curvilinear theory in his information processing paradigm.

Although Neisser's theory has generated much support in

the last few years, the numerous studies which fail to show

a curvilinear relationship between stimulus and response

cast doubt upon this position.

Two other theoretical positions have directed their

attention to the perceptual defense/sensitization phenomenon.

The psychoanalytic approach as first espoused by Blum (1954)

and Nelson (1955) adheres to a two-process or two-stage

perceptual process. The first process, where the stimuli

act directly on the unconscious if the stimuli are con-

sistent with libidinal desires, results in perceptual sensi-

tization. The second process, where the ego defense mech-

anisms act on the stimuli, results in perceptual defense.

Numerous writers, including Wiener and Schiller (1960)

have objected to any explanation which supports a dual process.

The two-process viewpoint posits two distinct processes -- a

supraliminal one and a subliminal one. In a series of three

experiments, the authors pointed out that there is a lack

of sufficient evidence to support subliminal perception.

Wiener and Schiller demonstrated that what had been thought

to be subliminal perception can be interpreted as the per-

ception of partial cues, which requires only a one-process







view and therefore offers evidence which challenges the

psychoanalytic position.

A final theoretical framework is provided from the

works of Snygg and Combs (1959). In their phenomenological

theory, the authors have contended that goals and values

exert a selective effect on perception and thus, markedly

affect behavior. In addition, Snygg and Combs have commented

that anxiety is in a state of being threatened, but one

in which the object of threat cannot be clearly and precisely

differentiated, resulting in less clear perception. Accord-

ing to the authors, under this state of anxiety (threat),

"we have no choice but to defend our self concepts when

they seem to us to be severely threatened" (p. 172).

Therefore, when threatened (as by anxiety-provoking stimuli),

the reaction is likely to be perceptual defense.

The research which indicates perceptual sensitization

or a dual phenomenon presents evidence that phenomenological

theory fails to explain however.

Although various theoretical frameworks have been

exposed, the theories can be categorized to a degree

into 1) those theories which accept perceptual defense

and attempt to explain why the phenomenon occurs (Howes

word probability theory and response bias theory); 2) those








theories which accept perceptual defense and attempt to

explain perceptual defense in terms of their viewpoint

(learning theory and Snygg and Combs phenomenonological

theory); 3) those theories which accept the duality

of perceptual defense/perceptual sensitization (psycho-

analytic theory, biological adaptation theory, Postman's

hypothesis theory, and Neiser's information processing

theory); and 4) those theories which accept perceptual

sensitization (Lysak as well as Datson). See Table 1

for a breakdown of theories and their respective positions

in perceptual defense/sensitization.








TABLE 1

CLASSIFICATION OF THEORIES IN
PERCEPTUAL DEFENSE/SENSITIZATION


THEORY


DEFENSE


SENSITIZATION


DEFENSE AND
SENITIZTATIOM


Word Probability
Theory X


Response Bias
Theory X


Biological
Adaptation X
Theory


Hypothesis
Theory X


Lysak's
Theory X


Learning
Theory X


Information
Processing
Theory


Psychoanalytic
Theory X


Phenomenological
Theory X








- continued


SENSITIZATION


DEFENSE AND
SENSITIZATION


Table 1


THEORY


DEFENSE


Datson's
Theory X












CHAPTER III
CONCLUSIONS FROM THE LITERATURE REVIEW
AND HYPOTHESES GENERATED

The problems involved in perceptual defense/sensi-

tization seem to revolve around two major issues --

methodology and theory. In terms of theoretical frame-

work it is now apparent that disagreement exists as to

whether perceptual defense, perceptual sensitization, or

both perceptual processes can be found. A number of

theories have attempted to account for or defend one or

both processes. In addition, two positions have emerged

with respect to the relationship between racial attitude

and perceptual defense/sensitization.

Regarding methodology, the use of ascending Method

of Limits in almost every study has been scrutinized. In

particular the use of negative indicators and the limited

response range have been discussed, as well as the issues

of frequency, response bias and selective attention.

Although the two problems of theory and methodology

seem, on the surface, to have little overlap, it appears

that a shift in the experimental/methodological approach

to the phenomenon could provide an understanding of how

personal attitudes and psychological conflicts influence

46







our perceptual processes, thereby providing some insight

into a theoretical resolution of the problem.

A methodology which aims to reach this goal would

require: 1) The response should fall on a continuous scale

which allows for differentiation into perceptual defense,

perceptual sensitization, and non-defended perception.

2) The response should be nonverbal or require as little

verbalization as possible. 3) The response should utilize

the same modality as the perception itself.

In addition, Minard (1965) listed six points which

should be heeded if the methodology is to avoid some of

the pitfalls previously discussed.

1) Measurement of perceptual defense/sensitization should

avoid the use of unexpected or socially unacceptable stimuli.

2) The stimuli employed should be chosen so that they

are personally emotion arousing for the subjects and that

they incorporate adequate controls for word frequency.

3) Results should not be averaged for stimuli which differ

greatly in their emotion-arousing properties.

4) Results may be obscured if the scores of subjects

with "quite different" personalities are averaged.

5) Averages and comparisons which involve different

situations should be avoided, because the experimental

situation may serve to alter the effective level of stimulus







emotionality (changing the sex of the experimenter, for

example).

6) Obtain measures of response bias by the use of some

control.

The major methodological inadequacies for previous

perceptual defense/sensitization studies are cited in

Table 2.

Following the advice of Minard, as well as the three

methodological refinements cited above, this study has

attempted to overcome methodological problems in perceptual

defense/sensitization and provide evidence to add credence

to a theoretical position.

Within the framework of this methodology, any of

the three theoretical positions seem to have potential

merit. However, the position which espouses the duality

of defense and sensitization has generated no support in

the literature when personality variables are considered.

Since the subjects are being grouped on the basis of racial

attitude (a personality variable), it seems highly unlikely

that the duality position would be supported by the present

study. However, an analysis of individual respondents

within each prejudice group would indicate the occurrence

of this phenomenon, should it be so indicated.

Since support has been generated in the literature














AUTHOR


49


TABLE 2

MAJOR METHODOLOGICAL INADEQUACIES
OF PERCEPTUAL DEFENSE/SENSITIZATION STUDIES


INADEQUACY


C 0 w 0
c iW *H3 jOJ
OU .HC 0 C 0 O0
0 4 20 Q H a w J0 0 10
-I Em ou 0 4 *- *r C r0

4i 0 F C Uo Q) aj o 0 o o 0 a 3 c :r o
C O) -' M) :J 0L I '-A E W -r- >
cn n o 'P r a o ,4 O Q m a) U 0W H O X O
a (D E a) e o a0 w w x U., a( t- aj L o u i ^ c
n 'i o E- Uo > CL (n M a % S vl u) yIl ma 0
Beier &
X X >: X X X
Cowen
Berger x X X x x _
Blackwell x \;_ X x__
Blum -X _ x
Bootzin r&
Natsoulis X X X X
Bootzin &
Stephens X X X
Broadbent X X X _
Brown &
Rubenstein X X
Bruner &
X X X X X X X
Postman
Byrne &
Sheffield X X X X X X X
Sheffield x x x x x
Cable X X X X
Carpenter
et al. X X X X
Chapman X X
Chapman &
X X
Feather
Cowen &
X X X X X
Beier
Cowen &
Ce x x X X X X X
Obrist
Datson X X X X
Davis X X X X X X
Dorfman X X X X X
Dulaney X X X X
Erikson X X X X X X
Erikson
(1963) X X








Table 2 continued


AUTHOR


INADEQUACY


L I4 U)
0 a C *)
c 1 0 1 1 C 0
o0 C 0 0
.- E *-1 U) CQ *4J C CD O
4 3 -1 9E N o0 O 0 .- 4U) -I J o C
uow cQ u )* CO U) r- a- 4- J ,4
Cr C H 3- j Q) ) U > CU Q- 0 4 e
p 0 F, E O ) a 0 a o 1 0 o o u ) 3 0
U 0 l f-- .0 3 P1 Q-4 U] '-, CL I r-- E : -r-I >
,U) U )0 O i DI 0) O L- p I Q) U) Q C ,-I 0 X 0
Q ) E -i o L) 0 u (). I 3 ( o ua ,0 c P
C;4 i 0 E0 a cO a" 5 c n ro Do a


Eriksen &
Brwne X X X X
Browne
Forest
osX X X X X X
et al.
Freeman X X X X Xx X
Fulkerson X X X X X X
Goldiamond
& x x x x
Hawkins
Goldstein X X X X X _
Gordon_ X X X
Greenbaum x
Hochberg X X X X X
et al.
Howes and
Solomon
Howie X X X X X X X
Kates and
Klein X X X X X
Kempler &
Wiener ____X
Kleinman X X X X X___
Kurland X X X X
Lacy x x x x
et al.
Lazarus X X X X X
et al.
Lazarus & x X X X X
McLeary
Luchins X X X X X X X
Mathews & x X
Wertheimer _







Table 2 -


AUTHOR


INADEQUACY


V)l ECU 0 -4 41 M H to
U o C N o 0-i ) ( U 0 Co
.r-1 c: *H 3 -4 >- ) n pr c: u > c: L-4 0 41 u e
p 0 r-1 p E c -r-1 0 ) 0 c 00 o U a 3 U: a 0
I a r-4 41 P 3 C 4 -) I M l 3 1-14 >
Cn IUn) 0o a cr o 0 d ) a) ru c, l -. O0 X O
Q) a) E 44 0 ) a) 4 U -r-4 Q) 3 0) P U 0 oU 4- C C: P
C_________ 0 4 H > vn '0 a- n 3o ce D c: W) un I) c (.
McGinnies X X X X X X X
Minard _X
Nelson x X X X X
Porzemsky x
Postman & x
Postman & X X X X X X X
Bronson
Postman &
Bruner X X X X X X X
Postman &
Leytham X X X X X X X
Postman &
Solomon X X X X X X X
Postman
et al. X X X X X X X
Pustell x x x x
Ruiz &
X X X
Krause
Sauber X X
Sensening X X
et al.
Solomon &
X X X X X X X
Howes _
Spence X X X X
Spielberger X X
Van deCastle X X X X
Vanderplass X
X X X X X X
& Blake
Whittaker
X X X X X X X
et al.
Wiener X X X X X X
Wiener &
Schiller X X X X X X


continued








Table 2 continued


INADEQUACY


0 b0 C 3
c- --T-i o- QJ 13 o
C 0 Q) U 0
0 u .- HC 00D l C I0 O
) E 4o 44-J H ra |- C C
I N 3 O l ( -1l CN0LU C U 0 C
r- C-tt 1 .-U1l -- U l Q U C:J U > ( C r- 0 i 4
0 ,4 W- O H w 0 0 0 I- 0
Li 0- r-4 4-J :J 0 0 CDWI M4 >
n u ro O 4 cr ( oj 0 4 C O C ( I V- Q-- rl 0 o O
O) GU O (I C -H ( LI :- 0 O O U I- C O
n 1 Ln o E > !4 U rj0 -a cn --"1 -n -- ______________

Zajonc &
Nieuwenhuyse X X X
____ _ . ________ ______ _______I __ ______


Zigler
Yopse


Zuckerman


AUTHOR


--







for both the sensitization and the defense arguments,

two-tailed hypotheses have been postulated. Although the

power of the statistical tests is weakened by such a

strategy, the opportunity of discovering the direction

of a significant difference, if any exists, is increased.

Based upon this thought, the following hypotheses are

offered in order to investigate the visual depth perception

of racially connotative words and their control words as

well as ego-alien words and their control words by high,

average, and low prejudiced male and female high school

students.

1) Highly prejudiced individuals will perceive racially

connotative and control words at significantly different

distances.

2) Average prejudiced individuals may or may not perceive

racially connotative words at significantly different

distances.

3) Low prejudiced individuals will not perceive racially

connotative and control words at significantly different

distances.

4) High, average, and low prejudiced individuals will per-

ceive ego-alien and control words at significantly different

distances.

5) There will be a significant difference in the depth per-







ception of racially connotative words between highly preju-

diced and low prejudiced individuals.

6) There will be a significant difference in the depth per-

ception of racially connotative words between highly prejudiced

and average prejudiced individuals.

7) There will be a significant difference in the depth

perception of racially connotative words between average

prejudiced and low prejudiced individuals.

8) There will be no significant difference in the depth

perception of racially connotative control words between

highly prejudiced, average prejudiced, and low prejudiced

individuals.

9) There will be no significant difference in the depth

perception of ego alien words between highly prejudiced,

average prejudiced, and low prejudiced individuals.

10) There will be no significant difference in the depth

perception of ego alien control words between highly pre-

judiced, average prejudiced, and low prejudiced individuals.

11) There will be no significant differences between males

and females within the highly prejudiced, average prejudiced,

and low prejudiced groups with regard to racially connotative

racially connotative control, ego alien, and ego alien con-

trol words.




55


12) There will be no significant difference between the

forward and backward starting positions with regard to

words and prejudice group classification.












CHAPTER IV
METHODOLOGY

Subjects

Eighty-four Caucasian freshman and sophomore high

school students were separated into three groups by means

of The Social Situations Questionaire (Kogan and Downey, 1954).

These students were randomly selected from the master

role of an Alachua County, Florida high school; subjects

for the second phase of the experiment were then chosen

from these students. One experimental group (designated

as the high prejudiced group) was composed of those

students whose score on the attitude survey fell at

least one standard deviation above the mean. A second

experimental group (designated as the low prejudice

group) was composed of those students whose score on the

attitude survey fell at least one standard deviation

below the mean. A control group (the average prejudice

group), composed of a random selection of students

whose score on the attitude survey fell between plus

one and minus one standard deviation, was also employed.

Instrument Selection

Eighty-four subjects were administered an adaptation of







the Social Situations Questionaire (See Appendix 4). This

is a fifteen item, Guttman-type scale developed by Kogan

and Downey (1956) to measure discriminatory attitudes to-

ward Negroes. The instrument was modified slightly in

1974 (the word "colored" was replaced by the word "black").

Shaw and Wright (1967) report that "this scale appears to

be above average in validity for the measurement of

attitudes toward the Negro...." Reliability is average

for a scale of this sort (.73).

Experimental Procedure

During the experimental procedure, the subject was

first dark adapted for approximately two minutes. Then,

the subject observed for five seconds a word painted in

two-inch letters with luminescent paint on a flat black

card. Immediately after the word was removed, a lumines-

cent marker placed on a track below the word was shown

to the subject. The subject was then required to tell the

experimenter whether to move the marker closer to or farther

away from the subject, so that the marker was "directly

below the spot where the word appeared or the same distance

away as the word." The word was objectively positioned at

a distance of four feet in the median plane from the front

of the instrument. The marker was positioned at either

the near position (one foot in front of where the word







appeared) or the far position (one foot behind where the

word appeared).

Apparatus (See figures 1 and 2)

The experimental apparatus was a box-like structure

eight feet long, one foot wide, and one foot high. It was

constructed of three-eighth inch plywood with all walls

solid except for the top which was open. The interior of

the box was painted flat black and all devices were painted

the same color or made of black construction paper (i.e. the

paper upon which the words are printed).

Centered in the apparatus at the zero foot mark (four

feet from each end of the eight foot long box), and eight

inches off the floor of the apparatus, the words appeared.

A window slide through which the words were slid, was at-

tached by a hinge to a bar running across the width of the

apparatus at the top. After the word was exposed for five

seconds, the viewing slot was closed by means of a hinged

cover operated by a pulley, and the window slide device

lifted out of the box. The luminescent marker was then

inserted and the viewing slot opened. On the floor of the

apparatus a track two and one-half inches wide extending

the entire length of the box, served to guide the marker.

The marker, two inches by two inches in size, was pushed

along the track by a wooden standard ruler attached to the







base of the marker. The wooden standard ruler was ten feet

long and extended through a covered hole in the back of the

apparatus, and was able to move the luminescent marker to

either extremity of the apparatus. The ruler was also

employed to tell exactly how far the words were perceived

from the zero inch point (the zero inch point being the

point where the word was exposed).

The entire apparatus was mounted on a table, so

that it was eye level to the subject, when the subject was

seated before it. When seated in the dark room, the subject

was able to see only the luminescent words or the lumines-

cent marker in the approximate median plane. When the words

were visible, the luminescent marker was removed from the

box; when the marker was present, the hinged word holder

or window slide was placed in an upward position out of the

box. With the equipment not visible, the environment was

free from distracting and confounding stimuli. To prevent

the subject from seeing or estimating the size of the appa-

ratus, a "wall" was constructed so that the only contact with

the apparatus was by looking through the viewing slot. After

each trial, the viewing slot was closed and the measurement

taken and recorded.

Stimuli (See Appendix I)

Two words were presented that were thought to be ego-







alien to each subject, as defined by Porzemsky (1969), eg.

PIG. For each ego-alien word, defined as inharmonious with

one's total self, the subject was also presented a neutral

control word which was perceptually similar except for a

change in one letter, eg. PIN.

Three words were also employed that were thought to

be racially connotative, eg. NIGGER. The three racially

connotative words used were chosen prior to the experiment

from a list of ten words by five students serving as judges

(see Appendix 2). Again, for each racially connotative

word employed, a neutral control word which was perceptually

similar except for a change in one letter were presented to

each subject.

Starting Position

There were two trials for each word. For one trial,

the starting position for the luminescent marker was one

foot closer to the subject than the zero inch point where

the word appeared. For the other trial, the starting position

of the luminescent marker was one foot further away than the

position of the word. The initial starting position as well

as the order of presentation of the words was randomly assigned.

Recording Device

The distance that each word was perceived from the zero-

inch mark was recorded on the experiment recording sheet to

the nearest quarter inch (see Appendix 6).







Statistical Procedures

The primary analysis of the data consists of individual

t-tests and analysis of variance. Two-tailed t-tests were

carried out to test for differences between ego-alien and

control words for the high, average, and low prejudiced

groups. In addition, two-tailed t-tests were carried out

to test for differences between racially connotative and

control words for the high, average, and low prejudiced

groups. Two-tailed t-tests were also employed to test for

sex differences and starting position differences.

Analysis of variance was employed to test for differences

between the three groups in depth perception of racially

connotative, racially connotative control, ego alien and

ego alien control words.

All told, ten two-tailed t-tests and two analyses of

variance were conducted.







FIGURE 1--APPARATUS (Open View)


/ I \
/ I \
/ I \
luminescent hinged standard
marker word holder ruler


FIGURE 2--APPARATUS (Front


ruler
runner
runne r


View)


- -front wall


- -viewing slot












CHAPTER V
ANALYSTS OF THE DATA

In the initial stage of the study eighty-four students

responded to the racial attitude survey. Twelve students

failed to complete all items on the survey, responded more

than once to one or more questions, or failed to complete

the data sheet. The results for these twelve students were

ruled invalid and therefore not included in the data. For

those seventy-two students who correctly completed the racial

attitude survey as well as the data sheet, Table 1 shows

the frequency distribution. A low score is indicative of

low prejudice while a high score is indicative of high

prejudice.

For the purposes of this study, those students who

scored twenty-seven or less on the racial attitude survey

(less than or equal to one standard deviation below the mean)

became the low prejudiced group in the second part of the

study. Those subjects who scored forty-four or greater

(more than or equal to one standard deviation above the mean)

became the high prejudiced group in the second phase of the

experiment. The average prejudice group consisted of a

random selection of fourteen students whose scores fell







between twenty-eight and forty-three (between one standard

deviation below the mean and one standard deviation above

the mean.)

The data for the second phase of the experiment (for

those students in the low, average and high prejudiced

groups) was collected on the "Experiment Recording Sheet" --

see Appendix 6. Analysis of the data by means of individual

t-tests and analyses of variance provided information bearing

upon the hypotheses and questions previously mentioned in

Chapter 3.

The first hypothesis, which suggested that highly

prejudiced individuals would receive racially connotative

words and their control words at significantly different

distances, was affirmed. Table 2 indicates a significant

difference at the .01 level. For the average prejudiced

individuals, a significant difference in depth perception

distance was also found between the racially connotative words

and their control words. In this instance, a significant

difference at the .05 level was found, as delineated by

Table 2. No finding had been suggested in the second hypothesis.

Finally the third hypothesis, which suggested that low prej-

udiced individuals would not perceive racially connotative

words and their control words at significantly different

distances, was affirmed. Table 2 shows no significant difference.







Hypothesis 4 stated that all three groups would perceive

ego alien words and their control words at significantly

different distances. Table 3 shows a significant difference

at the .05 level for the high prejudiced group and a

significant difference at the .01 level for the low

prejudiced group. There was no significant difference found

between ego alien words and their control words for the

average prejudiced group. However, the difference did approach

significance (.14).

The fifth hypothesis postulated that a significant

difference in the depth reception of racially connotative

words between highly prejudiced and low prejudiced individuals

would exist. Table 4 shows the difference to be significant

at the .05 level.

The sixth hypothesis, which suggested that there would

be a significant difference in the depth perception of

racially connotative words between the high prejudiced and

average prejudiced groups, was affirmed. Table 4 shows a

difference at the .05 level.

However, no significant difference was found in the

depth perception of racially connotative words between the

low prejudiced and average prejudiced groups as delineated

by Table 4. Therefore, hypothesis 7 is rejected, although

the difference approached significance (.17).







TABLE 3

FREQUENCY DISTRIBUTION FOR THE RACIAL
ATTITUDE SURVEY


SCORE

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37


FREQUENCY

9

0

1

2

3

4

-J



4
3

1






3




4
2






9

5


SCORE

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54


FREQUENCY

4

6

9
I?

0

2



3
9
2

3

2

3

2

0

0

1

1

1

0

1







TABLE 4

DISTANCE MEANS OF THE RACIALLY CONNOTATIVE WORDS
AND THEIR CONTROL WORDS FOR THE
HIGH PREJUDICED, AVERAGE PREJUDICED, AND LOW PREJUDICED
GROUPS



RACIALLY CONNOTATIVE CONTROL

High
Prejudiced 9.88 (S.D.=4.61) -0.15a (S.D.=3.69)

Average
Prejudiced 4.03 (S.D.=3.82) -0.68b (S.D.=3.27)

Low
Prejudiced 1.58 (S.D.=3.52) -0.05 (S.D.=2.39)


at=5.30, p(.01 between racially connotative words and their
control words for the high prejudiced group.

bt=2.25, p<.05 between racially connotative words and their
control words for the average prejudiced group.


An analysis of variance revealed no significant

difference in the depth perception of the racially

connotative control words between the high prejudiced,

average prejudiced, and low prejudiced groups. Table 5

indicates that no significant difference exists, thereby

affirming hypothesis eight.

Table 6 also fails to indicate a significant difference,

in this instance in the depth perception of ego alien words

between the three groups. Acceptance of the ninth hypothesis

is indicated by the analysis of variance.







TABLE 5

DISTANCE MEANS OF THE EGO ALIEN WORDS
AND THEIR CONTROL WORDS FOR THE HIGH PREJUDICED,
AVERAGE PREJUDICED, AND LOW PREJUDICED GROUPS.



EGO ALIEN WORDS CONTROL WORDS

High
Prejudiced 5.98 (S.D.=4.07) 1.00a (S.D.=3.01)

Average
Prejudiced 1.53 (S.D.=2.94) -1.20 (S.D.=3.17)

Low
Prejudiced 3.15 (S.D.=2.91) -0.55b (S.D.=2.33)


at=2,47, p<.05 between ego alien words and their control
words for the high prejudiced group.

bt=2.76, p<.01 between ego alien words and their control
words for the low prejudiced group.


Analysis of variance also points to an acceptance

of the tenth hypothesis. Table 7 shows no significant

difference in the depth perception of the ego alien

control words between the high prejudiced, average prejudiced

and low prejudiced groups.

Male/female differences are examined in Tables 8 and

9. As expected, a significant difference was found among

both high prejudiced males and females in the depth

perception of racially connotative words versus their

control words. A similar finding was discovered among the

average prejudiced male and female groups. However, as




69


TABLE 6

ANALYSIS OF VARIANCE OF RACIALLY CONNOTATIVE WORDS
BETWEEN HIGH PREJUDICED, AVERAGE PREJUDICED,
AND LOW PREJUDICED GROUPS





SOURCE SS DF MS F

Among Groups 363.72 2 181.86

Within Groups 591.44 27 21.91 8.30*

Total 955.16 29


Post-hoc Test of Significance (Turkey's Test)

X (High Prejudiced) X (Average Prejudiced) X (Low Prejudiced)

---- 5.85* 8.30*

---- 2.45



- -p<.05


anticipated, no significant difference in the depth

perception of racially connotative words versus their

control words was found for either the male or female

low prejudiced groups. Table 8 delineates the significant

differences.

Whereas no male/female differences were discovered

for the racially connotative words versus their control

words, the data comparing mnle/female differences for the







TABLE 7

ANALYSIS OF VARIANCE OF RACIALLY CONNOTATIVE CONTROL WORDS
BETWEEN HIGH PREJUDICED, AVERAGE PREJUDICED,
AND LOW PREJUDICED GROUPS



SOURCE SS DF MS F

Among Groups 2.56 2 1.28

Within Groups 321.84 27 11.92 0.11

Total 324.40 29










TABLE 8

ANALYSIS OF VARIANCE OF EGO ALIEN WORDS
BETWEEN HIGH PREJUDICED, AVERAGE PREJUDICED,
AND LOW PREJUDICED GROUPS



SOURCE SS DF MS F

Among Groups 101.41 2 50.71

Within Groups 444.47 27 16.46 3.08

Total 545.88 29







TABLE 9

ANALYSIS OF VARIANCE OF EGO ALIEN CONTROL WORDS
BETWEEN HIGH PREJUDICED, AVERAGE PREJUDICED,
AND LOW PREJUDICED GROUPS


SOU


RCE _SS DF i MS F


Among Groups


25.55 2 12.78


Within Groups 384.65 27 10.54


1.21


Total 410.20 29


ego alien words versus their control words revealed un-

expected findings. Table 9 indicates a significant

difference for high prejudiced males in the depth perception

of ego alien words versus control words, but the difference

for high prejudiced females failed to reach significance.

For the average and low prejudiced groups, however, the

opposite result was found. Table 9 indicates a significant

difference in the depth perception of ego alien words and

their control words for average and low prejudiced females,

but the difference for average and low prejudiced males failed

to reach significance. There appears to be no explanation

for this unexpected finding. It should be pointed out,

though, that the t scores for the high prejudiced female

group, as well as the average and low prejudiced male groups

all approach significance.

Tables 10, 11, and 12 examine the effect of the

















4C
0 CJ C,





0 0 0 0
M O O O.O -

70 4' I ,N 0 0 0
Snl O o o o o r

S[ 0 a a a
0-' 0-, 0 0 0 0





0
a)
z O o o O 0 0 0



-O 4- 4 4- 4.j
0* 0* *, ,
UO P- C C C:






T 3 C0o C o




F-A (n F? 0 0 0 0
S U - O- U U

0 D 0 0 r



- o r m -i r- *r-i r-i










4J 4 4' 0 i
0 r 0) 0C 0 0
rom -." o, ^; o o

5- Ca) a) a) a)














->- DO D O 4- Q
0 0 1 1l U- vi r4





0 n .s 4 H a ) r4 r' -





D 00- u .r U-4 u oo r) F
n m u ca vi 4- -I lU)









,0 >O O O O 4O J O



0P C4 r-4a H * p
C)J CD a C)C Ca >C







V) C0 -C ca. Caa 0l toa 4-J .
a- )) c o a)

-l -0 0 Or- 0 0


V > O1 0 4 0 C'> 0 -U 0












PI <- P. Q9J 4-4 0 4410 U 44 4410
r C a < P-, U- OL. C0 4- 3 4- -< T





















































































a)

.r-I
-o

*H 5-


-o
aO

a) -r4




>( -I


< C-,


U)
In


0


-4
0


C
r-l


0
ci





0




rn
C


)


0o
c 1
a E



o


-4 *

a)

0 )
( q)


aI-


.6 0









II -4
0O







C14
1 -J
4- 0
CZ (J4


1
0


r-4
0

4-1


U












u
0
*c








5-
0







40
rc



TOC

0 U





*el


o E
O U





( U
C0





mIn
4I

U O
3 -d


r4-4

uO




U 0L-


In
-u
14
0
0
O
3


.4
s-I

C
0
U
-i










0
a(U














0
O






c0






.)3 Q
r4J






To 3
In




CM




43 U
Ci-
0 0


b 4A
a). -i
Cd











*1-

II

.1J







starting position -- near or far -- on the depth perception

of the racially connotative words versus their control

words and the ego alien words versus their control words

for the high prejudiced, average prejudiced, and low

prejudiced groups.

As expected, a significant difference in the depth

perception of racially connotative words versus their

control words was found for the high prejudiced groups

at both the near and far starting positions, as indicated

by Table 10. However, no significant difference was

found for the depth perception of ego alien words

versus their control words for either the near or far

starting positions. The difference did approach significance

for both starting positions though.

For the average prejudiced groups, similar results

were evident in comparing ego alien words and their

control words. As with the high prejudiced group, no

significant differences could be found for either

starting position. In addition, only the near starting

position difference approached significance. The comparison

of the racially connotative words and their control words

by starting position for the average prejudiced group

produces more confusion. Whereas, Table 11 clearly

points to a very significant difference (p<.01) between




























r.4 .
0 0-

Ob
u 0
C
OM

O
La)
*H U


0 u

C


cn4
4 -)



O }ri

,-40
01





CCL
u



O
O *r4
0 r-
> 4-J

*H






Or 0



UUCO




0
4 C



'-4
u C







03
S-
. cn
0 U
v<


r-4

SC o

O
C 0


0 H








0.
(0





u D


-Wa)

*H U
3









C
CC

0





*a H
3C

> .H






0 0
4LJ






r-4 0
c0




0 0

U *






ITj



-0
O M
C C





41 0
a


MC
CO 0
*H *i
0 M
Ocn
n-J 4
4-1 0
U) 0.


bOC
*H r-

can U
04















































0





















CN





Cl)
*




























mO

-I"
















cn
*










-1o

4/1 n
U) 0


toc
C 0
Sr-4 *-r-


CO
41 0
cij In.






























z

JLn

0
o










O

0

-ZO

U


woc:
CO: 0


d n

V)oa


O
N
CM







1'1





0





cn
o


























oo
0



MC








C: O
r- -
DU
*-
wo


0 0.
-J r4
*~I
.LnJ
(I,


0


cn
(/)
,a


0

,--!
0
41

O.
u ::




a
, *P
0 Q-





03





"4 CL


0

0





MO
0 Ln
to






01




01
U (






mJ i
0 C








the depth perception of racially connotative words and

their control words for the near starting position, no

significant difference for the far starting position

could be found. Perhaps, the difference in the perception

of racially connotative words versus control words lessens

as the "apparent" distance of the words becomes greater.

When the luminescent marker is placed at the far starting

position (two feet further than the placement of words at

the near starting position), the effect of the racially

connotative words may be reduced. It should be noted,

however, that the difference did approach significance.

The fact that the difference was only significant

for the near starting position may have theoretical impli-

cations as well. This finding adds credence to the

perceptual defense position, in that defense against racially

connotative words was stronger when the stimulus was apparently

nearer to the subject.

No significant differences were found between the

depth perception of racially connotative words and their

control words for the near or far starting positions for

the low prejudiced group. Table 12 indicates the anticipated

result. However, a significant difference was discovered

between the ego alien words and their control words for the

near starting position but not for the far starting position.







In fact, the difference for the far starting position

did not approach significance.

When individual results are examined with respect to

differences in the perception of racially connotative and

control words across groups, the differences are accentuated.

Appendix 6 displays the performances of subject A (from the

high prejudiced group) and subject B (from the low prejudiced

group). Close comparison of the two performances exhibits

the large difference in distance between each of the

racially connotative words versus the perceptually similar

control words for subject A and the small or lack of

difference in distance between each of the racially connota-

tive words versus the perceptually similar control words

for subject B. For example, a difference of three and

one-half inches was found between subject A's perception

of the word "nigger" and the word "bigger" at the far

starting position, while a difference of only one-fourth

of one inch was found in subject B's performance.

The direction of the difference is also significant,

in that the racially connotative words were always perceived

at a greater distance by subject A than the control words,

in comparison to subject B where no consistent pattern was

evident.

In summary, the predicted differences in the depth







perception of racially connotative words versus their

control words for the high prejudiced and average prejudiced

groups were found. Differences in the depth perception

of racially connotative words between the three groups

were also found. However, hypotheses concerning the

depth perception of ego alien words versus their control

words as well as the depth perception of ego alien words

between groups had mixed results, failing to confirm

some hypotheses. When male/female differences were

explored, no significant differences were found among the

three groups between racially connotative words and

their control words, but once again, mixed results were

found with respect to the ego alien words and their

control words. Similar results were also found when

near and far starting positions were compared. Anticipated

results occurred in all but one comparison of the racially

connotative words and their control words. However,

comparisons of the near and far starting positions with

respect to ego alien words and their control words failed

to substantiate the hypothesis. Finally, when individual

performances of high and low prejudiced subjects on the

depth perception instrument were compared as illustrated

in Appendix 6, the differences become clearer.











CHAPTER VI
SUMMARY AND CONCLUSIONS

The major purpose of this study was to investigate

whether or not personal attitudes influence perceptual

processes. In particular, the problem to which this

paper was addressed was whether or not racial attitude

affects visual depth perception. If an effect could be

documented, in which direction would the effect lie?

Perceptual defense would be supported if high prejudiced

individuals perceived racially connotative words at a

significantly greater distance than control words while

the opposite result would lend support to the perceptual

sensitization viewpoint. If no effect were found, the

result would support the duality position.

Twelve hypotheses were generated to deal with this

problem. Of greatest concern was an examination of visual

depth perception of racially-connotative words and their

control words among high prejudiced, average prejudiced,

and low prejudiced subjects. For control purposes, visual

depth perception of ego alien words and their control

words for the three groups was also examined. In addition,

hypotheses dealing with gender and starting position (near

or far) were also generated and examined.

81







The Sample

The sample consisted of 84 Caucasian freshman and

sophomore high school students who were randomly selected

from the master role of an Alachua County, Florida high

school.

Treatment of the Data

The first phase of the experiment consisted of the

completion of the racial attitude survey and an accompanying

information data sheet. Twelve subjects were dropped from

the experiment because of a failure to successfully complete

the attitude survey and/or the data sheet. From the results

of the attitude survey, three groups were formed which

comprised the subjects for the second phase of the study.

Those subjects who scored one standard deviation or more

below the mean comprised the low prejudiced group while

those whose scores on the racial attitude survey fell one

standard deviation or more above the mean comprised the

high prejudiced group. A random selection of those whose

scores clustered around the mean (between +1 and -1 S.D.'s)

on the attitude survey comprised the average prejudiced

group. The subjects from these three groups then individu-

ally completed the second part of the experiment -- the

depth perception black box.







Results

As predicted, the differences in the depth perception

of racially connotative words versus their control words

were found for the high prejudiced and low prejudiced

groups. Differences in the depth perception of racially

connotative words between the three groups were also found.

In all instances, the high prejudiced group perceived the

racially connotative words at a greater distance than did

the average prejudiced group. A similar difference in

distance was found between the average prejudiced group and

low prejudiced group, with the average prejudiced group

perceiving the racially connotative words at a further

distance than did the low prejudiced group.

Unlike the positive results attained from the hypothesis

which investigated the racially connotative words, one of

those hypotheses examining ego alien words and their control

words failed to reach significance. No significant difference

was found between ego alien words and their control words

for the average prejudiced group, although the difference

did approach significance. Significant differences between

the ego alien words and their control words were found for

the high prejudiced and low prejudiced group. As predicted

no significant differences were found in the depth percep-

tion of ego alien words or their control words among the

three groups.







In examining male/female differences, no differences

were found in the depth perception of racially connotative

words versus their control words; however, a difference in

the depth perception of ego alien words versus their control

words did appear. Whereas high prejudiced males perceived

ego alien words at a significantly further distance than

control words, a similar finding did not result for high

prejudiced females. However, for the average and low

prejudiced subjects, the opposite finding was evident.

Average prejudiced and low prejudiced females perceived

the ego alien words at a significantly further distance

than the control words, but no significant difference

was found for average prejudiced and low prejudiced males.

In relationship to starting position (near versus far),

no significant difference was found between near versus far

starting position in the depth perception of racially

connotative words for the three groups although the differ-

ence approached significance. However, differences between

near versus far starting position were evident for the ego

alien words and their control words for the low prejudiced

group.

Conclusions

The major conclusion that can be drawn from this study

is that racial prejudice (as a personal attitude) influences









visual depth perception. Of particular interest is the

finding that the depth perception of racially connotative

words varies conversely with the prejudice level of the in-

dividual, i.e. the more prejudiced an individual, the further

back that individual is likely to perceive a racially conno-

tative word. In conjunction with this finding, the highly

prejudiced individual was found to perceive the racially

connotative words at a significantly further distance thaii

the control words. These two findings offer support for the

theoretical formulations which espouse only a perceptual

defense argument to the problem, rather than a sensitization

or duality persuasion. Therefore, the evidence from this

study offers support for both the learning theory as well as

phenomenological points of view.

The learning theory account of the perceptual defense

phenomenon, as described previously in the literature review,

adheres to a stimulus-response model. Anxiety is considcrc-d to

be a punisher, so a response which elicits anxiety is less

likely to occur again in the future. In the tachistoscopic

experiments, the word or picture presented served as the

stimulus while the verbal statement, galvanic skin response,

etc. of the subject was the response. Within this dynamic,

stimulus generalization is added, so that a response learned

to one stimulus may also be elicited by other stLiruli which







resemble the original stimulus. If anxiety is thought of

as a learned response to a conflict-stimulus word presented

tachistoscopically the response may generalize and be

elicited by another stimulus, such as a neutral word which

physically resembles the conflict stimulus. In the

experiment conducted for the purposes of this study, a

neutral word resembling the conflict-stimulus except for

a change in one letter was employed to test for any gener-

alization effect. None was evident, so the learning theory

position needs to be questioned.

In contrast, the findings of the present study appear

to add more credence to the phenomenological position of

Snygg and Combs. Their theory has purported that values

and attitudes exert a selective effect on perception, which

in turn, markedly affects behavior. When threatened, there

appears to be a narrowing of the perceptive field to the

object of threat (tunnel vision) as well protection of the

perceptions the individual already holds. Within this

paradigm, anxiety is seen as a state of being threatened.

Under this state of anxiety (threat), the individual reacts

by defending against the anxiety-provoking stimulus. The

object of threat cannot be clearly and precisely differen-

tiated, resulting in less clear perception. In the present

experiment, Snygg and Combs' concept translates into defense







against the anxiety-provoking stimulus, in this instance,

the racially connotative words. As a protection of the

self concept against these words, it seems that the

more prejudiced individual tends to "push" the racially

connotative words further back. In other words, the

threat produces a psychological defense against the

anxiety-provoking stimuli resulting in a concrete behavioral

change -- viewing the racially connotative words at a

further distance than the control words.

The findings of this experiment also offer support to

the study conducted by Sensening, Jones, and Varney (1973).

Employing a different modality than that utilized in the

present study, the authors found evidence of perceptual

defense by highly prejudiced individuals in the comparison

of photographs of black and white subjects. It would be a

reasonable generalization to state that the more prejudiced

the individual the more likely he or she is to perceptually

defend against racially connotative words.

A second conclusion generated from this study is to

diminish the results of previous studies which failed to

control for methodological inadequacies. The present

dissertation demonstrated that the methodological problems

cited by Minard as well as by this author could be remedied.

Among the methodological problems addressed and dealt with







in the present research were the use of a continuous response

scale, the deemphasis of verbalization in the response, the

avoidance of unexpected stimuli, the use of emotion arousing

stimuli, the categorization of stimuli by their emotion-

arousing properties and the categorization of "different"

personality subjects (on the basis of prejudiced levels).

The utilization of this improved methodology may have

played a role in clearing up or at least improving the

state of the past theoretical confusion in perceptual

defense/sensitization research. When methodological problems

are diminished, as is the case in this study, then the

results of the research indicate the presence of only one

process in perceptual defense/sensitization. Perceptual

sensitization or the dual paradigm of defense and sensi-

tization are not indicated, while perceptual defense is

readily demonstrated.

Recommendations for Further Study

There is an important need to continue perceptual

defense/sensitization studies which employ refined metho-

dologies to control for the inadequacies in previous re-

search. Replication of this present study would prove

to be a worthwhile endeavor, to substantiate the present

findings.

A number of refinements of the present study would







serve to improve the validity. First, increasing the size

of the population would magnify the likelihood of attaining

significance. However, according to James R. Barrell (Personal

communication, 1975), the smaller the sample size, the

greater is the opportunity to find true differences.

Secondly, using different populations such as

college students, middle-aged adults, young children,

rural or urban populace, or populations of different

income groups or socioenconomic status would serve to

substantiate the present findings or amend the present

research.

It would also be interesting to modify the present

study to view the effect of racial prejudice on the

visual depth perception of Negroes. By simply changing

the racially connotative words from "nigger" and "black" to

"honkey" and "whitey" for example, an experimenter would

be able to determine if highly prejudiced Negroes employed

perceptual defense or if highly prejudiced Negroes instead

utilized perceptual sensitization or both defense and

sensitization.

Beyond the scope of confining research to racial

prejudice and its influence on visual depth perception,

investigating the influence of other measurable modalities

(i.e. anxiety, depression, and anger) on visual depth







perception would be a logical point to continue the vein

of this present study. In addition, experimentation

employing different methodologies than the one used in this

dissertation might be explored to measure not only the

influence of racial prejudice on visual depth perception,

but also the influence of numerous values, attitudes, and

psychological processes on other forms of perception (i.e.

figure-ground discrimination, auditory perception, etc.)

One might for example vary the stimulus from the racially

laden words to educationally laden words, to test if those

students with the poorest academic records or those students

who misbehave most often defend more against the education

ally laden words than those students who perform well

academically or those who are well-behaved. Some insight

into the dynamics of the student's misbehavior or poor

academic performance might be gained from such research.

From a practical point of view, the depth perception

box, in and of itself, seems to have value. Although a great

deal of investigation is still needed on the validity and

reliability of such an instrument, it is interesting to

speculate on the possible utility of the depth perception

box. As a means of identification, the instrument has

many possibilities. For example, utilizing stimuli which

relate to homosexuality (i.e. gay, queer, etc.) might be a




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs