Automatic and effortful memory processing by students with and without mental retardation

MISSING IMAGE

Material Information

Title:
Automatic and effortful memory processing by students with and without mental retardation
Physical Description:
x, 166 leaves : ill. ; 29 cm.
Language:
English
Creator:
Thomas, Suzanne B., 1948-
Publication Date:

Subjects

Subjects / Keywords:
People with mental disabilities -- Education   ( lcsh )
Memory in children   ( lcsh )
Special Education thesis Ph.D
Dissertations, Academic -- Special Education -- UF
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1994.
Bibliography:
Includes bibliographical references (leaves 157-165).
Statement of Responsibility:
by Suzanne B. Thomas.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002023957
oclc - 32885988
notis - AKL1498
System ID:
AA00011819:00001


This item is only available as the following downloads:


Full Text











AUTOMATIC AND EFFORTFUL MEMORY PROCESSING BY STUDENTS
WITH AND WITHOUT MENTAL RETARDATION

















By

SUZANNE B. THOMAS
















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

UNIVERSITY OF FLORIDA

1994














ACKNOWLEDGEMENTS

Many people have been instrumental in the completion of

this dissertation and the steps leading to it. They deserve

more thanks than this acknowledgement can convey. I am

grateful to my committee chairperson, Dr. Charles Forgnone,

and committee members Dr. Michael Nunnery, Dr. Jeanne

Repetto, Dr. Stephen Smith, and Dr. Stuart Schwartz for

their guidance, patience, and support. I also thank Dr.

Joan Curcio and those faculty members from my master's

programs at the University of North Texas who encouraged me

to pursue my dream.

I am grateful to those people who instilled in me the

importance of learning and growing in our profession. They

taught me the importance of individualization and

normalization before those ideas were in vogue. I thank

Leslie, Dennis, Robin, and Mable. Special thanks go to Dr.

Jerry Vincent who taught me that administrators with a

vision can make a difference.

Friends too numerous to mention provided words of

encouragement and support throughout this project. Linda,

Joe, Tracy, and Michael gave me the motivation and

confidence to persevere. The assistance of Sybil Brown made

the study possible. Special words of thanks are extended to









Jesse Todd who was there through the trials and triumphs,

the tears and celebrations.

The greatest debt of gratitude I owe to my family,

especially my parents, Wilson and Freida Thomas, for

supporting and encouraging me and for teaching me values.

They taught me to care about others, to work hard, and that

with desire anything is possible.
















TABLE OF CONTENTS


Page

. ii


ACKNOWLEDGEMENTS . . .


LIST OF TABLES . .


ABSTRACT . . . ix

CHAPTERS


1 INTRODUCTION . .


Objectives . .
Rationale . .
Theoretical Rationale .
Educational Rationale .
Definition of Terms .
Delimitations of the Study .
Limitations of the Study .
Summary and Overview of Remaining


Chapters


2 REVIEW OF LITERATURE . .. .19

Theory of Automaticity . 19
Structure of Memory ....... .23
Structure of Memory and Mental Retardation 25
Structure of Memory and Automaticity ... .27
Level of Processing . 28
Level of Processing and Mental
Retardation . . 30
Level of Processing and Automaticity 30
Strategy Use . . 31
Strategy Use and Mental Retardation 32
Strategy Use and Automaticity .. .34
Application of Memory Models . .. 34
Encoding Condition . ... 35
Developmental Trends . 42
Structural Processes. . .47
Strategic Processes . ... 53

3 METHODS AND PROCEDURES . .61


Null Hypotheses . .
Selection and Description of Participants
Research Methods . .


. 61
63
. 65









Procedures . . .. 68
Memory Task . . 69
Recall Test: Test of Effortful
Processing . .. 71
Relocation Test: Test of Automatic
Processing . . 72
Follow-up: Test of Secondary Memory 73
Data Treatment . ... .74
Analyses . . 75
Supplemental Analyses . 76
Summary . . 77

4 DATA PRESENTATION AND ANALYSIS . .78

Participant Characteristics . .. 78
Recall and Relocation Performance .. .82
Recall Performance . .. .83
Relocation Performance . 87
Recall and Relocation Interaction 91
Post Hoc Analysis . .. .96
Item Retention . .. .96
Study Time . . .. .97
Intrusion Errors . .. .99
Summary of Results . .. 100

5 DISCUSSION AND IMPLICATIONS . .. .102

Purpose and Objectives . ... 102
Study Methods . ... .103
Results of the Investigation .. 103
Interpretation and Discussion of Related
Findings . . 114
Developmental Level . ... 114
Encoding Condition . ... 115
Problems and Limitations . ... 118
Implications and Suggestions for Further
Research . . ... 121
Implications .. . .. 121
Suggestions for Future Research .. .126
A Final Comment . 127

APPENDICES

A RESEARCH APPROVALS . .. 128

B PARENT INFORMED CONSENT DOCUMENT ... .130

C INSTRUCTIONS FOR EXAMINERS . .. 132

D SCRIPTED INSTRUCTION FOR EXPERIMENTAL ENCODING
CONDITIONS . . 134









E DATA SHEETS FOR RECORDING RECALL AND
RELOCATION RESPONSES . 145

REFERENCES ................. ... ... .. 157

BIOGRAPHICAL SKETCH . .. .166















LIST OF TABLES

Table Pace

1 Pictures Used as Stimuli . ... 67

2 Interrater Reliability . .. 74

3 Percent of Personal Characteristics by
Developmental Level . .. .79

4 Percent of Personal Characteristics by
Encoding Condition . .. 79

5 Percent of Personal Characteristics by
Encoding Condition within Developmental
Level . . 80

6 Chi-Square Analysis of Encoding Condition by
Personal Characteristics within Developmental
Level . . .81

7 Means and Standard Deviations of Recall and
Relocation Performance by Developmental Level
and Encoding Condition . ... 83

8 Mean Number of Items Recalled by Developmental
Level and Encoding Condition . .. 84

9 Source Table for Effect of Developmental Level
on Recall Performance, Day One ... .85

10 Source Table for Effect of Encoding Condition
on Recall Performance, Day One .. 85

11 Source Table for Effect of Developmental Level
on Recall Performance, Day Two ... .86

12 Source Table for Effect of Encoding Condition
on Recall Performance, Day Two ... .87

13 Mean Number of Items Relocated by Developmental
Level and Encoding Condition . .. 88

14 Source Table for Effect of Developmental Level
on Relocation Performance, Day One ... .88









15 Source Table for Effect of Encoding Condition
on Relocation Performance, Day One .. .89

16 Source Table for Effect of Developmental Level
on Relocation Performance, Day Two .. .90

17 Source Table for Effect of Encoding Condition
on Relocation Performance, Day Two .. 90

18 Source Table of Repeated Measures ANOVA for
Recall . . 92

19 Source Table of Repeated Measures ANOVA for
Relocation . .... .94

20 Percent Change by Developmental Level and
Encoding Condition . ... 97

21 Means and Standard Deviations of Study Time by
Developmental Level and Encoding Condition 98

22 Correlation Coefficients of Study Time to
Recall and Relocation Performance by Encoding
Condition and Developmental Level ... .99

23 Correlation Coefficients of Intrusion Errors
to Recall and Relocation Performance by
Encoding Condition and Developmental Level 100


viii















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

AUTOMATIC AND EFFORTFUL MEMORY PROCESSING BY STUDENTS
WITH AND WITHOUT MENTAL RETARDATION

By

Suzanne B. Thomas

August 1994


Chairperson: Charles J. Forgnone
Major Department: Special Education

The purpose of the study was to examine information and

memory processing by students with and without mental

retardation. Two objectives were stated. The first

objective was to determine if differences in effortful and

automatic processing existed as hypothesized from the theory

of automaticity. The second objective was to determine if

memory processing differed by developmental level dependent

upon (a) whether the information was automatically or

effortfully processed or (b) whether clustering or semantic

strategies were applied.

In this study, a picture recall and relocation test was

completed. Students (N=180) aged 10 through 15 were

participants in the study. The students represented three

developmental levels: students classified as trainable

mentally handicapped, educable mentally handicapped, and








students without mental retardation. Within each

developmental level, students were randomly assigned to one

of three encoding conditions: nonsemantic, semantic, or

clustered. A follow-up test was conducted approximately 24

hours following the initial examination. A split-plot ANOVA

with repeated measures was conducted to analyze immediate

and 24-hour delayed recall and relocation performance.

The analysis revealed significant differences in the

number of pictures correctly recalled and relocated.

Students without mental retardation recalled and relocated

more pictures correctly than students classified as educable

mentally handicapped who recalled and relocated more than

trainable mentally handicapped students. Significant

interaction effects resulted when retention interval was

included in the analysis. As relocation performance, a test

of automatic encoding of spatial location, was not

insensitive to the intervening variables of developmental

level and encoding condition, the theory of automaticity

could not be supported.

The results of this investigation may contribute to

increased precision in the design of educational programs

for adolescents with mental retardation. Conclusions and

recommendations for further research are provided.














CHAPTER 1
INTRODUCTION

The focus of the study reported herin was memory and

information processing by persons with mental retardation.

Memory processing is crucial in the development of

intellectual behavior, problem-solving abilities, and

adaptive functioning (Turnure, 1991). Memory has been

defined as (a) the retaining and recalling of past

experiences, (b) the capacity to behave in a way modified by

past experiences, or (c) the power to reproduce or recall

what has been learned (Katims, 1987a).

Researchers of theories of information processing

assert that memory is determined either by (a) the structure

of memory, (b) the use of memory strategies, or (c) the

level of processing. Structural features of memory refer to

those aspects of the memory system which are not

programmable. These structures cannot be varied or changed,

although structural capacities may emerge as a function of

age and experience (Brown, 1974). Memory strategies, called

control processes, may include the use of such techniques as

rehearsal, mnemonics, clustering, and semantic stories to

increase recall (Katims, 1987a). The level-of-processing

approach recognizes the importance of both strategies and

prior experiences in information processing (Craik &








Lockhart, 1972). According to the control processes and

level-of-processing approaches, recall is enhanced by the

use of strategies to increase information organization, by

deeper processing, which includes semantic processing, and

by elaboration techniques (Schwartz & Reisberg, 1991).

Memory-processing theories based on the structural deficits

approach are used to define mental retardation. Approaches

based on strategic deficits are used to develop techniques

to improve learning and memory performance (Lupart &

Mulcahy, 1979). The theoretical framework for these three

approaches to memory processing are reviewed in more detail

in Chapter 2.

Additionally, it is hypothesized that memory is

influenced by how information is acquired or encoded. In

the theory of automaticity, Hasher and Zacks (1979, 1984)

described two types of mental or cognitive processes:

effortful and automatic. Effortful encoding involves

strategic processes in the intentional application of such

operations as clustering, imagery, rehearsal, and

elaboration. Automatic processes, on the other hand, occur

incidentally, without attention or awareness (Posner &

Snyder, 1975). Automatic processes are used to encode such

item attributes as spatial location, frequency, and temporal

order and are not sensitive to encoding variables or

individual differences (Hasher & Zacks, 1979, 1984).

Differences in memory processing experienced by persons

with mental retardation have been the subject of prior








investigations. From these research efforts, persons with

mental retardation have been reported to be deficient in (a)

structural processes, (b) the spontaneous use of adequate

control processes, (c) active mediational devices, and (d)

the ability to strategically transform input (Brown, 1974).

Early research into the memory capacities of students with

mental retardation was focused on the structure of memory

functioning. Differences in memory performance between

persons with and without mental retardation were interpreted

to be either developmental differences resulting from slower

mental development and a lower level of attainment or

differences resulting from a deficiency in one or more

cognitive processes. Later researchers deemphasized

structural-based processing and concentrated on the use of

control processes (Ellis, Meador, & Bodfish, 1985; Glidden,

Bilsky, Mar, Judd, & Warner, 1983). Investigators reported

the ability to use control strategies spontaneously may not

develop until early adolescence and may be further delayed

for persons with mental retardation (Brown, 1974).

Although the findings of these early research efforts

have contributed significantly to the understanding of

mental retardation, changes have been noted in the

population of students involved in these studies. MacMillan

(1989) reported that research literature published in the

1960s and early 1970s, which serves as a basis for current

understanding of the cognitive characteristics of persons

with mental retardation, was conducted with samples of








students with educable mental handicaps. This population

generally included persons with intelligence quotient (IQ)

scores close to, or exceeding, the current upper limit for

defining mental retardation and who had been long-term

residents of large institutions (Forness & Kavale, 1993). A

large percentage of these study participants had cultural-

familial retardation and, therefore, cognitively resembled

their nonretarded peers. As the functioning level of

students currently classified with mental retardation

differs from the samples used in the original studies,

MacMillan (1989) urged caution in extrapolating findings of

these early studies to today's population as the "time-

period specific validity may be of little relevance

for current educators" (p. 13). Turnure (1991) urged the

inclusion of students with more severe disabilities in the

study of memory processing to increase the appreciation of

individual differences and to analyze and understand the

learning processes of the variations in mental retardation

subgoups.

In addition, the current emphasis on placing students

from all developmental levels in regular classrooms with

nonretarded peers of similar chronological age was not

addressed by early researchers. Comparative data generally

available from previous research equate persons with mental

retardation with persons with similar mental ages, college

students, or the elderly (Diech, 1974; Dugas & Kellas, 1974;

Ellis, Woodley-Zanthos, & Dulaney, 1989a; Lamberts, 1979;








Merrill, 1992). Detterman (1979) questioned this approach

in reviewing previous methodological flaws in the study of

memory of persons with mental retardation. Bray and

Ferguson (1976) reported that performance differences noted

when comparing persons with mental retardation and persons

matched for mental age may be attributable to previous task-

related or situational experiences. Turnure (1991) pointed

out that previous studies of automatic encoding and

retention were completed with heterogeneous populations, a

practice that limits comparisons, generalizations, and

usefulness in applying results to educationally relevant

situations.

Detterman (1979) reviewed the importance of studying

memory functioning of persons with mental retardation. From

a review of previous research efforts, Detterman concluded

that a better understanding of memory can lead to (a)

increased knowledge of intellectual functioning and how the

mind operates, (b) a better understanding of the source of

individual differences and prescriptions for rectifying

differences not based solely on IQ scores, and (c)

clarification of how the memory processes of persons with

mental retardation differ from persons without retardation.

According to Detterman, the study of memory of persons with

mental retardation contributes to increased precision in

assessment and intervention and provides a theoretical basis

in developing remedial techniques.








This study was designed to examine the effortful and

automatic information processing by students with and

without mental retardation. In their theoretical framework

of automaticity, Hasher and Zacks (1979, 1984) hypothesized

that memory for spatial location should not benefit from

either encoding variables (instructions) or individual

differences. The use of strategic behaviors or strategic

tactics is reported to be inadequate in persons with mental

retardation, but it is not known how the use of these

behaviors varies by developmental level (Belmont & Mitchell,

1987; Brown 1974). Therefore, students classified with

educable mental handicaps (EMH), trainable mental handicaps

(TMH), and peers without mental retardation of similar

chronological age were included in the present study to

determine if memory processing is similar for these three

groups.

Objectives

The objectives of the study were twofold: The first

objective was to investigate if differences in effortful and

automatic processing exist as hypothesized by Hasher and

Zacks' theory of automaticity (1979, 1984). The second

objective was to determine if the memory processing of

persons with and without mental retardation differs by

developmental level dependent upon (a) whether the

information was automatically or effortfully processed or

(b) whether clustering or semantic strategies were applied.








As has been suggested, the model presented by Hasher

and Zacks differentiates effortful and automatic processing.

Effortful memory functions are voluntary and intentional and

require considerable memory capacity. Automatic functions,

on the other hand, occur without intention, function at a

constant level under all circumstances, and require minimal

cognitive energy. Encoding of item recognition requires

effortful processing, while frequency of occurrence and item

location have been found to be automatically encoded (Ellis,

Katz, & Williams, 1987; Ellis & Allison, 1988; Hasher &

Zacks, 1979, 1984). In this study, effortful processing

was determined by recall of item information. Memory for

spatial location was used to test automatic processing.

According to this model, only effortful processes were

expected to be enhanced by the use of control strategies and

to be sensitive to developmental trends (Ellis et al., 1987;

Katz, 1987).

Rationale

Theoretical Rationale

Researchers of general theories of information

processing assert that memory is determined either by (a)

the structure of memory, (b) the use of memory strategies,

or (c) the level of processing. For persons with mental

retardation, deficits have been described as an

interrelatedness of these factors. Meador and Ellis (1987)

hypothesized that the differential performance of persons

with mental retardation and their nonretarded peers on








short-term memory tasks is due to deficits in both

conscious, effortful processes and in the automatic aspects

of encoding. In this study, automatic and effortful

processing and control strategy use were investigated

relative to developmental level.

Researchers of automatic, or incidental, learning

report that memory that does not depend on strategic

processes is developmentally insensitive (Brown, 1974).

Involuntary memory is the result of incidental learning and

accompanies active exploration of the environment.

Automatic processing requires minimal attentional resources

and does not tap memory capacity (Katz, 1987). According to

Hasher and Zacks (1979, 1984), both frequency of occurrence

and spatial location can be successfully encoded

automatically as well by persons with mental retardation as

by their nonretarded peers.

In contrast with incidental learning, voluntary memory

is reported to develop gradually as a child develops.

Gradual development from involuntary, passive memory to

voluntary or active memory parallels the hypothesis that

intentional strategy use is dependant on maturity level.

Performance differences between persons with mental

retardation and nonretarded peers have been found to occur

under task conditions that require the active use of

semantic techniques or require cognitive effort (Brown,

1974; Merrill, 1990).








Persons with mental retardation may apply strategies

differently, fail to adopt strategies for information

processing, or inadequately integrate more than one

dimension from stimuli to solve problems (Brown, 1974). As

performance improved following strategy instruction supplied

by Belmont and Mitchell (1987), these researchers concluded

that persons with mental retardation were not deficient in

the ability to apply memory strategies. Deficiencies

resulted from the use of different techniques to apply

strategies. Bray and Turner (1987) supported the findings

of Belmont and Mitchell and concurred that most strategic

deficiencies of persons with mental retardation were

performance anomalies rather than production deficiencies.

Ellis (1970) and Gerjuoy and Spitz (1966) hypothesized that

performance problems, which result in deficient use of

short-term memory storage, vary with intelligence and show

developmental changes.

Educational Rationale

In addition to theoretical significance, various

educational implications may be derived from the results of

this investigation for both classroom teachers and

educational leaders. Prior researchers of memory processing

provided evidence that original learning conditions are

important aspects of recall (Turnure, 1991). Other factors

such as the use of study time to increase strategy use

(Turner & Bray, 1985) and perceptions of task difficulty

(Belmont & Mitchell, 1987) also can mediate strategic







10

behaviors and memory effectiveness. Results from this study

may clarify and extend previous information on structuring

learning situations to increase the effectiveness of

classroom instruction, especially classrooms for students of

various developmental levels.

Current focus on normalization and functional skill

training has accentuated the need to develop more effective

strategies for helping people with mental retardation recall

more adaptive information such as telephone numbers, street

addresses, or job task lists (Kramer, Nagle, & Engle, 1980).

An increased understanding of the processing of relocation

and frequency of occurrence data can guide instruction in

such adaptive areas as shopping and completion of work tasks

(Ellis et al., 1989a). Understanding the hypothesized

distinction between automatic and effortful encoding of

temporal cues will help devise training programs employing

techniques using the cues of size, shape, or location to

mediate differences in learning.

Additionally, the increased understanding of memory

processing furnished by this study is important to persons

in educational leadership positions. Educational

leadership, particularly the administration of special

educational programs, has evolved as the field of special

education has experienced a period of developmental

maturation (Lipp, 1992). Administrators increasingly are

required to be instructional leaders and managers for all

students (Van Horn, Burrello, & DeClue (1992). The American








Association of School Administrators (AASA) (1991) has

documented the changing role of school leaders. According

to AASA, a leader in a restructured school needs increased

knowledge and skills in curriculum content and development,

pedagogy, and theories of education. Effective leadership,

therefore, requires information about curriculum,

instructional devices, modes of instruction, and related

services to meet the individual needs of all students

(Biklen & Taylor, 1985).

The increased understanding of memory-processing

functions gained in this study can contribute to leadership

development as called for by AASA. This understanding of

memory processing will enable the school administrator to

(a) provide guidance in program development and curricula

decisions, (b) provide teacher support and advance

professional development opportunities, (c) develop

communication among general and special educators,

specialists, and parents, (d) act as a change agent in

developing programs and policies, allocating resources, and

evaluating outcomes, and (e) review and assist in

development of individual educational plans (Biklen &

Taylor, 1985; Lipp, 1992; Van Horn et al., 1992).

Definition of Terms

Automatic processing is the incidental processing of

information with minimal attention or awareness. In the

present study, automatic processing was assessed by recall

for spatial location.







12

Clustered encoding is using a patterning technique that

involves grouping items by categories, or clusters. In this

study, participants were prompted or cued to name items from

the same conceptual category to increase clustering.

Control strategies are strategies such as rehearsal,

semantic, mnemonic, or stories that increase the

organization of information in order to increase recall. In

the present study, the affect on recall of two control

strategies, clustered encoding and semantic encoding, was

assessed.

Educable mental handicap (EMH) is defined by Rule 6A-

6.03411 of the Florida Administrative Code and the Florida

State Plan as (a) a general measured level of intellectual

functioning two or more standard deviations below the mean

and generally falling between two and three standard

deviations below the mean, (b) the assessed level of

adaptive behavior below that of other students of the same

age and socio-cultural group, and (c) the demonstrated

subaverage level of performance in academic, preacademic, or

developmental achievement. In the present study,

participants were considered to be educable mentally

handicapped if they were classified as EMH within their

respective school district and had an IQ score of 59 through

73. The upper limit of 73 was used to account for the

possibility of a plus or minus three standard error of

measurement in the intelligence test score.








Effortful processing is the use of voluntary,

intentional effort to encode information. These efforts may

include the use of such techniques as rehearsal, imagery, or

elaboration and may be affected by characteristics of the

individual. In the present study, effortful processing was

assessed by recall for item information.

Encoding is the process by which certain aspects or

features of a stimulus establish an internal representation

of the event for storage and retrieval. In the present

study, nonsemantic, semantic, and clustered conditions were

used to encode information.

Free-recall procedure is a procedure in which subjects

are asked to recall stimuli in any order.

Immediate recall is the ability to remember just-

presented stimuli. In the present study immediate recall

was assessed by requesting participants to name pictured

items without delay following picture presentation.

Long-term memory provides storage for vast quantities

of information that is remembered but not currently needed

or active. The term secondary memory, currently preferred

by behavioral psychologist, is used in the present study to

describe this memory store. Secondary memory was assessed

by 24-hour delayed recall in the present study.

Nonsemantic encoding is the failure to use previously

known information to give meaning to and increase memory for

a new event or stimuli. For the present study, nonsemantic

encoding was cued by asking the participant to provide the








name of a pictured item without prompting additional

information about the item.

Peer without mental retardation of similar

chronological age is, for the purpose of this study, defined

as a student age 10 through 15 who is not receiving special

education services and whose Comprehensive Test of Basic

Skills (CTBS) scores indicate functioning within the grade

level average of students taking the test in this district.

Primary memory, also known as short-term memory, is a

part of memory processing that is limited in size. New

incoming information will displace the current contents of

this memory store.

Recall is remembering stimuli from prior presentations.

Recall requires memory to be searched with prior context as

a referent point and explicitly demands recovery of

connection between items and context. Listing previously

presented words and fill-in-the-blank tests are examples of

tests of recall. In the present study, recall was assessed

by asking the participants to name previously seen pictures.

Relocation procedure is a procedure in which subjects

are asked to recall the location of previously seen stimuli.

This procedure is a measure of memory for spatial location.

In the present study, placing pictures on their original

location on a page was used to test relocation ability.

Retention interval is a measure of the amount of time

information is held in either primary or secondary memory







15

storage. In the present study, both immediate and secondary

retention intervals were assessed.

Secondary memory is a vast memory storage not dependent

on ongoing activity for maintaining ability to remember.

This memory store may be referred to as long-term memory.

Semantic encoding is the use of previously known

information not associated with any specific event or

stimuli to give meaning to and increase understanding and

memory for a new event or stimuli. In the present study,

semantic encoding was cued by requesting participants to

relate interesting information about a picture being shown.

Short-term memory, the second memory store in the modal

model of information processing, receives analyzed

information from sensory memory and provides a small,

limited storage where information is held for processing.

Short-term memory is sometimes known as active or working

memory; however, the term primary memory is preferred.

Trainable mental handicap (TMH) is defined by Rule 6A-

6.03411 of the Florida Administrative Code and the Florida

State Plan as (a) a general measured level of intellectual

functioning two or more standard deviations below the mean

and generally falling between three and five standard

deviations below the mean, (b) the assessed level of

adaptive behavior below that of other students of the same

age and socio-cultural group, and (c) the demonstrated

subaverage level of performance in academic, preacademic, or

developmental achievement. In the present study,








participants were considered to be trainable mentally

handicapped if they were classified as TMH by the school

district and had an IQ score of 30 through 55.

Delimitations of the Study

This study was delimited by geographical restrictions,

subject selection, and the scope of measurement procedures.

Students from one school district, Pinellas County, a large

district located in west central Florida, were included.

Participants were students aged 10 through 15 classified as

EMH, TMH, or peers without mental retardation with similar

chronological ages. Classification of retardation and IQ

scores available from the district were accepted. Etiology

of the mental retardation was not considered. Students with

sensory impairments were not included, and consideration was

not given to sex or socioeconomic status of the

participants.

Additionally, the study was delimited by the procedure

used to measure the dependent variables. A picture recall

and relocation task was used to measure recall and

relocation ability based on procedures used in previous

investigations of automatic and effortful memory processing

by persons with mental retardation (Ellis et al., 1987;

Ellis et al., 1989a; Katz, 1987).

Limitations of the Study

Generalization of the results of this study are limited

by subject selection. Participants were randomly selected

for participation from among those meeting the age and








developmental level criteria available from the

participating school district. A bias, however, may have

existed in the receipt of parental consent to participate.

Participants were randomly assigned to experimental groups.

Summary and Overview of Remaining Chapters

The purpose of this study was to determine if the

ability of students with and without mental retardation to

recall prior stimuli supports Hasher and Zacks's theory of

automaticity. The memory functioning of students of three

developmental levels was explored. Although persons with

mental retardation are reportedly deficient in the

structural features of memory processes as well as the

efficient and spontaneous use of control procedures, it was

hypothesized by Hasher and Zacks (1979, 1984) that

information that is automatically processed will not be

sensitive to these developmental differences. Therefore,

automatically processed information will be recalled equally

well by persons at various developmental levels. An

extension of this hypothesis is that information that

requires effortful processing will be recalled

differentially, dependent on developmental level. A better

understanding of memory processing for students with and

without mental retardation will aid in structuring learning

conditions and in increasing instructional effectiveness.

In Chapter 2, a conceptual framework is considered in

which three theories of information processing are

presented: the structural or modal model, the use of







18
organizational or control processing strategies, and the

level of processing approach. Hasher and Zacks's theory of

effortful and automatic processing is reviewed.

Additionally, an analysis of literature relating the

application of these theories to persons with mental

retardation is included. In Chapter 3, the hypotheses are

stated and the methodology for implementation of the study

is discussed. The presentation and results of the

statistical analysis are included in Chapter 4. Chapter 5

consists of a discussion of the results in terms of support

for information-processing theories and significance for

educational decision making.














CHAPTER 2
REVIEW OF LITERATURE

Information-processing theories have developed to

explain how information is acquired, held in storage, and

recalled when needed. To establish a foundation for

understanding memory processing, this chapter begins with a

review of Hasher and Zacks' (1979, 1984) theory of

automaticity and three theoretical approaches to memory

processing: the structure of memory, the level-of-

processing approach, and the use of memory strategies. The

chapter concludes with a review of studies applying these

theories to persons with mental retardation.

Theory of Automaticity

The theoretical model developed by Hasher and Zacks

(1979) is based on two types of information processing:

automatic and effortful. With automatic processing,

information is acquired incidentally, without attention or

awareness. Automatic processing requires minimal cognitive

energy, is rapid, makes few demands on attention, and is not

sensitive to individual characteristics. A process is

defined as automatic if it can be executed even though

cognitive resources are diverted elsewhere (Merrill, 1992).

Strategic processes are intentionally used to encode

information with effortful processing. Effortful








information processing is slow, sequential, attention

demanding, and subject controlled; further, may be affected

by such individual variables as IQ, instructions, and

motivation (Schneider & Shiffrin, 1977). According to this

theoretical model, some stimuli require effortful processing

to be encoded into memory. Other stimuli, including

frequency of occurrence and spatial location, require only

automatic processing to be integrated into memory systems

and can be as successfully encoded by persons with or

without mental retardation (Hasher & Zacks, 1979, 1984;

Katz, 1987).

Schneider and Shiffrin (1977) hypothesized that new

learning requires controlled processing while well-learned

sequences become automatic. Hasher and Zacks (1979)

incorporated this interpretation into their explanation of

the two sources of automatic processes. First, automatic

processes may be genetically determined and vary minimally

with age, intelligence, culture, education, or early

experience. The second source of automatic processes,

according to Hasher and Zacks, is extended practice.

Automatic and effortful processing, therefore, work

simultaneously. Memory performance is dependent not only on

what the person does but also on what the person already

knows.

Researchers of incidental learning have found that

memory that does not depend on effortful, strategic

processing is developmentally insensitive (Merrill, 1992)








although the ability to focus voluntarily on task-relevant

information may not develop until early adolescence and may

be further delayed for persons with mental retardation

(Brown, 1974). Merrill (1992) reported that encoding

differences between persons with and without mental

retardation may be due to (a) deficits in effortful but not

automatic processing, (b) differences in both automatic and

effortful processing, or (c) differences in processing

resources that may vary with developmental level.

Performance differences occur under task conditions that

require the active use of semantic information or other

techniques requiring cognitive effort (Brown, 1974; Merrill,

1990).

Greene (1986) reviewed the efforts of researchers who

investigated the hypothesis that frequency of occurrence is

encoded automatically. According to Greene, a number of

researchers have demonstrated that subjects who know they

will be tested for frequency information do no better than

those who are told to expect a free-recall or an unspecified

memory test. Thus, support was offered for automatic

encoding of this variable. Katz (1987) examined prior

research on the automaticity hypothesis and memory for

spatial location. Some researchers have reported finding no

individual differences in automatic processing of location;

other researchers have found age-related differences. Katz

presumably agreed with the view of Ellis et al. (1987) that

these differences may not reflect differences in location








memory per se "but rather reflect the influence of

developmentally sensitive task-related variables on the

criterion test" (p. 22). To minimize the possibility of

interference with automatic processing in memory tests,

Ellis et al. (1987) made the following recommendations:

Task-related tests should include (a) minimal opportunities

for effortful or strategic processing, (b) instructions and

requirements equally understood by all subjects, (c)

location cues clearly and equally differentiated, and (d)

item information that is encoded equally well by all

subjects.

The significance of studying the theory of automaticity

with persons with mental retardation is twofold. First,

deficits in automatic processing may affect memory directly

and compromise the ability to discriminate among stimulus

events. Additionally, automatic processes may serve as

building blocks without which the development of effortful

processes is precluded (Katz, 1987). Katz went further to

assert that if information that is processed automatically

by nonretarded persons is processed effortfully by retarded

persons, limited attentional resources are consumed "leaving

little for the operation of strategic processes" (p. 49).

Belmont and Mitchell (1987) agreed that deficient

fundamental processing may preclude the efficient

application of any memory strategies. Therefore, according

to Belmont and Mitchell, if fundamental (automatic)

processing deficits exist, "the possibility of deficiencies








at the next higher level (strategies) would seem to be

largely irrelevant" (p. 95).

Researchers have relied on the structural, level-of-

processing, and strategic approaches to explain memory

functioning. Each of these theoretical approaches will now

be reviewed. How these three approaches relate to the

theory of automaticity, and how mental retardation has been

interpreted within each approach is emphasized.

Structure of Memory

The structural features of memory refer to those

aspects of the memory system which are not programmable and

cannot be varied or changed (Brown, 1974). Starting in the

1950s, cognitive psychologists borrowed from the

explanations of electronic processing of information to

hypothesize that memory is made up of three discrete memory

stores: sensory memory, short-term memory, and long-term

memory. This approach, which is known as the modal model,

focuses on how information is obtained and transferred among

the memory stores as it is progressively incorporated deeper

within the memory system.

Katims (1987a) explained that the structure of memory

approach does not necessarily refer to the physiological

structure of the brain; it refers to structural features or

limitations in the sense of rapid decay of information, lack

of flow of information among the various stores, or the

inability to encode stimuli for retrieval. According to the

structural approach, sensory memory provides a momentary









record of raw, unanalyzed events. Short-term memory

provides a small repository where information is held by

rehearsal techniques while being processed. Long-term

memory provides permanent storage for unlimited quantities

of information while it is not being used (Schwartz &

Reisberg, 1991).

In the modal model, short-term and long-term memory are

different memory mechanisms. Short-term memory is small, of

a fixed size (holding approximately seven items), is easily

accessed by simple attention, and is occupied by whatever

entered the most recently. Long-term memory is accessed

when information is attended to and contemplated. The

capacity and stability of storage increases as information

flows from primary, or short-term, to secondary, or long-

term, memory (Dugas, 1976). The organization of information

can increase the amount of information that can be held in

short-term memory. However, as the size of memory chunks

increases, the number of chunks that can be held in short-

term memory decreases (Schwartz & Reisberg, 1991; Sitko &

Semmel, 1972).

Support for the modal model is provided by research

into free-recall performance. A U-shaped serial-position

curve has consistently resulted from these experiments and

is indicative of high recall for primacy and recency events

and low recall for items presented in intermediate positions

(Schwartz & Reisberg, 1991). Recency events held in short-

term memory are those events experienced the most recently.









Primacy events, those initially experienced, are held in

long-term memory as they received more in-depth processing.

In addition, Paivio and Begg (1981) reported modality

differences for recency items; recall is better when items

are presented auditorially than when the presentation is

visual. Implications from this line of research, according

to Paivio and Begg, are that one is likely to remember what

one has just heard better than what one has just read. That

modality difference may not hold true for primacy events,

however.

Although the modal model provides a basis for the

understanding of memory functioning, the model has been

criticized for failure to (a) provide an explanation for

continuity in the flow of information among the memory

stores (Brown, 1974), (b) consider the strategies and

activities of the subject (Brown, 1974; Schwartz & Reisberg,

1991), or (c) adequately explain recalling of previously

learned material in a specific order (Schwartz & Reisberg,

1991). Schwartz and Reisberg (1991) illustrated this last

phenomenon, termed long-term recency effect, with subjects

who recalled the American presidents in order in a free-

recall procedure.

Structure of Memory and Mental Retardation

Both short-term and long-term memory deficiencies have

been documented to occur in persons with mental retardation.

However, not all research findings are consistent. Brown

(1974) reported that persons with mental retardation are








customarily described as deficient in short-term memory but

not long-term memory although it is unclear whether the

acknowledged deficiencies reflect a structural deficiency or

a lack of strategy application. Equal recall of recency

items by persons with and without mental retardation on

serial-recall tests have led researchers to conclude that

there is no difference in primary or short-term memory.

Turnure (1991) completed a review of long-term memory and

mental retardation and concluded that while data on long-

term memory is inconclusive, evidence is accumulating that

the long-term memory capacity of persons with mild mental

retardation functions within normal ranges on a variety of

tasks.

To apply the modal model to persons with mental

retardation, it is necessary to interpret these findings in

terms of structural versus strategic deficiencies. Katims

(1987a) confirmed that evidence supports IQ-related

individual differences in memory performance as a result of

nonstrategic processes. For persons with mental

retardation, the stimulus trace may be both shortened in

duration and lessened in intensity resulting in a failure to

learn and remember as well, given the same stimulation or

effort (Katims, 1987a). Ellis (1970) acknowledged that

persons with mental retardation may suffer also from

rehearsal deficits as determined by a failure to profit from

increased exposure time on tasks.








Developmental theorists contend that short-term memory

deficits are not structural problems but deficiencies in the

spontaneous use of strategic devices that limit access to

long-term memory stores (Butterfield, Wambold, & Belmont,

1973). Overall memory performance differences have been

attributed to deficits in rehearsal, although Hale and

Borkowski (1991) questioned this view. According to Hale

and Borkowski, "it is unclear whether differences in primary

memory in mentally retarded individuals are attributable to

structural deficits, rate of forgetting, or processing

efficiency" (p. 514). Brown (1974) asserted that

developmentally related limitations to structural capacity

may set limits to the extent to which strategies can be

trained, and these limitations may restrict the ability to

generalize strategy use.

Structure of Memory and Automaticity

Intelligence-related differences in recall have been

noted when strategic processes were not involved. Ellis et

al. (1985) provided evidence that persons with mental

retardation experience automatic processing deficits. These

researchers found that students with mental retardation

encoded fewer items on a test of automatic encoding but,

once encoded, stimulus durability was unrelated to IQ.

These authors favored an automatic processing interpretation

of intellectual differences, called for additional research

in this area, and invited reconsideration of the idea that








the relationship between intelligence and memory was due to

effortful processing alone.

A shift in investigations from structural memory

capacities to investigations of processing and strategic

behaviors resulted from evidence of rehearsal deficits that

affect how information is acted upon in short-term storage.

Additional focus was provided to ways in which strategic

deficits can be remediated with instruction and support.

The essential difference in structural and control failures,

according to Katims (1987a), is the susceptibility of the

deficit to training.

Level of Processing

The level-of-processing approach maintains that deep

processing, thinking about the meaning of stimuli, is the

best way to establish memory. This approach preserves the

central theme of the modal model; memory improves as

information is processed. Focus moves from the structural

features of memory, however, to processing, or analysis, of

the information. Thus, the determinants of memory shift

from the physical features of the information to its

semantic features and the strategic behaviors employed by

the subject (Lupart & Mulcahy, 1979).

Memory is viewed as on a continuum by the level-of-

processing approach. The terms primary and secondary memory

are preferred for categorizing memory functions as the terms

short-term and long-term memory are seen as providing a

misleading distinction about duration of processing time








(Brown, 1974; Schultz, 1983; Schwartz & Reisberg, 1991).

Emphasis is given to the qualitative, as opposed to

quantitative, aspects of analysis performed on the stimuli.

Primary and secondary memory are not seen as separate and

discrete memory stores but provide for a seamless flow of

information as deeper processing occurs. The strength of the

memory trace is determined by the dimensions along which the

stimulus is encoded (Craik & Tulving, 1975). The more

elaborately information is analyzed and encoded, the more

durable the memory trace (Schultz, 1983). Concerns with this

theory have been noted by researchers, however, in defining

and measuring depth of processing and in accounting for

other strategies a person may use to enhance memory

(Schwartz & Reisberg, 1991).

Some researchers contend that elaborative processing is

yet another approach to information processing while other

researchers have argued that elaboration is not a separate

principle from processing depth: elaboration simply

involves deeper processing (Schwartz & Reisberg, 1991).

Elaborative processing involves thinking about the multiple

aspects of an event's meaning. In elaborative processing,

something is added to what is being learned in order to make

it more meaningful (Scruggs & Laufenberg, 1986). More

elaborative or richer memory strategies make connections

between what is learned and what is known. These

connections aid in deeper processing and enhances memory.

Therefore, with the elaborative approach, the depth to which







30
an individual is able to process information is dependent on

the expertise, prior knowledge, and expectations the

individual brings to the situation (Schwartz & Reisberg,

1991; Scruggs & Laufenberg, 1986).

Level of Processing and Mental Retardation

Persons with mental retardation may experience both

mediational and production deficits which affect the level

to which information is processed. Mediational deficits

exist when a person is unable to employ potential mediators,

integrate more than one dimension from stimuli, or

consolidate new information with previously known

information in problem solving situations. Production

deficits may delay the ability to select and attend to

relevant aspects of stimuli (Brown, 1974). Schultz (1983)

suggested that persons with mental retardation may be

capable of achieving shallow encoding as well as persons

without retardation, but they may have difficulty in

achieving deep encoding. Schwartz and Reisberg (1991)

attributed this difficulty to either a lack of relevancy of

the memory test or a reduced amount of prior knowledge (a

deficient schematic network) the person brings to the

learning situation.

Level of Processing and Automaticity

According to the level-of-processing approach, incoming

information is subjected to various levels of analysis. The

strength and durability of memory is positively related to

the degree to which the stimuli has been elaborated at








encoding (Craik & Lockhart, 1972; Craik & Tulving, 1975).

Craik and Lockhart (1972) minimized the distinction between

incidental and intentional processing and argued that it is

the level of processing that predicts future memory

performance. Schultz (1983) suggested that incidental

memory performance increases as a function of encoding

processing depth. Stimuli processed at a deep semantic

level will result in a more persistent memory trace.

Mar and Glidden (1977) reported that several stimuli

attributes may determine the depth to which the stimuli is

processed. Preliminary, or automatic, encoding involves

physical features of the stimuli while deeper, effortful

processing involves semantic analysis. Intention to

memorize is critical in memory performance. Mar and Glidden

acknowledged that the failure to encode beyond preliminary

stages may contribute to deficits in associative clustering

and memory processing.

Strategy Use

Mnemonic strategies that increase the organization and

semantic content of information improve memory for to-be-

remembered material (Paivio & Begg, 1981; Schwartz &

Reisberg, 1991; Sitko & Semmel, 1972). Rehearsal, chunking,

grouping and other organizational devices, and encoding by

transforming from one stimulus code to another (e.g., from

visual to auditory stimuli) represent types of strategy use.

According to the general strategy use hypothesis,

performance is increased to the maximum possible for a task








when an optimum strategy is properly executed. A properly

executed strategy is one that guides the construction of an

effective sequence of tactics (Belmont & Mitchell, 1987).

Active strategy use requires the ability and intention to

plan one's performance in advance. General skills are

needed to engage voluntarily in mnemonic activity; specific

skills are required to carry out specific mediational

routines (Brown, 1974).

Criticisms of the strategy use approach as a

comprehensive explanation of learning and memory have

centered on the inability to define consistently and measure

strategy use, the inconsistent application of the construct,

and the view by some researchers that strategy use offers a

total, rather than partial, theoretical explanation for

learning and memory deficiencies (Belmont & Mitchell, 1987).

Strategy Use and Mental Retardation

Persons with mental retardation have been shown to

exhibit both qualitative and quantitative limitations in

employing rehearsal, clustering, and other organizational

strategies to aid in memory and learning (Brown, 1974;

Gerjuoy & Spitz, 1966). Deficits have been attributed to

(a) failure to employ strategies spontaneously, (b) the

application of strategies in a way different from their

original intent, or (c) an unawareness of the need for

strategic intervention (August, 1980). The active use of

rehearsal strategies may depend on the attributes of the

person, the instructions received, the perception of task








difficulty, or the particular task or situation (Belmont &

Mitchell, 1987; Brown, 1974; Ellis, 1970).

Limited success has been shown in increasing recall

abilities of persons with mental retardation through the use

of categorical lists, temporal blocking, generation of

mediating sentences for word triads, and rehearsal training

(Glidden et al., 1983). Organizing information by recurring

redundant patternings can reduce information processing

demands and decrease memory load. Persons with mental

retardation are less efficient at recognizing and using

organizational patterns, however, unless redundancy is high

(Spitz, 1966) or they are given explicit instructions to

attend to conceptual relationships (August, 1980). Brown

(1974) observed that failure to employ rehearsal strategies,

even after training to do so, may be a function of capacity

limitations.

Gerjuoy and Spitz (1966) explored the efficacy of

training programs to increase the ability to organize and

cluster input. They found that this training did facilitate

immediate recall. The ability to generalize this

organization to new tasks, the long-term effectiveness of

such training, and the efficiency of rehearsal strategies in

enhancing elaborative processing appear limited, however

(Brown, 1974; Engle & Nagle, 1979; Glidden et al., 1983).

The lack of ability to retain and transfer strategy use to a

novel situation has been cited as a defining characteristic

of mental retardation (Belmont & Mitchell, 1987).








Strategy Use and Automaticity

Although much real-world information is automatically

encoded and depends on free-recall retrieval, focusing

attention on the use of control processes has highlighted

teaching persons with mental retardation to produce and

spontaneously use strategies to decrease performance

deficiencies (Glidden et al., 1983). Nonetheless, strategy

instruction may decrease future occurrences of deficits in

incidental learning (Dugas, 1976; Fox & Rotatori, 1979).

The inability to use organizational strategies may result in

an inability to screen out irrelevant information or attend

to appropriate mediators in novel situations (Bilsky,

Whittemore, & Walker, 1982).

Application of Memory Models

In this part of the chapter, literature relevant to the

study of the effect of encoding condition on recall by

persons with mental retardation is reviewed. Selection of

studies for the review was based on both the construct and

the subjects investigated. A study had to meet the

following criteria to be included in the review.

1. The study had to include a test of memory

functioning. Tests were included for both short-term and

long-term measures of both recall and recognition memory.

2. The subjects must have been thoroughly described

and some of the participants had to include persons with

mental retardation, between the ages of 10 and 21. Studies







35

that involved participants exclusively who were younger than

10 or older than 21 were not included.

3. The study must have been empirical in nature with

the intervention or tests used described in sufficient

detail to permit replication.

4. The study must have been cited in the literature

since 1967.

A total of 138 studies were located for this review.

The analysis of the findings of these studies is organized

to reflect the application of encoding conditions and the

theories of memory processing to persons with mental

retardation. Therefore, the findings of the studies are

synthesized by the following areas: (a) encoding condition,

(b) developmental trends, (c) structural processes, and (d)

strategic processes.

Encoding Condition

Encoding condition refers to the establishment of an

internal representation of a stimulus for storage and

retrieval. Several investigators have studied factors that

influence encoding and the effect of encoding condition on

recall. Results of investigations of the effect of encoding

condition on memory of students with mental retardation are

organized around tests of encoding condition and influences

to information encoding.

Tests of encoding condition. From a study that included

30 students with mild mental retardation (average IQ 62.5)

and 30 nonretarded peers, Ellis et al. (1985) reported that








persons with mental retardation were encoding-deficient

because fewer items were encoded by the students with mental

retardation. Once encoded, item durability was unrelated to

IQ, however. Ellis et al. further determined that students

without retardation were able to encode faces more

accurately than students with retardation. The researchers

had predicted that this variable would be encoded

automatically and, therefore, encoded equally well by both

persons with and without retardation. In addition, those

students who claimed to use strategies had more accurate

recall. These results favor a general interpretation that

persons with mental retardation experience a difference in

automatic processing, according to these authors. Diech

(1974) also provided evidence for deficits in automatic

processing and incidental learning in students with mental

retardation following research involving 18 students with an

average IQ of 51.0 and 18 kindergarten students. Diech

pointed out, however, that it was important to note that the

students with retardation did experience some incidental

learning.

In contrast to the findings of Diech, Schultz (1983),

who studied 12 students classified as educable mentally

retarded whose average IQ was 65 and 12 children without

mental retardation, affirmed that the students with mental

retardation who received an unexpected recognition test

retained information as well as peers without retardation.

No deficits in incidental learning were apparent. Hillman







37

(1972) used a total of 90 students between the ages of 9 and

13 to study the effects of question type and position on the

learning among children with mental retardation. No

differences between groups matched for mental age on an

incidental criterion test were found. Song and Song (1969)

evaluated the effectiveness of discriminating reading skills

based on the visual memory ability of 26 adolescents

classified as high readers (average IQ of 60) and 28

adolescents classified as low readers (average IQ of 59).

Reproducing designs from memory, an incidentally coded

event, was determined to be one discriminator of reading

ability. Reading achievement of persons with mental

retardation was not related as much to their ability to

perceive but to their ability to remember what was

perceived.

In an investigation of the level of processing

framework, Lupart and Mulcahy (1979) investigated incidental

and intentional learning of 42 students classified as EMH

and 42 nonretarded students from the forth, fifth, and sixth

grades. Only partial support was expressed for the

hypothesis that recall performance was positively related to

depth of processing. A significant difference in recall was

noted between incidental and planned intentional conditions

but not between incidental and intentional conditions.

Depth of processing did influence the durability of memory.

Factors influencing encoding. Incidental learning was

influenced by prior instructions on categorization in a








study by Fox and Rotatori (1979). Dugas (1976) included 40

institutionalized students with mental retardation in an

investigation of recall differences of familiar and control

stimuli due to subject-paced and experimenter-paced

presentations. Recall followed the classic, bowed serial-

position curve; recall differences due to stimulus

presentation and pretraining were restricted to primacy

effects. Subject-paced presentation also affected primacy

effects for students without mental retardation.

Familiarization training to acquaint subjects with the

stimuli did not affect primacy effects for students with

mental retardation but did for those without retardation.

The combined results of this study support the hypothesis of

the positive effect of rehearsal on both encoding and recall

of primacy events.

Mar and Glidden (1977) used free- and cued-recall tasks

to study the effect of semantic and acoustic encoding on

seven encoding groups. Each group consisted of 14 students

and each group had a mean IQ of 60. Both semantic encoding

and semantic retrieval cues were needed to produce maximum

recall. Semantic cues did not influence free recall

patterns. Therefore, the authors concluded that persons

with mental retardation may not be able to use elaborative

encoding processes or retrieve items once elaboration has

been used.

Bray and Ferguson (1976) studied the effect of cues to

forget previously presented information with students with








mental retardation. In experiment I, 16 students with an

average IQ of 66.8 and average age of 9.8 were compared with

22 nonretarded students with an average age of 7.4. In a

replication of experiment I, a second group of 16 students,

average IQ of 70.6 and average age of 10.7, were compared

with 24 nonretarded students with an average age of 6.9.

Passively acquired information was easily lost. This

passively acquired information interfered only minimally

with recall of actively acquired information. Bray, Turner,

and Hersh (1985) also investigated the developmental course

of selective remembering. Forty-eight students, 16 from each

of three age groups, were studied. Students 11 years of age

without mental retardation were able to use forget cues, but

those of similar ages with retardation (mean IQ of 69.3)

were not able to take advantage of such cues. Two

additional groups of students with mental retardation were

studied by these researchers. Those with an average age of

15 and those with an average age of 18 were not able to

eliminate interference from to-be-forgotten items although

this skill was available to study participants of similar

ages without retardation.

Differences in serial position effects following fixed-

set or variable-set procedures led Phillips and Nettelbeck

(1984) to conclude that the encoding procedures used affect

the performance of children and persons with mental

retardation more than encoding procedures affect nonretarded

adults. Ten students with retardation (average IQ of 67),







40

10 nonretarded students from the third and forth grades, and

10 college students were included in this investigation.

Additionally, the type of stimulus used had a profound

influence on the encoding scan rates of students with mental

retardation in a test of high-speed memory scanning

conducted by Maisto and Jerome (1977). Six students without

mental retardation and six students with mild retardation

(average IQ 72.4) were involved in this investigation.

Merrill (1992) examined differences in the speed of

encoding of 15 students with an average IQ of 60.2 and 15

college undergraduates. Merrill determined that differences

in processing speed resulted from encoding that required

attentional resources. The size of memory load influenced

encoding times for all subjects. Decision processes were

executed more efficiently when resources were allocated to

their execution. These findings do not support the

assumption that familiar stimuli are encoded automatically.

Merrill concluded that there may be memory processes that

can be executed without attentional resources if necessary,

but are performed better and faster when resources are

allocated.

In support of theories of automatic processing, memory

for spatial location was unrelated to age, intelligence, or

type of instruction in a study by Ellis et al. (1989a). A

total of 28 students from the second grade, sixth grade,

college students, and students with mental retardation

(average IQ of 62.9, average age of 16.41) were included by








Ellis. The inability to relocate items was not inherent of

either low IQ or Down syndrome. Both age and intellectual

trends were shown in the number of items recalled, however.

Summary. Studies of the effects of encoding conditions

on recall have achieved mixed results. Hillman (1972) and

Schultz (1983) offered support for Hasher and Zacks (1979,

1984) theory of automatic and effortful processing when

their research produced no deficits in incidental processing

of information by persons with mental retardation.

Automatic encoding deficiencies were acknowledged by Diech

(1974) and Ellis et al. (1985).

Possible explanations for these discrepant findings

were offered by several researchers who demonstrated that

encoding of information and therefore, incidental learning,

may be influenced by prior instruction (Dugas, 1976; Fox &

Rotatori, 1979), the procedures used (Maisto & Jerome, 1977;

Phillips & Nettelbeck, 1984), the size of the memory load

(Merrill, 1992), the type of information to be encoded

(Ellis et al., 1989a), and the ability to benefit from cues

(Bray et al., 1985; Mar & Glidden, 1977).

In the present study, the effect of encoding conditions

on recall was investigated. Both immediate recall and

durability of memory were used to assess if differences

exists across developmental levels with actively and

passively acquired information. Differences in recall and

retention in item identification and item relocation would

support the theory of automaticity because evidence would be








provided that these two attributes are processed

differently. Differences in locating recalled and

unrecalled items would suggest further independent

processing of these two attributes and provide additional

support for the theory of automaticity.

Developmental Trends

Differences in recall abilities based on developmental

levels have been investigated by several researchers.

Researchers have compared the memory performance and use of

memory strategies by students with different classifications

of mental retardation with students matched for mental age,

or nonretarded peers of similar chronological ages.

Trends in memory processing. Calfee (1969) investigated

short-term retention as a function of memory load with

groups of students classified as moderately, severely, and

educable mentally retarded. Nine students with an average

IQ of 65, 12 students with an average IQ of 44, and 10 with

an average IQ of 37 were studied. Recall performance and

ability to encode and organize information decreased with

age and IQ while forgetting increased with age and IQ.

Recognition memory remained fairly constant.

Baumeister (1974) analyzed developmental trends across

a wide variety of materials and modalities on tests of

serial memory span thresholds with 15 students from each of

three groups (second graders, sixth graders, and students

with mental retardation with an average IQ of 71.7 and an

average age of 11.0). Lamberts (1979) showed that the








memory spans of adolescents with mental retardation (35

students, average IQ of 37.8, average age 14.58) were

comparable to the memory span of students matched for mental

age (25 students from kindergarten and nursery school,

average age 4.42) when the stimuli were not abstract. The

students with mental retardation did less well when the

stimuli were linguistic. The strong age effect supports the

hypotheses of the growth of memory span with age (Lamberts,

1979). McLaughlin, Stephens, and Moore (1971) used a visual

memory task (reconstruction of geometric shapes) to evaluate

reconstructive memory systems. Forty-eight students with an

average IQ of 90-110 and 48 with an average IQ of 50-75 were

studied. Both groups included students from age 8 to 18.

As the age of the subject increased, memory drawings more

closely resembled the original configuration. These

researchers suggested that reconstructive memory systems are

developed prior to the development of evocative systems.

Stephens, Anderson, and Garrison (1971) observed that

the performance of 35 nonretarded study participants matched

for both mental age and chronological age exceeded the

performance of students with mental retardation on a

seriation task (reproducing pictures of stimuli arranged by

decreasing size). Improvement following instruction did not

occur in the youngest students with mental retardation,

leading the authors to conclude that seriation schema were

incompletely developed. Improvements by subjects without








retardation at a 6-month follow-up test was indicative of

the establishment of a seriation schema.

Brown, Campione, and Barclay (1978) trained two groups

of students classified as EMH on anticipation and rehearsal

strategies that included a self-checking component. Group

one included 21 students, aged 10, with an average IQ of 70.

Participants from group two were 13 years of age and also

had an average IQ of 70. Compared to a control group,

students trained in self-checking routines took more time

studying, recalled more from a prose passage, and recalled

information more clearly related to thematic constructs from

the passage. Brown and her colleagues pointed out that

these patterns are characteristically seen in students

developmentally more advanced. Lower-functioning students

(mean mental age of 6) showed no effects of the training at

the 1 year follow-up. More advanced students (mean mental

age of 8) both maintained original rote recall ability and

generalized the ability to recall the gist of the prose

passage to novel situations.

Trends in strategic skills. In an investigation of the

relationship of mental age to retentional capacities and the

use of retention strategies, MaBane (1976) observed that

both retention capacity and strategy use were directly

related to level of intelligence. The results of this

study, which included a total of 18 students of two levels

of retardation (group one mean IQ 74, group two mean IQ of

50), reflect short-term memory capacity constraints on the








use of rehearsal techniques related to developmental level.

Students with moderate and severe retardation did not

benefit from external organization (categorical clusters)

although other researchers had found this technique to be

successful in aiding students with EMH to increase recall

(Burger & Erber, 1976). These researchers, who observed 16

students with an average IQ of 39.12, hypothesized that

lower functioning students were not able to understand

organizational or categorical labels.

Swanson (1977) used both pictures and actual objects

for a serial recognition tests with 10 students classified

as learning disabled (mean IQ 100.3) and 10 students

classified as EMH (mean IQ 73.3). The students classified

as learning disabled were able to make use of three-

dimensional material while the EMH students could not

effectively process the same concrete material. No

differences were seen in recognition memory for pictures.

Sugden (1978) reported that developmental trends in the

use of rehearsal strategies apparent in 15 nonretarded

students from each of three grade levels (6th, 9th, and 12th

grades) were almost totally absent in a similar number of

students with mental retardation matched for chronological

ages. Lupart and Mulcahy (1979), who studied a total of 84

students, contended that the ability to increase memory

performance through adopting memory strategies may be more

related to mental age than to IQ. No significant

differences in gains in recall across incidental,








intentional, and planned-intentional conditions were noted

by these researchers. Engle and Nagle (1979) identified

semantic processing deficits with mildly retarded students

13 years of age but not with those 10 years old. They

determined that to be able to use semantic encoding

strategies, the student must be able to use externally

provided prompts and learn to provide prompts for internal

mediation. In their study, developmental trends were shown

in these abilities. The IQ's of study participants ranged

from 50 to 75. Engle and Nagle concluded that the semantic

network is not sufficiently developed by age 10 to allow

categorical decisions to facilitate recall. This hypotheses

is supported Reichart, Cody, and Borkowski (1975), who

studied 44 students with mental retardation whose IQ's

ranged from 30 to 70. They found that only those students

with an IQ greater than 50 were able to acquire and transfer

strategy use.

By contrast, even the youngest students in a study by

Becker and Morrison (1978) had increased memory performance

when encouraged to use categorization skills. Measures of

clustering were studied with 57 nonmentally retarded

students, 87 students classified as learning disabled, and

91 students classified as EMH. Students with lower mental

ages needed assistance in organizing materials while

students with higher mental ages perform better when allowed

to structure their own organization.








Summary. Developmental trends in recall and

recognition memory have not been produced consistently;

however, researchers generally have described the ability to

organize, encode, and retrieve information as related to

both age and developmental level (Calfee, 1969; McBane,

1976; Swanson, 1977). Developmental trends in memory

performance may be the effect of the modalities and

materials employed (Baumeister, 1974; Lamberts, 1979;

McLaughlin et al., 1971) or the inadequate use of strategies

(Engle & Nagle, 1979; Lupart & Mulcahy, 1979; Reichart et

al., 1975; Stephens et al., 1971; Sugden, 1978). Evidence

of the ability of younger students with mental retardation

to benefit from strategy instruction has been inconclusive

(Becker & Morrison, 1978; Burger & Erber, 1976).

In the present study, the memory processing of students

of three developmental levels was assessed. Of specific

interest was the effect on memory of level of retardation

under various encoding conditions. If developmental trends

were evident in automatically processed information, support

would not be shown for the theory of automatic and effortful

processing because, according to Hasher and Zacks (1979,

1984), age and ability do not affect the memory of

automatically processed information.

Structural Processes

Structural processes describe how information is

processed, retained, and flows among the various memory

stores. In this section, studies in which researchers








assessed structural processing mechanisms and structural

deficits in students with mental retardation are reviewed.

Structural processing. In a classic study of structural

versus strategic processing, Glidden and Mar (1977)

evaluated the ability of 80 students with and without

retardation to retrieve information from storage. Students

with retardation (average IQ of 60) retrieved about 60% of

the number of items retrieved by those without retardation.

This difference in recall rates could not be accounted for

solely by the number of items available in storage. It was

hypothesized that both accessibility and availability

deficits existed.

Persons with and without mental retardation processed

auditory and visual information in similar ways, according

to Katims (1987b). Twenty-four students with educable

mental retardation and 24 similar age peers were studied.

Quantitative differences were noted under all three

treatment conditions used by this researcher. No

differences between auditory and visual memory span for

letters resulted when 15 students classified as trainable

mentally retarded (IQs ranging from 27 to 55) were tested

(Varnhagen, Das, & Varnhagen, 1987). Students with Down

syndrome (13 students with IQs from 28 to 45) did not differ

from those with other etiologies. In addition, Varnhagen et

al. demonstrated that persons with mental retardation have

less efficient information processing, faster stimulus

decay, and greater stimulus interference on both recall and








recognition tasks. These findings were offered as evidence

of structural short-term memory deficits. Reid and Kiernan

(1979) also reported modality-related encoding deficiencies

when examining the abilities of six students with severe

retardation (average IQ of 47) to encode spoken words and

manual signs.

Structural deficits. Stratford (1979) determined that

test behavior was similar for all three groups studied in a

test of discrimination memory as participants from all

developmental groups appeared to attempt to rehearse input

and not randomize stimuli presented. Study participants

included students with Down syndrome (average mental age of

3.8), students with mental retardation with other etiologies

(average mental age of 3.4), and nonretarded students

(average mental age of 5.7). The author confirmed poor

short-term memory existed for size, form, and order for

students with mental retardation, both with and without Down

syndrome.

Studies of secondary memory deficits have led several

researchers to conclude that, for persons with mental

retardation, primary memory is intact. Secondary memory

deficits may result from deficits in rehearsal strategies

that mediate the transfer of information from primary to

secondary memory (Ellis et al., 1985). Dugas and Kellas

(1974) determined that the 14 students with mental

retardation they studied (average IQ of 71.75) were at a

distinct disadvantage in the use of primary memory, however.









On an immediate recognition memory test, students with

mental retardation had higher average reaction times for

retrieving information and for making serial comparisons.

Burack and Zigler (1990) used two tests of intentional

memory and one test of incidental learning to investigate

differences in the memory functioning of persons with

organic and nonorganic retardation. Thirty-five students

without retardation (average age 9.34, average IQ of 94.63),

33 students with familial retardation (average age 13.49,

average IQ 63.64), and 40 students with organic retardation

(average age 15.78, average IQ 59.2) were studied. Results

on incidental learning tasks were no different from

intentional test results. Etiology should be considered

when studying memory processing differences, according to

these authors, due to differences in structural systems,

differences in utilization of structural capabilities, and

differences in task-related motivations. Reichart et al.

(1975) provided evidence that strategy use and recall did

not differ by etiology when they examined rehearsal strategy

use by students whose IQ ranged from 30 to 70.

Engle and Nagle (1979) interpreted the results of a

study of encoding strategy use by 42 students with mild

retardation (IQ range 50-75) as supportive of deficits in

declarative knowledge. Declarative knowledge was defined as

the amount and complexity of knowledge existing in long-term

memory and the number of associations automatically elicited

when an item is presented for memory. Engle and Nagle








described deficits in declarative knowledge as "not quite

but almost both structural and strategic" (p. 28).

Deficits, according to Engle and Nagle, are not structural

or physiological per se, but are based on automatic

processing not under conscious control. Strategy use, on

the other hand, requires conscious associations. With

practice and an increase in the number and richness of

available associations, automaticity develops. Ellis et al.

(1985) agreed that some nonstrategic tasks may depend on

semantic knowledge and, therefore, be sensitive to

individual differences. Engle and Nagle (1979) believed

that for students classified as EMH to remember as well as

nonretarded peers, instructional strategies should focus not

only on developing strategic repertoires but also on

equating the extent, richness, and automaticity of

declarative knowledge. Teaching techniques should be used

that enhance the amount of existing knowledge and

experiences (Engle & Nagle, 1979).

Glidden and Mar (1977) reported that one aspect of an

accessibility deficit in persons with mental retardation

involved failure to spontaneously use mnemonic strategies

consistent with semantic organization. Glidden et al.

(1983) echoed concerns expressed by Engle and Nagle that the

use of semantic strategies may be hindered by the lack of

comprehension or failure to understand the to-be-remembered

information. Glidden and her colleagues used two groups to

investigate the efficacy of semantic processing on free








recall. Group one consisted of 48 students with an average

IQ of 62.9. Participants from group two contained were 36

students with an average IQ of 68.1. Glidden et al.

concluded that the problems linking stimuli may account for

developmental delays in recall and comprehension.

Additionally, strategy use may increase semantic knowledge

on other tasks and thus be more beneficial than rote

rehearsal strategies.

Summary. Studies of memory and retrieval led Glidden

and Mar (1977) to conclude that both accessibility and

availability deficits may exist for persons with mental

retardation. Similarity in processing styles between

persons with and without retardation was seen by Katims

(1987b) and Stratford (1979) although quantitative and

modality related differences have been noted as well

(Katims, 1987b; Reid & Kiernan, 1979). Some researchers

have observed differences in recall ability related to

etiology, while no etiology-related differences were

identified by other researchers (Burack & Zigler, 1990;

Reichart et al., 1975).

Studies of structural processes of memory can provide

information concerning deficits in short-term memory, long-

term memory, or both. Stratford (1979) explained that, for

students with mental retardation, short-term deficits

existed in memory for size, form, and order. According to

the theory of automaticity, this information should be

encoded automatically and not sensitive to developmental








differences. The present study included analysis of short-

term and long-term memory of automatically and effortfully

processed information to determine if developmental trends

occur.

Strategic Processes

Several strategies have been shown to be at least

minimally effective in increasing memory processing for

persons with mental retardation. In this section, the use

of specific task strategies, clustering, and semantic or

elaborative strategies are reviewed.

Task-specific strategies. Overlearning, or additional

presentations of stimulus materials, was assessed with a

group of 36 students classified as educable mentally

retarded (average IQ 73.0) and 24 third-grade students

matched for mental age (Raskin, 1970). Overlearning was

necessary to overcome deficits in long-term perceptual

memory for the study participants with mental retardation.

Ross and Ross (1978) divided 33 students into three

treatment groups to determine the effectiveness of imagery,

rote repetition, and control conditions. The groups average

IQs were 65.73, 66.27, and 68.36, respectively. Imagery was

superior to rote memory training for increasing multiple

associate learning. The results of a study by Fox and

Rotatori (1979) supported earlier work by Brown (1974) that

specific instructions and task-specific strategies should be

used with persons with mental retardation to maximize

learning.









The positioning of questions in a story strategy

affected learning and recall in a study by Hillman (1972)

using 90 students aged 9 to 13. Questions which followed

presentations of story sections, close temporal proximity

between critical information to be learned and questions,

and questions of the same type for training and criterion

testing were most effective. When Burger and Erber (1976)

allowed 16 students with an average IQ of 39.12 to select

to-be-remembered items, item preference was more important

than practice effects in stimulating learning and recall.

Raskin (1969) indicated that prior training may hinder the

perception of apparent movement. Training was provided to a

total of 30 students. One group ranged in age from 8 to 13

(average IQ 71.8); the age range for group two was 11 to 15

(average IQ 73.6). More experience in training sessions was

required by students with retardation to establish enduring

perceptual learning.

For students with an IQ of less than 50, a cumulative-

rehearsal strategy was most efficient for inducing strategy

transfer when Reichart et al. (1975) studied 44 students

with IQs ranging from 30 to 70. For higher-functioning

students, a cumulative-clustering strategy enhanced learning

and transferred to new information for short-term recall,

but the use of this strategy dissipated by a 2-week follow-

up. Use of a cumulative-rehearsal strategy showed greater

consistency in duration patterns over 1-hour, 24-hour, and

2-week follow-up testing. Strategy acquisition and transfer







55
was limited by mental age as participants with lower IQs did

not acquire, transfer, or profit from active rehearsal

strategies.

Clustering. Gerjuoy and Spitz (1966) used two

experiments to investigate the growth of clustering and free

recall as a function of age, intelligence and practice. Two

groups of institutionalized students (20 students per group

with an average IQ of 72.05 and 52.95, respectively), 19

students matched for mental age (average age 9.81), 14

students matched for chronological age (average age 14.70),

and 20 college students were included in the study. Both

presenting stimulus words in categories and requesting

subjects to recall words in categories increased recall for

students with mental retardation. A combination of these

procedures produced the greatest increases. In a study of

organizational strategies to increase free recall of verbal

learning, Sitko and Semmel (1972) observed that 30 students

classified as EMH (average IQ 69.9) experienced less

clustering and recall on categorical lists and on stimulus

lists of highly associative noun pairs than 30 nonretarded

peer of similar chronological ages.

Evans (1977) used 24 students from each of three

groups to study associative-clustering tendencies. Students

from the sixth grade, seventh grade, and those with mental

retardation with a mean IQ of 73 comprised the groups. When

novelty-condition lists and organized-condition lists were

assessed to determine the ability to transfer associative-







56
clustering techniques, results indicated that recall process

factors were similar for students without retardation as for

those classified as borderline mentally retarded. Gerjuoy

and Alvarez (1969) studied the transfer of clustering

techniques from a list presented categorically to a task

requiring the recall of randomly presented words with 60

students with an average IQ of 59.4 and 60 nonretarded

peers. Both students with and without mental retardation

benefitted from external clustering imposed by the examiner.

Neither practice nor familiarity with the task increased

recall or clustering performance on the randomized list.

Baumeister (1974) illustrated that forced chunking

(paired-associate stimuli) increased the serial memory span

of 15 students with mental retardation (average IQ 71.7) as

well as 15 students each matched for mental and

chronological ages. Becker and Morrison (1978) attested

that blocking and sorting techniques increased measures of

clustering for both students classified as learning disabled

as well as those with EMH.

In a study comparing the use of prompts by 56 students

with mental retardation (mean IQ 74) and 56 fifth-grade

students, August (1980) determined minimum-sort prompts to

be as effective as maximum-sort prompts for improving recall

and clustering. Minimum cues included instructions to sort

words by category. Maximum-sort prompts included training

and practice in categorization. As no benefit resulted from

explicit instructions in generating organizational schemas,








August emphasized the importance in isolating factors

responsible for IQ-related disparities in memory

performance.

Semantic and elaborative strategies. Engle and Nagle

(1979) compared the techniques of semantic encoding,

repetitive rehearsal, and acoustic encoding (thinking of the

sound of the word and repeating the initial sound) for

increasing the recall of 42 students with IQs ranging from

50 to 75. Free recall was enhanced by the use of semantic

encoding on both training and unprompted follow-up word

lists. No difference in strategy use was seen on a recall

test 7 months later until the semantic strategy was

prompted. Following prompts, the semantic group showed

greater improvement. Students with mental retardation

required additional time to construct rich, more elaborative

memory traces when Schultz (1983) compared 12 students with

an average IQ of 65 to 12 nonretarded children. As the

students classified as EMH were increasingly slower in

encoding at progressively deeper processing levels, Schultz

asserted that persons with mental retardation either have

difficulty in combining information or the interaction

between the subject, processing level, and decision time is

indicative of short term memory deficits.

Riegel, Taylor, and Danner (1973) used two groups of

students classified as EMH to study the effect of mnemonic

strategy use on language development. Group one consisted

of 29 students with an average IQ of 72.3. Group two had 32







58
participants with an average IQ of 72.5. Following training

to seek and utilize associations between stimuli, no

differences were demonstrated between the groups on sorting

and recall measures.

Merrill and Bilsky (1990) used subject, verb, and

subject-verb cues to encode sentences semantically into

memory with 33 students with mild retardation (average IQ

62.9), 33 nonretarded 4th-grade students, and 33 nonretarded

10th graders. All groups showed some advantage to using

two-word, subject-verb cues; students with mental

retardation showed the smallest recall gains. The ability

to utilize holistic representations, therefore, was

determined to be a function of both age and intelligence.

It was hypothesized that persons with mental retardation may

not be able to perceive sentences as a single, integrated

unit.

Story strategies were used to increase semantic

encoding in two studies by Glidden and Warner (1983, 1985).

In the first analysis, 60 students with an average IQ of

66.63 were studied. The second experiment included 24

students with an average IQ of 66.7. Stories generated by

the study participants facilitated recall of those students

without retardation. This semantic strategy failed to

increase the learning and memory of those participants with

mental retardation, however, unless order of recall was

disregarded. Instructions to recall stimuli in serial order

appeared to change how the information was processed. When








order of recall was disregarded, less reliance was made on

input organization and flattened serial position curves

resulted. Follow-up from both of these studies affirmed

that the advantages of semantic story use was maintained

over short intervals but faded at longer (8-month)

intervals. An added finding of these studies was that, as

recall and retention were related to the quality of stories

produced, persons with mental retardation seemed to do

better when mediators were provided externally. Subject-

created stories involved only minimum semantic processing

while experimenter-created stories increased retention and

immediate recall. Additional training may be needed in

story creation and creating meaningful retrieval cues (e.g.,

story titles) if subject-composed story strategies are used

with persons with mental retardation (Glidden et al., 1983).

Summary. Several strategies have been studied with

students with mental retardation. Overlearning, imagery,

single- and multiple-word cues, rehearsal and clustering,

and questioning techniques have been used to increase memory

performance (Hillman, 1972; Merrill & Bilsky, 1990; Raskin,

1970; Reichart et al., 1975; Ross & Ross, 1978). Both

semantic encoding and story strategies have shown promise

for increasing memory through increasing depth of processing

(Engle & Nagle, 1979; Glidden & Warner, 1983; 1985) although

questions exists concerning the durability of these

techniques. Brown (1974) warned that strategy training







60
should be task specific to be effective for students with

mental retardation.

In the present study, the use of semantic and non-

semantic encoding, clustering, and clustering-probed recall

techniques were investigated on a picture recall and

relocation task to determine if these strategies were

affected by developmental differences and produce

differences in short-term and long-term retention. If no

differences were exhibited for recall of location across

encoding conditions, support would be evidenced for the

theory of automaticity. Hasher and Zacks (1979, 1984)

hypothesized that input for spatial location will be

automatically processed and, therefore, not sensitive to

instructions or intention to process.














CHAPTER 3
METHODS AND PROCEDURES

In this study a picture recall and relocation test was

used to assess the effect of encoding condition and

developmental level on memory of students with and without

mental retardation. Performance on recall and relocation

measures was compared across encoding conditions and

developmental levels to determine support for the theory of

automatic processing as hypothesized by Hasher and Zacks

(1979, 1984). Three experimental encoding conditions were

used: nonsemantic, semantic, and clustered. Participants

of three developmental levels were assessed: educable

mentally handicapped (EMH), trainable mentally handicapped

(TMH), and peers without mental retardation of similar

chronological ages. The recall and relocation tests were

completed with each participant with a follow-up test of

secondary memory completed approximately 24 hours after the

initial testing. This chapter is divided into five

sections: Null Hypotheses, Selection and Description of

Participants, Research Methods, Procedures, and Data

Treatment.

Null Hypotheses

The following null-hypotheses were tested at the .05

level of confidence for each main effect or interaction:







62
HI There is no statistically significant difference in the

mean number of items recalled following presentation of

50 pictures as a function of (a) three developmental

levels of THM, EMH, and peers without retardation of

similar chronological ages; (b) three encoding

conditions of nonsemantic, semantic, and clustered; or

(c) two retention intervals of immediate and 24-hour

recall.

H2 There is no statistically significant effect on the

mean number of items recalled following presentation of

50 pictures as a function of the two-way or three-way

interaction of the (a) three developmental levels of

THM, EMH, and peers without retardation of similar

chronological ages; (b) three encoding conditions of

nonsemantic, semantic, and clustered; and (c) two

retention intervals of immediate and 24-hour recall.

H3 There is no statistically significant difference in the

mean number of items relocated following presentation

of 50 pictures as a function of (a) three developmental

levels of THM, EMH, and peers without retardation of

similar chronological ages; (b) three encoding

conditions of nonsemantic, semantic, and clustered; or

(c) two retention intervals of immediate and 24-hour

recall.

H4 There is no statistically significant effect on the

mean number of items relocated following presentation

of 50 pictures as a function of the two-way or three-








way interaction of the (a) three developmental levels

of THM, EMH, and peers without retardation of similar

chronological ages; (b) three encoding conditions of

nonsemantic, semantic, and clustered; and (c) two

retention intervals of immediate and 24-hour recall.

Selection and Description of Participants

Participants of this study were 180 students enrolled

in a large public school district in west central Florida.

In the present study, the memory functioning of adolescents,

aged 10 through 15, was assessed to keep with the findings

of several researchers that the ability to use control

strategies may not develop until early adolescence and may

be further delayed for persons with mental retardation

(Brown, 1974).

The 180 participants constituted three groups: 60 EMH,

60 TMH, and 60 peers without retardation of similar

chronological ages. These groups were developed by (a)

development of a pool of potential participants for each

developmental level who met the age and IQ or CTBS

restrictions and for whom permission to participate had been

received, and (b) random selection of 60 participants per

group from the pool for each developmental level. Within

each group, students were randomly assigned to one of three

experimental encoding conditions.

The names of 164 students without mental retardation

were made available to the researcher by the district.

These students received requests for consent to participate.









From the 103 who furnished parental consent, an available

pool of 85 resulted after it was determined that the

predefined age and CTBS parameters were met. A random

number selection procedure was used to select 60

participants for that developmental level.

The same procedures were followed for the students

classified as EMH and TMH. Of the original 139 names of

students classified as EMH made available by the district,

75 comprised the available pool of those with parental

consent and meeting the age and IQ eligibility criteria.

For students classified as TMH, the names of 88 students

were made available by the district. An available pool of

73 resulted from which the random selection procedures were

used to select the 60 participants.

Classification of the students with mental retardation

was based on local district classifications. Participants

classified as THM were required to have an IQ score of 30

through 55. Participants classified as EMH were restricted

to students with an IQ score of 59 through 73. The

participants who were not retarded were required to have

scores indicating functioning on grade level on the

Comprehensive Test of Basic Skills (CTBS) and not receiving

special education services. These restrictions were imposed

in order to decrease the possibility of the overlap of IQ

scores among the groups. The IQ scores were obtained from

psychological records available from the school district







65
and, for participants classified as EMH or TMH, these scores

must have been no more than 4 years old.

Research Methods

Participants were solicited from those enrolled in the

cooperating school district who met the age and

classification guidelines. Prior to conducting the study,

the respective approval of the University of Florida

Institutional Review Board and the Director of Planning and

Policy for the Pinellas County School District were obtained

and are contained in Appendix A. A Parent Informed Consent

Document (Appendix B) was sent to each parent and informed

consent was obtained for each participant.

Each participant was seen individually. Approximately

30 minutes were required to complete the experiment. A

follow-up test probe, which paralleled the original

experiment, was conducted approximately 24 hours following

initial testing.

The procedures used in this experiment are based on

techniques used by Ellis et al. (1987) and Gerjuoy and Spitz

(1966). Ellis et al. used recall of picture identification

and picture location to test automatic and effortful

processing. Gerjuoy and Spitz (1966) studied clustering of

nouns to determine the effect of externally provided

associative clustering on recall. Pictures were used for

the current experiment because the use of written words

limits neutrality of the response task with persons with

mental disabilities (Lamberts, 1979) and pictures are less








likely than written words to invite the use of strategies

which interfere with automatic encoding (Katz, 1987).

The stimulus for this experiment consisted of a book of

pictures. The book contained 50 pictures, plus an 8-picture

practice set. Plastic photo pages were lined with black

posterpaper and inserted into a spiral ringbinder. When

open, 2 pages provided an area of 37 cm by 26 cm and were

divided into quadrants by the ringbinder and a horizontal

white line. Four pictures were included in the 2-page area,

one in each quadrant. The pictures were all black and white

and were selected to represent up-to-date and easily

recognizable reproductions of common objects. The 50

pictures represented five items from each of 10 conceptual

categories. Table 1 lists the pictures used as stimuli.

An 8-picture sample task was contained in the first 2

pages of the book followed by 12 pages of four pictures each

and 1 page of two pictures. Two blank pages, with quadrants

outlined, were included at the end of the book for the

relocation test. The pictures were randomly arranged except

that the four pictures in a viewing area each came from a

different conceptual category. A set of 50 randomly

arranged pictures identical to those in the book was used

for the relocation test.

To ensure recognizability of the pictures, that objects

and conceptual categories were within the repertoire of

typical students meeting the selection criteria, and that

the scripted instructions were specific, comprehensive, and









Table 1
Pictures Used as Stimuli


Animals
Cat
Cow
Fish
Horse
Rabbit

Toys
Ball
Bat
Bicycle
Kite
Doll

Kitchen
Utensils
Cup
Fork
Pan
Spoon
Toaster

School
Supplies
Book
Computer
Pencil
Ruler
Scissors


Body Parts
Ear
Eye
Foot
Hand
Lips

Clothing
Belt
Dress
Hat
Shoe
Suit


Furniture
Bed
Chair
Desk
Sofa
Table


Food
Apple
Hamburger
Ice Cream
Milk
Pie

Vehicles
Airplane
Car
Fire truck
Train
Truck


Tools
Hammer
Ladder
Paintbrush
Saw
Shovel


understandable to the participants, a pilot test of the

memory task and recall and relocation phases was conducted.

Participants in the pilot study were 5 students without

mental retardation aged 10 through 15 and 10 students aged

10 through 15 classified as TMH. Any pictures not easily

recognizable were replaced with others that had been readily

named by the pilot participants. The scripted instructions

were revised as needed based on feedback from the pilot

test.








Procedures

To ensure reliability in presentation to each

participant, scripted instructions were developed and typed

on heavy paper for ease in handling by the experimenter.

The experimenter was either the author, an undergraduate or

graduate education student, or a person experienced in

working with students in special education. Special

training was provided to the examiners which consisted of a

review of the procedures with the author, observations of

sessions conducted by the author, and observations of

sessions conducted by each examiner until 100% reliability

in administration procedures was obtained. Reliability

among experimenters was maintained by the interrater

reliability procedures outlined in the Data Treatment

section of this chapter.

There were four phases for each participant: memory

task, recall test, relocation test, and follow-up. Figure 1

displays the phases.

The follow-up test of secondary memory was administered

by the examiner administering the initial memory recall and

relocation tasks. Each session was recorded on an audio

cassette recorder for use in determining reliability of both

encoding instructions and recording of responses.

Instructions for examiners are included in Appendix C. The

complete scripted instructions for each experimental

condition are presented in Appendix D. Data sheets used for

recording responses are included in Appendix E. Data sheets










Encoding Condition

Nonsemantic Semantic Clustered





Memory Task



Recall Test

Test of Effortful Processing
Yield: Number of Items Recalled




Relocation Test

Test of Automatic Processing
Yield: Number of Items Relocated




Follow-Up
Recall Relocation


Figure 1. Experimental Procedures Based on Encoding
Condition


and scripted instructions were color coded and assembled in

packets based on encoding condition for ease in handling and

increased reliability.

Memory Task

Activity 1: Encoding instructions. The experimental

condition assigned to each participant determined the

encoding instructions received by that participant.

Participants from Group 1, the nonsemantic-encoding

instruction group, were told to "name each picture."









Participants from the semantic-encoding condition, Group 2,

were told to name each picture and either tell what the

object is used for or tell something interesting about the

picture. Participants from Group 3, the clustered encoding

group, were asked to name each picture and then name another

item from the same conceptual category. All participants

from all groups were told that they would be asked to recall

as many pictures as possible following completion of the

task.

Activity 2: Materials. Participants were asked to

look at the pictures at a self-paced rate. They were not

allowed to turn back to the pages previously seen.

Activity 3: Practice set. Participants were shown the

8-picture practice set and were asked to recall as many

pictures as possible. Participants not able to name any two

pictures from the practice set were excluded from the

remainder of the task. An additional participant was

randomly selected from the available pool of potential

participants meeting selection criteria to replace any

participant excluded due to inability to name the items in

the pictures presented. No mention of location was made in

the instructions.

Activity 4: Task. When each subject was comfortable

with the procedures of the experiment, the main task was

introduced. Participants were told that the task for this

part of the experiment was the same as the practice set,

just longer. If the participant was unable to name a









picture, the experimenter provided the name. Appropriate

substitutions (e.g., substituting shoe for boot) were

accepted and noted on the data sheet by the examiner.

Inability to name pictured items correctly could indicate

difficulty encoding item information or lack of

understanding of task requirements. Both possibilities were

seen by Ellis et al. (1987) as task-related variables which

could interfere with tests of automatic processing.

Therefore, participants not able to name two or more

pictures were excluded from the test and an additional

participant was randomly selected from those meeting the

eligibility criteria and for whom parental permission had

been received.

Recall Test: Test of Effortful Processing

Activity 1: Recall activity. After viewing and naming

the pictures, participants were given 4 minutes for recall.

Free recall was requested from participants from Groups 1

and 2. For free-recall procedures, participants were asked

to name as many pictures as could be remembered. For

participants from Group 3, recall was requested in

categories. For example, participants in Group 3 were asked

to name as many vehicles as could be remembered.

Activity 2: Recall outcomes. As participants recalled

the items pictured, the names were recorded by the examiner.

Total study time was recorded as was position recall.

Incorrect names consistently used, for example, objects









named incorrectly and recalled with this same name, were

accepted.

Relocation Test: Test of Automatic Processing

Activity 1: Relocation activity. Following the 4-

minute recall interval, participants were asked to place the

pictures in their original locations using the blank pages

at the back of the picture book. Identical procedures were

used for item relocation for all experimental conditions.

A set of pictures identical to those in the picture

book were randomly arranged and used for the relocation

test. Participants were shown one picture at a time and

asked to name the quadrant where it had first appeared. No

feedback was given as to the correctness of the response.

Two or more pictures from the same location were not tested

consecutively.

Activity 2: Relocation outcomes. The picture location

named by the participant was recorded by the examiner to

determine the total number of pictures correctly relocated.

This task was performed to test the ability to recall

location when encoding had not been provided for that task.

Differential processing of item recall and item relocation

data would provide support for Hasher and Zacks' theory of

automatic and effortful processing. Further support for the

theory of automaticity would be offered if no differences

were noted in automatic processing of location across

developmental levels and encoding conditions.








Follow-up: Test of Secondary Memory

Activity 1: Recall follow-up. The recall exercise was

repeated approximately 24 hours after the initial testing.

During follow-up testing, the participants were not shown

the picture book. Only the recall and relocation phases of

the test were repeated. Participants were again given a 4-

minute recall period with free recall requested from Groups

1 and 2 and clustered recall requested from Group 3.

Activity 2: Relocation follow-up. Participants were

presented with a reordered version of the 50 pictures and

the relocation test was repeated using the same procedures

as used in the original relocation test. As test-wiseness

now existed for the relocation test, relocation data could

no longer be considered a measure of automatic processing.

This phase was repeated, however, to determine if

deterioration of automatically processed information

occurred differentially based on encoding condition or

developmental level.

Turnure (1991) reported an empirical basis for

identification of any test over 20 seconds as a test of

transfer of information from short-term to long-term

(secondary) storage. Turnure questioned the use of this

short-delay interval as a simple repetition test, however,

and recommended a minimum extended retest criterion interval

of 1 hour. For the current test, a 24-hour delay interval

was selected to be consistent with previous research using

similar recall and relocation exercises (Ellis et al., 1987,

Katz, 1987) and to avoid potential floor effects that might









exist with a longer delay interval with the group classified

as TMH.

Data Treatment

The recording of 20% of testing sessions was reviewed

to verify the accuracy in the presentation of instructions

as well as accuracy in the recording of the other measured

variables. Additionally, 20% of the data sheets were

reviewed to verify correctness of computations. All

reliability verifications were completed by persons who (a)

were independent of the examiner, (b) had advanced training

in special education, and (c) received specific training on

data recording procedures from the researcher. Any errors

located in either data recording or data computation were

corrected.

Interrater reliability of recordings by the

experimenter and independent transcriber was calculated by

the percentage agreement method of number of agreements plus

number of disagreements times 100. Percentage agreements

are shown in Table 2.


Table 2
Interrater Reliability



Recording of Data
Response Item % Agreement
Instruction presentation 97%
Number items recalled 99%
Number locations recalled 98.85%

Computation of Data % Agreement
Number items recalled 100%
Number items relocated 100%









For this experiment, dependent variables included

number of pictures recalled and number of picture locations

recalled. An optimum score of 50 was available for both

variables. Independent variables included encoding

condition (nonsemantic, semantic, and clustered),

developmental level (EMH, TMH, and nonretarded), and

retention interval (immediate and long term).

Analyses

Inferential statistics were conducted for each

hypothesis. The latest version of Statistical Analysis

Systems (SAS) was employed to complete data analysis

including Lindquist Type III (split-plot) analysis of

variance (ANOVA) and Tukey's procedures. The SAS program is

widely use in the field of social sciences for statistical

analysis. A .05 level of significance was chosen as the

criterion for rejecting or failing to reject the null

hypotheses.

The results of each hypothesis were analyzed by a

Lindquist Type III ANOVA procedure (developmental level by

encoding condition by retention interval) with repeated

measures on the last dimension. The Lindquist Type III

ANOVA design is appropriate for use when three factors are

involved and repeated measures are made across all levels of

one factor. The Lindquist design, also known as a 3-factor

ANOVA with repeated measures on one factor or a split-plot

factorial ANOVA, allows the researcher to answer seven

research questions: three concerning main effects, three








with first-order (two-way) interactions, and one with

second-order (three-way) interactions (Huck, Cormier, &

Bounds, 1974). The main effects were used to test

hypotheses one and three. Interaction effects were used to

analyze hypotheses two and four.

Tukey's studentized range statistic was used in post

hoc analysis to investigate significant effects and to

determine whether there were significant differences between

the means for each of the three developmental levels. The

Tukey procedure may be used in all cases where a significant

F is obtained in the ANOVA calculation and allows for

evaluation of all pair-wise comparisons (Bartz, 1981; Huck

et al., 1974). It can be used for groups of unequal size

(Bartz, 1981; Box, Hunter, & Hunter, 1978), and all tests

have a level of significance which is, at most, equal to

alpha. Thus, the possibility of Type I error, rejecting the

null hypothesis when it is true, is reduced (Winer, 1971).

Supplemental Analyses

Post hoc analyses were used to investigate the

independence of automatically and effortfully processed

information. Specifically, analyses were conducted to

examine retention and determine the relationship of study

time and intrusion errors to recall and relocation

performance. Retention was evaluated using a percent change

score to determine if deterioration in performance from day

one to day two occurred differently for recall and

relocation information. Pearson Product Moment Correlations








were calculated to determine if study time, the time used by

each participant to encode the 50-item set, was related to

either recall or relocation. The third post hoc analysis

investigated intrusion errors, or recalling items not in the

picture book. Correlation coefficients were calculated to

determine if the number of intrusion errors experienced by a

participant was related to his or her recall and relocation

performance.

Summary

In Chapter 3 the methods and procedures of the current

study were outlined. Data were collected through the use of

a picture recall and picture relocation task to determine if

the ability of students with and without mental retardation

to recall prior stimuli supported Hasher and Zacks' theory

of automaticity. A follow-up test of secondary memory was

conducted 24 hours after the initial recall and relocation

tasks. Inferential statistics were employed to describe the

variables and examine if differences occurred across

developmental levels, encoding conditions, and retention

intervals. Results of these procedures are presented in

Chapter 4.














Chapter 4
DATA PRESENTATION AND ANALYSIS

The results of the present study are presented in

Chapter 4. Results from the Lindquist Type III analysis of

variance (ANOVA) with repeated measures are provided and

analyzed as related to the hypotheses stated in Chapter 3.

This chapter is divided into four sections: Participant

Characteristics, Recall and Relocation Performance, Post Hoc

Analyses, and Summary of the Results.

Participant Characteristics

Students aged 10 through 15 from 11 schools in a

district in west central Florida were included in the study.

Of the total 180 participants, 52.2% were female and 47.8%

were male; 68.3% were white and 31.7% were minority.

Minority status included 1 student classified by the

district as Asian-American, 53 students classified as

African-American, and 3 students classified as Hispanic.

Socioeconomic status (SES) was based on eligibility for free

or fee-reduced lunch programs. Of the study participants,

45% did not receive lunch assistance while 55% did receive

such assistance. The 180 participants constituted three

groups: 60 EMH, 60 TMH, and 60 peers without retardation of

similar chronological age. Personal characteristics of the







79
participants, analyzed by developmental level, are presented

in Table 3.


Table 3
Percent of Personal Characteristics by Developmental Level


Mean Mean Race SESD Sex
IQa Age White Minority Yes No Male Female

Non- 12.7 95 5 20.0 80.0 40.0 60.0
retarded
EMH 65.7 12.0 35 65 91.6 8.33 48.3 51.7
TMH 44.1 12.6 75 25 59.3 46.7 55.0 45.0


a IQ scores not available for students without retardation
b SES = socioeconomic status. Yes = eligible for free or fee-
reduced lunch. No = not eligible for free or fee-reduced
lunch.


Participants were randomly assigned to one of three

experimental conditions: Group 1, nonsemantic encoding;

Group 2, semantic encoding; and Group 3, clustered encoding.

The characteristic of the participants by encoding condition

groups are presented in Table 4.


Table 4
Percent of Personal Characteristics by Encoding Condition



Mean Race SESa Sex
Age White Minority Yes No Male Female

Nonsemantic 12.3 70.1 29.3 58.6 41.4 58.6 41.4
Semantic 12.4 70.0 30.0 53.3 46.7 43.3 56.7
Clustered 12.6 64.5 34.5 53.2 46.7 42.0 58.1

SSES = socioeconomic status. Yes eligible for free or fee-reduced
lunch. No = not eligible for free or fee-reduced lunch.









Personal characteristic data further examined by

encoding subgroups within each developmental level are shown

in Table 5.


Table 5
Percent of Personal Characteristics by Encoding Condition
within Developmental Level


Mean Mean Race SES' Sex
IQa Age White Minority Yes No Male Female

Non-
retarded
Gr. 1 12.5 90.0 10.0 25.0 75.0 45.0 55.0
Gr. 2 12.6 95.0 5.0 25.0 75.0 40.0 60.0
Gr. 3 12.8 100 0 10.0 90.0 35.0 65.0
EMH
Gr. 1 66.3 12.1 38.9 61.1 88.9 11.1 66.7 33.3
Gr. 2 64.9 12.0 35.0 65.0 90.0 10.0 40.0 60.0
Gr. 3 65.9 12.0 31.8 68.2 95.4 4.6 41.0 59.1
TMH
Gr. 1 45.0 12.4 80.0 20.0 65.0 35.0 65.0 35.0
Gr. 2 44.4 12.7 80.0 20.0 45.0 55.0 50.0 50.0
Gr. 3 42.9 12.9 65.0 35.0 50.0 50.0 50.0 50.0


a IQ scores not available for students without retardation
b SES = socioeconomic status. Yes = eligible for free or
fee-reduced lunch. No = not eligible for free or fee-
reduced lunch.


In order to determine if significant differences in the

distribution of personal characteristics existed, chi-square

analyses were computed. A significant association was noted

between age and encoding condition (X2 = 12.423 (df=6) p =

.053) and between age and developmental level (X2= 20.189

(df=6) p = .003). For example, 20% (n=12) of the participants

in encoding Group 2 were 13 years of age while 29% (n=17) of









Group 1 participants were of this age. Furthermore, 40%

(n=24) of the students without mental retardation were 13

years old while only 15% (n=9) of the students classified as

EMH were 13 years old. Additional significant association

were noted between race and developmental level (X2 = 51.759

(df=2) p = .000) and SES and developmental level (X2 = 62.357

(df=2) p = .000). However, when encoding condition was

controlled for within each developmental level, significant

association for age, race, and SES no longer existed. Results

of chi-square distributions for age, race, sex, and SES are

shown in Table 6.


Table 6
Chi-Sauare Analysis of Encoding Condition by Personal
Characteristics within Developmental Level



Chi-square df P value


Age
Non-retarded 6.467 6 p = .373
EMH 7.621 6 p = .267
TMH 5.971 6 p = .426

Race
Non-retarded 2.105 2 p = .349
EMH .218 2 p = .897
TMH 1.600 2 p = .449

Sex
Non-retarded .417 2 p = .812
EMH 3.464 2 p = .177
TMH 1.212 2 p = .545

SES
Non-retarded 1.875 2 p = .392
EMH .668 2 p = .716
TMH 1.741 2 p = .419









As all p values were greater than .05, there was no

evidence to support the conclusion that there was a significant

association between personal characteristics and the encoding

condition within each developmental level. Therefore,

randomization to encoding condition successfully distributed

individual characteristics so that within each developmental

level individual characteristics were not associated with

encoding condition.

Recall and Relocation Performance

The number of pictures recalled and number of picture

locations recalled were the dependent measures in the present

study. An optimum score of 50 was available for both

variables. Independent measures included developmental level

(TMH, EMH, and nonretarded); encoding condition (nonsemantic,

semantic, and clustered); and retention interval (immediate and

long-term). Summary statistics of recall and relocation

performance by developmental level and encoding condition are

presented in Table 7.

The following sections contain an analysis of the effect

of developmental level and encoding condition on immediate and

24-hour delay recall performance (hypothesis one) followed by

an analysis of immediate and 24-hour delay relocation

performance (hypothesis three). The interaction of recall and

relocation with the independent variables are then discussed

(hypotheses two and four).









Table 7
Means and Standard Deviations of Recall and Relocation
Performance by Developmental Level and Encoding Condition



Recall 1 Recall 2 Relocation 1 Relocation 2

Non-
retarded
Gr. 1 17.8 19.9 36.5 28.0
(5.9) (7.0) (7.0) (6.7)
Gr. 2 20.6 22.0 39.6 29.9
(4.9) (4.5) (5.0) (6.7)
Gr. 3 28.9 31.0 39.2 30.9
(3.9) (5.0) (5.7) (6.4)
EMH
Gr. 1 11.4 11.1 29.7 23.6
(3.9) (4.0) (10.3) (8.5)
Gr. 2 13.2 13.4 33.5 21.1
(3.2) (4.0) (9.8) (8.8)
Gr. 3 22.6 21.8 28.6 22.0
(6.7) (5.7) (10.5) (7.9)
TMH
Gr. 1 8.6 8.6 20.5 15.1 r/
(3.8) (5.6) (9.1) (4.1)
Gr. 2 12.5 7.4 19.3 16.5
(7.4) (4.6) (9.0) (6.3)
Gr. 3 16.1 18.3 t/ 20.7 16.4 I
(7.7) (7.6) (11.9) (6.6)


Recall Performance

It was hypothesized


that there would be no statistically


significant difference in the mean number of items recalled

following presentation of 50 pictures as a function of (a)

three developmental levels of TMH, EMH, and peers without

retardation of similar chronological ages; (b) three encoding

conditions of nonsemantic, semantic, and clustered; or (c) two

retention intervals of immediate and 24-hour recall. A









significant main effect for either developmental level or

encoding condition for either day one or day two recall

performance would result in rejection of the null hypothesis.

The recall performance on both day one and day two for all

students is shown in Table 8 for both developmental level and

encoding condition.


Table 8
Mean Number of Items Recalled by Developmental Level and
Encoding Condition



Recall Day One Recall Day Two


Developmental Level
Nonretarded 22.4 24.3
EMH 16.1 15.8
TMH 12.4 11.4

Encoding Condition
Nonsemantic 12.6 13.3
Semantic 15.4 14.2
Clustered 22.3 23.7



Immediate recall. Recall performance on day one was

analyzed with ANOVA designs to determine the effect of

developmental level and encoding condition on immediate recall.

A significant main effect at the .05 level ([E (2,177)=31.47,

E<.001]) was found for developmental level. These results are

shown in the Table 9.








Table 9
Source Table for Effect of Developmental Level on Recall
Performance. Day One



Source df F p

Developmental Level 2 31.47 .0001 *

Error 177


*E< .05


A second ANOVA was completed to determine the effect of

encoding condition on recall performance. The significant

main effect which resulted from this analysis is shown in

Table 10 ([F (2,177)=32.87, R<.0001]).


Table 10
Source Table for Effect of Encoding Condition on Recall
Performance. Day One



Source df F p

Encoding Condition 2 32.87 .0001 *

Error 177


*E< .05


Follow-up analyses using Tukey's procedure was performed

to investigate the source of these differences in immediate

recall performance. Significant differences by all three

developmental levels and all encoding conditions were noted.

On day one, the students without mental retardation recalled

more than those classified as EMH who recalled more than

students classified as TMH. The clustered encoding group as a









whole recalled more than those participants in the semantic

group who recalled more than those from the nonsemantic group.

Significant differences were noted in the mean number of items

recalled by the clustered encoding group and the semantic

group and between the clustered and nonsemantic groups but not

between the semantic and nonsemantic groups.

Long term recall. Additional ANOVAs were completed to

investigate the effect of developmental level and encoding

condition on 24-hour delay recall. Again, a significant main

effect for developmental level was noted at the .05 level ([F

(2,177) = 48.72, p<.0001]). Results are shown in Table 11.


Table 11
Source Table for Effect of Developmental Level on Recall
Performance. Day Two



Source df F p

Developmental Level 2 48.72 .0001 *

Error 177


*E< .05


Tukey's follow-up procedures revealed significant

differences in long term recall for all three developmental

levels as again the students without mental retardation

recalled more than EMH who recalled more than TMH.

A significant main effect was noted at the .05 level for

encoding condition ([E (2.177) = 34.10, E<.0001]). Results of

the ANOVA are shown in Table 12.








Table 12
Source Table for Effect of Encodinc Condition on Recall
Performance. Day Two



Source df F p

Encoding Condition 2 34.10 .0001 *

Error 177

*E< .05



As on day one, significant differences were noted in

recall on day two between nonsemantic and clustered encoding

conditions and semantic and clustered but not between semantic

and nonsemantic conditions. In conclusion, both developmental

level and encoding condition made a significant difference in

recall performance on both day one and day two. Therefore, the

first hypothesis was rejected.

Relocation Performance

The third hypothesis was that there would be no

statistically significant differences in the mean number of

items relocated following presentation of 50 pictures as a

function of (a) three developmental levels of TMH, EMH, and

peers without retardation of similar chronological ages; (b)

three encoding condition of nonsemantic, semantic, and

clustered; or (c) two retention intervals of immediate and

24-hour recall. A significant main effect for either

developmental level or encoding condition for either day one or

day two relocation performance would result in rejection of the

null hypothesis. The relocation performance on both day one







88
one and day two for all students is shown in Table 13 for both

developmental level and encoding condition.


Table 13
Mean Number of Items Relocated by Developmental Level and
Encoding Condition

Relocation Relocation
Day One Day Two


Developmental Level
Nonretarded 38.4 29.5
EMH 30.6 22.2
TMH 20.1 16.0

Encoding Condition
Nonsemantic 28.8 22.1
Semantic 30.8 22.5
Clustered 29.5 23.0



Immediate relocation. Investigation of the effect of

developmental level on immediate relocation was performed

using an ANOVA design. This investigation revealed a

significant main effect ([E (2,177) = 63.36, E<.0001]) for

developmental level. Results of this investigation are

presented in Table 14.


Table 14
Source Table for Effect of Developmental Level on Relocation
Performance, Day One



Source df F p

Developmental Level 2 63.36 .0001 *

Error 177

*P< .05








Tukey's statistic computed to determine the source of

these differences provided evidence for significant

differences among all developmental levels. The students

without retardation relocated more pictures correctly than

those participants classified as EMH who relocated more

pictures correctly than TMH.

Further investigation of relocation performance using an

ANOVA design failed to reveal a significant main effect for

encoding condition ([F (2,177) = .43, p<.6480]). Results of

this investigation are shown in Table 15.


Table 15
Source Table for Effect of Encoding Condition on Relocation
Performance. Day One



Source df F p

Encoding Condition 2 .43 .6580

Error 177



Long term relocation. For 24-hour delay relocation

performance, a significant difference in the number of items

relocated was noted between developmental levels ([F (2,177) =

57.40, E<.0001]). The results are summarized in Table 16.

Using the Tukey's procedure, the investigator again found

significant differences in relocation performance for all

developmental levels. Students without mental retardation

relocated more than EMH who relocated more than students

classified as TMH.









Table 16
Source Table for Effect of Developmental Level on Relocation
Performance. Day Two



Source df F p

Developmental Level 2 57.40 .0001 *

Error 177


*P< .05


An additional ANOVA completed to investigate the effect

of encoding condition on relocation performance on day two

revealed no significant differences ([E (2,177) = .17,

E<.8480.]). The results are included in Table 17.


Table 17
Source Table for Effect of Encoding Condition on Relocation
Performance. Day Two



Source df F p

Encoding Condition 2 .17 .8480

Error 177



In summary, a significant difference was not shown in the

number of pictures relocated either on day one or day two due

to encoding condition. However, as developmental level had a

significant effect on the relocation performance of

participants on both day one and day two, hypothesis three was

rejected.




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EROS264SK_Y0M3QF INGEST_TIME 2012-09-24T14:39:10Z PACKAGE AA00011819_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES