Cues and strategies associated with the elimination of adult age differences in the recall of subject performed tasks

MISSING IMAGE

Material Information

Title:
Cues and strategies associated with the elimination of adult age differences in the recall of subject performed tasks
Physical Description:
x, 166 leaves : ill. ; 29 cm.
Language:
English
Creator:
Norris, Margaret P., 1954-
Publication Date:

Subjects

Subjects / Keywords:
Research   ( mesh )
Mental Recall -- Adult   ( mesh )
Mental Recall -- Aged   ( mesh )
Memory, Short-term   ( mesh )
Aging   ( mesh )
Department of Clinical and Health Psychology thesis Ph.D   ( mesh )
Dissertations, Academic -- College of Health Related Professions -- Department of Clinical and Health Psychology -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1990.
Bibliography:
Bibliography: leaves 157-165.
Statement of Responsibility:
by Margaret P. Norris.
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 - 001581155
oclc - 24641710
notis - AHK5062
System ID:
AA00009074:00001


This item is only available as the following downloads:


Full Text










CUES AND STRATEGIES ASSOCIATED WITH THE ELIMINATION OF
ADULT AGE DIFFERENCES IN THE RECALL OF SUBJECT
PERFORMED TASKS














By

MARGARET P. NORRIS


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


1990

































Copyright 1990

by

Margaret P. Norris

































Dedicated, in memory,

to my father











ACKNOWLEDGEMENTS

I would like to first extend my earnest gratitude

to my advisor, Robin West. She has provided endless

support including laboratory space, research

assistants, subject referrals, financial assistance,

and much more that has made the mechanics of a research

project work remarkably smoothly. Above all, I would

like to acknowledge Robin's role in teaching me about

scientific investigation. I entered graduate school

wanting to be trained in research and to the credit of

Robin, I never had misgivings along the way. Robin

taught me design, statistics, and writing skills in a

way which was never tedious. She has made future

research projects possible for me by faithfully

encouraging me to apply for grants, present my findings

at meetings, and by sustaining my enthusiasm about

research.

I also want to recognize Walt Cunningham's role in

my training in Geropsychology. He has sparked my

curiosity and taught me how to follow through with a

question so that I might reach some answers and some

new questions, but never a tangential digression. I











have benefitted from Walt's spirit for complexity and

all of the challenges it brings.

I want to extend my appreciation also to Rus Bauer

who has provided valuable input to my research

projects. I have been able to rely on Rus for insights

and ideas that were always received as constructive,

and never critical. Rus has been patient and impartial

during times when I was stumbling.

My gratitude and acknowledgement also is expressed

to Hugh Davis. Hugh has kept my education in

perspective when I needed to affirm my identification

with Clinical Psychology and by keeping fiction alive

for me when there was no time for reading novels. I am

very grateful for Hugh's uncanny ability to know when

it is time to listen and the wisdom to understand the

bigger picture.

A very important person who has given support

throughout this project is my dear friend, John Bauer.

He provided computer assistance, software, and hardware

that helped me meet my deadlines. He tolerated the

fact that this project did not get completed between

8:00 and 5:00. Above all, John has given me a personal

life that has been a wonderful escape from work. John

offered me rides on the Moto Guzzi just when my energy











needed to be replenished. The long discussions about

science, research, funding, and careers have helped me

understand my goals.

Finally, I want to thank my subjects for their

time and their interest in participating in memory

research.



This research was supported by grant MH09542 from

the National Institute of Mental Health and from a

Dissertation Award from the American Psychological

Association.












TABLE OF CONTENTS

Page

ACKNOWLEDGEMENTS ................................iv

ABSTRACT ........................................ ix

INTRODUCTION..................................... 1

Review of Memory Models in the Aging
Literature .................................. 3
The Significance of SPT Research for Aging.... 18
Motor Cues and Age Differences in SPT Recall..24
Object Cues and Age Differences in SPT
Recall ..................................... 31
The Influence of Strategic Processing on Age
Differences in SPT Recall................... 34
Age differences in SPT recall and Memory for
Cognitive Activities .......................39

EXPERIMENT 1 ......... ...........................48

Introduction................................... 48
Method ........................................ 58
Results ....................................... 62
Discussion .................................... 69

EXPERIMENT 2 .................................... 89

Introduction.. .................................. 89
Method ........................................ 95
Results ....................................... 98
Discussion ................................... 102

EXPERIMENT 3 ................................... 109

Introduction ................................. 109
Method ....................................... 115
Results ...................................... 118
Discussion ................................... 122

GENERAL DISCUSSION ............................. 136








vii












APPENDIX A HEALTH INTERVIEW ................... 143

APPENDIX B LIST OF SPT ITEMS .................. 144

APPENDIX C TEST INSTRUCTIONS .................. 146

APPENDIX D SPT ITEM LIST ...................... 150

APPENDIX E COGNITIVE AND SPT ITEM LIST........ 154

APPENDIX F TEST INSTRUCTIONS .................. 156

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

BIOGRAPHICAL SKETCH ............................ 166


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

CUES AND STRATEGIES ASSOCIATED WITH THE ELIMINATION OF ADULT
AGE DIFFERENCES IN THE RECALL OF SUBJECT PERFORMED TASKS

By

Margaret P. Norris

August, 1990

Chairperson: Robin L. West, Ph.D.
Major Department: Clinical and Health Psychology

The aging effect which has been observed in traditional

verbal recall tasks has recently been examined in activity

recall, a new paradigm for everyday memory research. In

contrast to verbal recall tasks, age differences in Subject

Performed Tasks (SPTs) are not consistently observed. Three

experiments explored variables which may influence the

absence of adult age differences in SPT recall. The

multimodal hypothesis suggests that older adults perform as

well as the young in SPT recall because older adults benefit

from the presence of motor and object cues. Partial support

for this hypothesis was found in Experiments 1 and 2.

Although young adults recalled more than older adults when

actions were enacted during encoding (motor encoding), age

differences in recall were eliminated in the motor retrieval

conditions. The presence of object during encoding did not

influence age differences in recall; however, object

ix









presence interacted with other variables (motor encoding,

list organization, and rate of presentation) to influence

recall. The effects of strategic rehearsal and list

organization were also examined. Older adults performed

better on organized lists when information was enacted at

either encoding or retrieval, resulting in an elimination of

age differences in recall. Rehearsal, on the other hand,

had no significant influence on age-related performance

differences. Finally, Experiment 3 explored factors that

could explain the observed discrepancies between studies of

SPT recall (no age differences often found) and cognitive

activity (age differences consistently found). The findings

of Experiment 3 indicated that the discrepant results could

not be attributed to presentation rate effects or the

differential use of objects. Rather, the age deficit in

cognitive activity recall appears to be related to the

complexity of the to-be-remembered activities and their

labels. It was concluded that the recall probability of

SPTs and activities may depend on specific item

characteristics that have not been investigated thoroughly.

The results were also discussed in terms of older adults'

reliance on task-provided cues and strategies. The

provision of contextual information during both encoding and

retrieval appeared to guide processing so that older adults

were able to bring their level of recall up to that of young

adults.














INTRODUCTION

Until recently, the typical assessment measure in

memory experiments has been based on verbal encoding of

words. Because of this tradition, our knowledge of age

differences in memory abilities is largely limited to

verbal memory tasks such as word list recall and

paired-associate learning (Kausler, 1982). Recently,

gerontologists have felt the need to expand task

domains, particularly to include memory tasks that

might be part of the everyday activities of older

adults. Activity memory is a fruitful new task domain

which provides an opportunity to extend our knowledge

of memory performance in the elderly beyond our present

focus on the traditional verbal memory tasks.

Several measures have been employed in the aging

and activity memory literature (Norris & West, in

press). Event memory refers to simulated everyday

activities. This requires subjects to complete a

related series of actions such as preparing to go on a

vacation or taking a tour of a building. Cognitive

activity memory has also been investigated in the aging

literature. These are sustained mental activities











which subjects solve during acquisition, such as

Weschler Adult Intelligence Scale (WAIS) items and

matching words to pictures. Finally, Subject Performed

Tasks (SPTs) are discrete one-step actions which

subjects enact during acquisition such as, "Cross your

fingers" and "Put the stamp on the book." In contrast

to the age-related memory deficit which is seen with

traditional verbal measures, studies which have

investigated SPTs have frequently found that older

adults recall the performed actions as well as young

adults (Backman & Nilsson, 1984, 1985; Dick & Kean,

1989). This research project investigated variables

which may account for the elimination of age

differences in the SPT recall. Three experiments

addressed issues which remain unresolved in the current

literature on age differences in SPT recall.

Experiment 1 investigated the relative benefit of motor

cues during encoding and retrieval for young and older

adults. In addition, the benefit of list organization

for young and older adults was examined. List

organization was again manipulated in Experiment 2, as

well as the presence of object cues for young and older

adults. Finally, Experiment 3 addressed discrepant

results concerning age differences in SPT recall and











memory for cognitive activities by varying item type,

rate of presentation, and presence of objects. As will

be later elaborated, age differences in SPT recall have

been eliminated (Backman & Nilsson, 1984, 1985; Dick &

Kean, 1989), whereas age differences in recall for

cognitive activities have consistently been found

(Kausler & Hakami, 1983; Kausler, Lichty, & Davis,

1985). The sources of these differences were examined.

The principal models of aging and memory will

first be reviewed in order to identify the theoretical

foundations of the issues investigated. The specific

variables which were manipulated in the experiments

will then be examined by reviewing the SPT literature.

Finally, the three experiments will be individually

reported.


Review of Memory Models in the Aging Literature

Extensive research on memory and aging has

provided evidence that there is an age-related decline

in secondary memory (Craik, 1977; Kausler, 1982; Poon,

1985). That is, age deficits are not typically found

in immediate recall or remote memory, but there is

evidence that older adults perform worse than young

adults on tasks which require them to learn new

information. Three models of the nature and causes of











secondary memory deficits in the elderly have been

identified (Poon, 1985). These include the biological,

processing, and contextual models.

The biological model for age-related memory

deficits emphasizes the neurochemical and anatomical

changes which occur with normal aging. The following

changes in the brain are associated with normal aging:

synaptic and neural loss, increased accumulation of

lipofuscan, neurofibrillary tangles, changes in neural

transmitter substances and synaptic morphology, a

reduction in brain size due to neural loss, atrophy of

brain cells, and reduced cerebral blood flow (Huyck &

Hoyer, 1982). These neuronal and structural changes

are more advanced and widespread in demented older

adults who have pronounced memory deficits (Moscovitch,

1982a).

While these and other physiological changes are

known to occur in the normal aging brain, it is not

understood which, if any of these changes, contribute

to the cognitive changes seen in the normal, healthy

elderly population. Some investigators have emphasized

the importance of neurofribillary tangles and senile

plaques in explaining age-related memory deficits

(Tomlinson, 1972). This explanation has gained











credence because tangles and plaques have been observed

to a greater extent in dementing patients who have

significant memory impairment. Other investigators

have emphasized the association between the particular

cognitive deficits of the aged and the structural

changes in the hippocampus and temporal lobe

(Moscovitch, 1982b). These approaches are not

necessarily contradictory. As pointed out by Squire

(1980), neurofibrillary tangles are most commonly seen

in the hippocampal and temporal areas in the

nondementing aging brain.

A neurochemical approach has focused on changes in

the activity and concentrations of brain

neurotransmitters to account for the memory deficits in

the elderly (Deutsch, 1973; Hines & Fozard, 1980).

Most specifically, a reduction in the hypothalamic

catecholamines has been investigated. A decline in

memory performance following the administration of an

acetylcholine antagonist and a reversal of this

condition with the administration of an acetylcholine

agonist has been observed in young adults (Drachman,

1977). These experimental manipulations are thought to

mimic the neurotransmitter changes that occur with

advanced age. In the aging brain, changes in











neurotransmitter availability result in diffuse

neuronal depletion. This neurochemical change may

account for the cognitive changes seen in the elderly.

A neuropsychological perspective, which associates

functional deficits in the elderly with the structural

changes in the brain, is potentially relevant to the

investigation of age differences in SPT recall.

Preserved learning capacity which has been observed in

amnestic patients has raised questions about the way

memory is organized in the brain and differences among

kinds of memory systems (N. J. Cohen, 1982; Moscovitch,

1982b). Evidence for preserved learning in amnestic

patients (i.e. Korsakoff's, post-encephalitic,

bilateral ECT, and diencephalic patients) comes from

the ability to perform the following: 1) perceptual-

motor tasks such as mirror-tracing and mazes, 2)

perceptual tasks such as mirror-reading, 3) cognitive

tasks such as the Tower of Hanoi, and 4) priming tasks

such as word-completion and lexical decision tasks

(N. J. Cohen, 1982). Amnestic patients are able to

learn how to perform these tasks after repeated trials,

despite their poor recall for having performed the

tasks.











N. J. Cohen (1982) proposes that these

contradictory performances might be explained by two

different memory systems. He suggests that the

observed skill learning in amnestics is "procedural"

knowledge, or a capacity for "knowing how". In this

memory system, information is represented implicitly in

the cognitive operations or procedures. Amnestics are

capable of learning the rules or procedures for

acquiring skills, as required on tasks such as

mirror-reading and the Tower of Hanoi. These skills

are not distinguished by their motor involvement, but

by the nature of the information which is governed by

rules or procedures (N. J. Cohen & Squire, 1980). In

contrast, those learning skills such as verbal and

spatial memory which are impaired in amnestics are

referred to as "declarative" knowledge or "knowing

that". In this memory system, information is

represented explicitly as facts or data, and is

accessed directly on demand.

Most experimental investigations of memory are

based on the declarative knowledge system (N. J. Cohen,

1982). Impaired memory performance in declarative

memory tasks is typically observed in amnestic patients

who have damage to the medial temporal or diencephalic











brain structures. N. J. Cohen (1982) proposes that

procedural and declarative knowledge are mediated by

neuroanatomically distinct memory systems, and this

accounts for the differences in performance in

procedural and declarative memory tasks in amnestics.

It is possible that memory for performed actions is

another example of procedural knowledge. An

age-related memory deficit may be observed with

declarative tasks such as word list learning and

paired-associate learning, whereas older adults may

perform as well as young adults on activity recall

because it is procedural in nature.

In summary, the biological model of age-related

memory deficits has focused on the following neuronal

and structural changes in the aging brain: an increase

in neurofibrillary tangles and senile plaques, changes

in catecholamine transmitter functioning, and the

structural changes in the hippocampal and temporal

areas of the brain. A neuropsychological perspective

suggests that distinct memory systems may account for

the presence of some intact learning skills in

amnestics.

Poon (1985) argues that the biological theories

are applicable to an elderly population with











pathological dysfunction, but this does not account for

the memory performance seen in healthy, community-

dwelling older adults. The absence of an age deficit

in recall in many situations and tasks suggests that

investigations should focus on identifying those

conditions in which memory changes are observed. Poon

(1985) points out that the current challenge in

research is the investigation of the frequently

observed interactions between individuals,

environmental properties, and the task demands. Two

other models of aging and memory, the processing models

and the contextual model, offer alternatives to the

biological explanations of memory deficits in the

elderly.

Two approaches have been taken in the processing

models of aging and memory. First, there has been an

attempt to isolate the stage of processing at which

age-related deficits occur. The stages of processing

include encoding (i.e., acquisition or learning of the

information), storage (i.e., retention of the

information), and retrieval (i.e., access of

information for recall). The second approach in the

aging and memory literature, the information processing

or levels-of-processing model, has attempted to











distinguish the cognitive operations which characterize

the way young and older adults process information

to-be-remembered. As shall be seen, there is much

overlap between these models.

Extensive research has been conducted in order to

determine if deficits in the elderly occur during the

storage, encoding, or retrieval stages of processing.

According to the storage deficit hypothesis,

interference from prior or subsequent material results

in weakened memory traces. Hence, the information

which was initially encoded is lost during the storage

stage. However, there is no evidence for greater

susceptibility to interference in older adults (Craik,

1977). Subsequent literature has focused on encoding

and retrieval deficits as possible sources for

age-related memory deficits.

According to the encoding deficit theory, older

adults do not initially learn or acquire the

information as well as young adults. Two areas of

investigation have provided support for the encoding

deficit theory. First, older adults perform as well as

young adults when initial learning is equated between

the two groups (Schaie & Geiwitz, 1982). That is,

older adults may require more repetitions to learn a











list of words than young adults, but once the list is

learned, young and older adults forget at an equal

rate. This implies that the age-related deficit in

memory is due to less efficient encoding by the

elderly. Second, older adults are thought to be

deficient in their use of strategies or mediational

processes such as rehearsal, organization, visual

association, and verbal elaboration which are presumed

to occur at encoding (Poon, 1985). Evidence for this

comes from the elimination of age differences in recall

when encoding conditions induce subjects to use such

strategies (Schaie & Geiwitz, 1982). Hence, when older

adults are not guided by such mediational instructions,

they recall less due to a deficit at the encoding

stage.

According to the retrieval deficit hypothesis,

older adults are capable of acquiring and storing

information, but they are not as proficient as young

adults in retrieving the information. Evidence for the

retrieval deficit hypothesis comes from the comparison

of young and older adults' performance on free recall,

cued recall, and recognition. In a frequently cited

study, Schonfield and Robertson (1966) found a

systematic age-related decline in free recall but not











in recognition. A decrease in the magnitude of age

differences in recall when support is provided during

retrieval suggests that older adults' decline in memory

performance is due to a retrieval deficit. Similarly,

age differences in cued recall are typically less than

in free recall (Laurence, 1967; Perlmutter, 1979).

This literature suggests that older adults are

disadvantaged relative to young adults when retrieval

support is not provided (Craik, 1977; Perlmutter &

Mitchell, 1982).

Firm conclusions about the encoding versus

retrieval debate have not been made. The literature

has not conclusively separated encoding and retrieval

(Poon, 1985). The evidence which has supported the

encoding deficit hypothesis could also be interpreted

as evidence for a retrieval deficit in the elderly

(Burke & Light, 1981). Support for both encoding and

retrieval deficits in older adults has led researchers

to speculate that elderly adults have a processing

deficit which is common to both the encoding and

retrieval stages (Craik, 1977; Perlmutter & Mitchell,

1982).

In contrast to isolating the stage of processing

at which age-related deficits occur, the information











processing and the levels-of-processing models of aging

and memory have focused on characterizing the

processing deficit observed in the elderly. According

to the information processing model, elderly adults

have deficits in their use of mediational techniques

such as organization, verbal associations, and visual

imagery in determining recall. Young adults have been

found to make greater use of elaborative and

organizational processes than older adults (Hultsch,

1969). However, when subjects are induced to use

various organizational strategies by manipulating the

encoding conditions, the performance of older adults

improves (Hultsch & Deutsch, 1981). Similarly, Hulicka

and Grossman (1967) found that older adults did not

report use of imagery in a paired-associates learning

task, but when instructed to use mediators the older

adults' recall approached that of young adults.

The levels-of-processing model, initially proposed

by Craik and Lockhart (1972), uses an orienting

paradigm in which incidental recall (that is,

performance on an unexpected memory test) is tested

after the subjects' attention has been directed to

particular features of the information

to-be-remembered. According to the levels-of-











processing model, items are encoded along a continuum

from shallow to deep levels depending on what features

were attended to as a result of the orienting task.

Hence, only shallow processing occurs when ones'

attention is directed to surface features such as the

phonemic characteristics of the item list, resulting in

a fragile memory trace. Directing attention to

semantic features of the item list, on the other hand,

requires deep processing and will result in a more

stable memory trace. The levels-of-processing model

hypothesizes that the age-related memory deficit is due

to shallow processing of the to-be-remembered

information in elderly adults. This is supported by

evidence that age differences are eliminated or

minimized when older adults are provided with semantic

orienting tasks at encoding (Craik & Simon, 1980;

Perlmutter & Mitchell, 1982).

A final perspective on human memory is the

contextual model. This model views cognition as a

dynamic process involving an interaction of the

physical, social, and psychological contexts in which

the individual executes the task. The information

which is learned depends not only on the individuals'

current cognitive processes, but also on the context in











which the information was learned, such as the specific

task demands, as well as the motivations and abilities

which the individual brings into the situation (Hultsch

& Deutsch, 1981).

Many contextual variables have been investigated

including individual expectation and motivations, item

familiarity, pre-experimental knowledge, and verbal

abilities (Poon, 1985). Much of this literature has

found age by test conditions interactions suggesting

that age-related memory deficits may be exaggerated by

experimental conditions which favor young adults. For

example, several studies using different test material

including word lists, content of television programs,

and prose found that age differences in recall are

eliminated when older adults with low verbal skills are

excluded (Bowles & Poon, 1982; Cavanaugh, 1983; Taub,

1979). Other investigators have used material that is

differentially familiar to older and young adults

(Botwinick & Storandt, 1980; Hultsch & Dixon, 1983;

Poon & Fozard, 1978). These studies found that older

adults perform as well, or better, than young adults

when the information to-be-remembered, such as

entertainment figures, socio-historic information, and

pictured objects were familiar to the older cohort.











These studies suggest that the types of materials used

to compare the memory performance of young- and older

adults are major determiners of the magnitude of the

observed age-related differences (Poon & Fozard, 1978).

According to the contextual model, recall will

improve when the to-be-remembered information is

presented within a meaningful context. Hence, the

contextual model has emphasized ecologically valid

measures which are meaningful to the individuals'

everyday life rather than the traditional laboratory

tests such as word lists and paired-associate learning

tasks. The role of context in everyday memory demands

has been pointed out by Burke and Light (1981). For

instance, remembering whether or not one has taken

one's medication depends to a great extent on the

specific contextual memories that accompany that

individual occasion and distinguish it from other

instances of the same event. The contextual model of

human memory has, therefore, been considered to be more

relevant to individuals' everyday memory performance

than traditional models.

As previously mentioned, there is overlap between

the models reviewed. Processes such as organization

and mediation which are emphasized in the information











processing model, can also be viewed as methods of deep

processing (Schaie & Geiwitz, 1982). In addition, the

processing deficits seen in elderly adults may be

applied to both the encoding and retrieval stages

(Craik, 1977). That is, recall will improve when an

event is deeply encoded and when similar deep cues are

provided at retrieval. Similarly, the provision of

cues in memory tasks, whether they are verbal

associations as examined by the information processing

model or item familiarity as examined by the contextual

model, can be viewed as guided memory tasks which older

adults perform as well on as young adults. It is

important to acknowledge the overlap in these theories.

Experimental findings can be interpreted in light of

more than one theory.

In summary, three models of age-related memory

deficits have been reviewed in order to provide a

theoretical background. The biological model points to

neurochemical and structural changes in the aging brain

which may account for the memory decline in older

adults. Processing models have focused on isolating

the stage of processing deficits and the level of

processing which characterizes the age-related memory

deficits. Age deficits have been implicated at both











the encoding and retrieval stages of processing, but

not at the storage stage. The information processing

model has focused on the reduced use of mediational

techniques, such as organization, verbal associations,

and visual imagery in determining recall in the

elderly. The levels-of-processing model hypothesizes

that the elderly process information at a more shallow

level than young adults, resulting in a less stable

memory trace. Finally, the contextual model of memory

proposes that the information which is learned depends

not only on cognitive processes, but also on the

context in which the information was learned. Hence,

age may interact with contextual variables such as

expectations and motivations, item familiarity, and

pre-experimental knowledge. The overlap of these

models makes it possible to interpret research findings

according to more than one model.


The Significance of SPT Research for Aging

Recall of Subject Performed Tasks (SPTs) was first

introduced by R. L. Cohen (1981) in order to

investigate the generality of commonly accepted laws of

memory. SPT recall requires the subject to perform a

list of activities, such as "lift the spoon", "rub your

eyes", and "name four colors", which is followed by











free recall. R. L. Cohen (1981) characterizes

traditional verbal recall tasks as a one-way

interaction between the subject and the environment;

that is, the environment exerts an effect on the

subject, but the subject does not exert any observable

effect on the environment. SPTs, on the other hand,

are two-way interactions because this event also

involves an observable manipulation of the environment

by the subject.

R. L. Cohen (1981, 1983) points out that

hypotheses concerning the laws of memory have been

exclusively based on verbal tasks such as free recall

of word lists. In order to test the generality of

these laws, he and his colleagues have investigated the

following memory laws in SPT recall: serial position

effects, strategic processing, and levels-of-processing

effects. In serial position analyses, an absence of

the primacy effect has repeatedly been observed

(Backman & Nilsson, 1984, 1985; R. L. Cohen, 1981,

1983). A primacy effect, that is, greater recall for

the first few items on a list, is thought to be

indicative of rehearsal of those earlier items (Rundus

& Atkinson, 1970). The absence of a primacy effect

suggests that rehearsal is not as important in SPT











recall as it is in word recall. In addition, there is

evidence that strategic processing plays a minor role

in SPT recall. Subjects report that they actively

tried to memorize word lists, but they performed SPTs

without active attempts to memorize them (R. L. Cohen,

1981). This occurred despite the subjects' equivalent

recall of words and SPT lists. Furthermore, unlike the

levels-of-processing effects found in word list recall,

orienting tasks which called for shallow versus deep

processing did not influence the rate of SPT recall

(R. L. Cohen, 1981). It is possible that an

age-related memory deficit is not observed in SPT

recall because older adults are thought to have

processing deficits (Craik, 1977), and therefore, may

perform better on measures such as SPT recall which do

not require deep or strategic processing. These

findings suggest that our knowledge about memory

performance should not be based solely on traditional

verbal recall tasks because other memory tasks, such as

SPTs, appear to be governed by different memory laws

and this may influence older adults' performance.

SPT recall is an important area of research for

understanding age-related changes in memory for several

reasons. First, there is a need for more ecologically











valid approaches to the assessment of memory changes in

the elderly. An ecological approach to memory

assessment has been advocated by gerontologists

(Baddeley & Wilkins, 1984; Botwinick, 1984). An

important ingredient in an adequate ecological

assessment is its relevance to the memory complaints of

older adults. Studies show that memory for activities

are among the most common problems identified by

elderly adults (Cavanaugh, Grady, & Perlmutter, 1983;

Chaffin & Hermann, 1983). Unlike traditional

laboratory tests such as word lists and

paired-associate learning, SPTs resemble the memory for

activities demands which occur in one's daily life.

SPT recall and memory for everyday activities are

similar in that they both provide motor cues, and in

R. L. Cohen's (1981) terms, both involve a two-way

interaction between the individual and the environment.

Another reason why the investigation of SPT recall

is important for understanding adult age differences in

memory is because the lack of age differences found in

SPT recall represents an instance in which older adults

learn new information as well as young adults. This

challenges the widely accepted notion that older adults

have a deficit in secondary memory. Geropsychologists











must ask whether secondary memory deficits in older

adults are limited to a particular class of tasks. As

previously pointed out, Poon (1985) asserts that the

current challenge of aging research is to specify those

conditions in which age-related memory deficits occur

by testing for interactions between age groups and task

conditions. The observed age (young vs. old) by task

(verbal task vs. SPT) interaction has been attributed

to the elderly compensating for cognitive deficits by

using various contextual aids provided by kinesthetic,

visual, auditory, and tactile cues (Backman & Nilsson,

1984, 1985). The source for this interaction needs

further investigation in order to more clearly

understand the reasons why age differences in memory

are eliminated in SPT recall and to more precisely

define those conditions under which older adults

perform as well as young adults.

Finally, SPT recall is an important area of

investigation for aging research because it provides an

experimental paradigm in which the relative importance

of encoding and retrieval operations in age-related

memory deficits can be controlled and investigated. As

previously pointed out, both encoding and retrieval

operations have been implicated in accounting for











age-related memory deficits (Craik, 1977; Smith, 1977).

The study of SPT recall allows for the experimental

manipulation of cues provided during both encoding and

retrieval.

The reasons for investigating the lack of age

differences in SPT recall have been pointed out. SPTs

are a more ecologically valid measure of memory

performance than traditional verbal recall tests.

Also, SPT recall represents an instance in which older

adults learn new information as well as young adults.

The reasons for older adults' improved performance on

SPT recall needs further investigation. Finally, SPT

recall provides a paradigm in which both encoding and

retrieval processes can be examined. The SPT

literature will next be reviewed in order to identify

those variables which have been thought to account for

the elimination of age differences in SPT recall.

These include: motor cues, object cues, strategic

processes, task-related differences, and rate of

presentation.











Motor Cues and Age Differences in SPT Recall

A critical distinction between SPTs and verbal

memory tasks is that SPTs are multimodal, whereas

verbal memory tasks are typically unimodal or bimodal

(Backman & Nilsson, 1984, 1985). SPTs are multimodal

in that all five sensory systems may be involved during

presentation of the material. SPT encoding may include

auditory, visual, motor, tactual, and even olfactory

and gustatory registration. For example, the

information is processed visually because there are

objects present. There is motor processing as the

subjects carry out the action required of the SPT

items, and there is tactual processing by the touching

of various objects. Furthermore, some SPT items such

as "smell the flower" and "eat the candy" involve the

olfactory and gustatory systems. Verbal memory tasks,

on the other hand, are typically presented either

auditorily or visually by reading the items to the

subjects or having the subjects read items from slides

or cards. Backman and Nilsson (1984, 1985) suggest

that older adults are capable of compensating for their

memory deficit when multimodal cues are available.

Their research has focused primarily on the influence











of such cues in eliminating age differences in SPT

recall.

The presence of motor cues is a particularly

important distinction between SPT and verbal recall.

Motor cues are almost always present in SPTs and

presumably never present in verbal recall tasks.

Hence, motor cues may be the most important sensory

system in the multimodal processing of SPTs. Evidence

to confirm the hypothesis that motor cues are important

in eliminating age differences in SPT recall comes from

the following findings: 1) Older adults were not as

adept as young adults in recalling sentence

descriptions of SPTs (Backman & Nilsson, 1984, 1985).

2) Age differences were also found in recall when

subjects imagined performing the task described in the

sentence (Backman & Nilsson, 1985). 3) Age differences

were also found for recall of sentences when the

objects were present suggesting that the object cue

alone was not critical for recall among older adults

(Backman, 1985). These results suggest that the motor

cues provided by the performance of SPTs accounts for

the elimination of age differences in SPT recall.

Others have compared age differences in memory for

motor and cognitive activities to examine the role of











motor cues in age differences in activity memory

(Lichty, Kausler, & Martinez, 1986). Subjects were

tested for recall of motor activities such as cutting

shapes, connecting rings, and clay modeling. They were

also given cognitive activities to recall such as

estimation of length, incomplete words, and word

search. Young adults recalled both types of activities

better than older adults and it was concluded that

motor involvement is an unimportant predictor in the

magnitude of age differences in activity recall.

However, other variables may account for the age

deficit found in activity memory by Lichty et al.

(1986). As the authors suggest, the activities which

they employed may be less familiar to older adults than

young adults. In addition, the information to-be-

remembered was unfamiliar--subjects had to recall task

labels such as "connecting rings" and "word search".

Lastly, subjects in the Lichty et al. (1986) experiment

engaged in the activities for three minutes, whereas

Backman and Nilsson (1984, 1985) had subjects perform

SPTs for five seconds. Hence, while there is some

evidence that motor cues are not the critical variable

in eliminating age differences in SPT recall, the age











deficit found in Lichty et al. (1986) may be explained

by other experimental procedural differences.

In addition to trying to clarify whether motor

cues are important in eliminating an age-related memory

deficit, the relative influence of motor cues during

encoding and retrieval is an important aspect of

investigating the lack of age differences in SPT

recall. Various investigators in SPT and activity

memory have attributed age-deficits on standard memory

lists to either encoding or retrieval deficits in the

elderly. R. L. Cohen (1981, 1983) maintains that

encoding mechanisms are not important in SPT recall.

His evidence for this includes the following: 1) a

primacy effect is not observed in SPT recall suggesting

limited use of rehearsal, 2) subjects report that they

performed SPTs without actively attempting to use

memorization strategies, and 3) there is no effect on

SPT recall when subjects' attention is to diverted to

deep versus shallow features of SPT items. R. L. Cohen

(1981) argues that encoding mechanisms are of little

importance in SPT recall and that a model of SPT memory

requires a shift of emphasis from encoding to retrieval

mechanisms.











In contrast, Backman and Nilsson's (1984, 1985)

experimental design which manipulates motor cues only

during encoding has emphasized the importance of

encoding in SPT recall. They propose that young adults

recall verbal tasks as well as SPTs because they are

able to take the initial event and spontaneously recode

the event in terms of other contextual cues. The

elderly, on the other hand, require the additional

contextual cues during encoding in order to bring their

level of recall up to that of young adults. This

implies that the elderly have an encoding deficit.

Kausler and his colleagues have implicated both

encoding and retrieval deficits in memory for cognitive

activities. There is no difference in recall between

intentional and incidental instructions, providing

further evidence that activity memory is independent of

rehearsal and that encoding mechanisms are relatively

unimportant (Kausler & Hakami, 1983). Hence, the age

deficit they found in activity memory may be

attributable to a retrieval deficit in the elderly

(Kausler & Hakami, 1983). An age deficit in

recognition for cognitive activities, however, also

implies that the elderly have an encoding deficit in

activity memory (Kausler, Lichty, Hakami, & Freund,











1986). Age differences in activity recall are

predicted when young adults are more likely to encode

contextual information which is again elicited at

retrieval (Kausler & Lichty, 1988).

Activity memory studies have manipulated motor

cues during encoding (Backman, 1985; Backman & Nilsson,

1984, 1985; R, L. Cohen, 1981; Dick & Kean, 1989;

Kausler & Hakami, 1983; Kausler et al., 1986). That

is, items were presented with motor cues by having the

subjects perform the tasks as the experimenter read the

items. Recall, however, required a verbal description

of the tasks. This procedure does not allow for an

investigation of the influence of motor cues during

retrieval on activity recall. In more recent studies

using an event paradigm, motor cues have been

manipulated during both encoding and retrieval allowing

for the comparison of the following task conditions: 1)

verbal encoding/verbal retrieval, 2) verbal encoding/

motor retrieval, 3) motor encoding/verbal retrieval,

and 4) motor encoding/motor retrieval (Norris & West,

1988; Saltz, 1988). Norris and West (1988) found that

age differences in event memory were eliminated only in

the motor encoding/verbal retrieval condition, thus,

replicating the findings from earlier studies (Backman











& Nilsson, 1984, 1985; Dick & Kean, 1989). In

contrast, Saltz (1988) found support for the retrieval

deficit hypothesis in a study which investigated

subjects' recall for sentences (e.g., "The teacher

pointed a finger at the blackboard"). Motor enactment

of the sentence verbs was varied at both encoding and

retrieval. Motor processing during retrieval was

differentially beneficial to older adults suggesting

that cues during retrieval are more important in

eliminating age differences in recall.

In summary, motor cues are a critical distinction

between SPT and traditional verbal recall tests, and

hence, may be an important influence in eliminating age

differences in SPT recall. There is evidence that

motor cues during encoding are important in eliminating

age differences in SPT and event recall. However, some

investigations have found evidence that retrieval

mechanisms may be more important than encoding

mechanisms. The relative benefit of motor cues during

encoding and retrieval is addressed in Experiment 1 by

manipulating motor cues during both encoding and

retrieval.













Object Cues and Age Differences in SPT Recall

It has been argued that older adults perform

better on SPT recall than traditional verbal memory

tasks because SPTs involve multiple sensory systems

(Backman, 1985; Backman & Nilsson, 1984, 1985). As

previously reviewed, there is support that motor cues

are important in eliminating age differences in SPT

recall (Backman & Nilsson, 1984, 1985; Dick & Kean,

1989). However, it may be that the presence of motor

cues per se is less important than the multimodal

property of SPTs. In order to address the multimodal

hypothesis there must be evidence that older adults

benefit from cues other than motor cues. The

importance of object cues during encoding in SPT recall

has also been investigated.

Motor cues were varied in a study which compared

young and older adults on recall of SPTs and sentences

with objects present (Backman, 1985). Age differences

were eliminated in SPT recall, whereas young adults

recalled the sentences with objects present better than

older adults. This suggests that the presence of

objects during encoding plays a minor role with respect

to the lack of age differences in SPT recall and that











the critical task properties in eliminating age

differences in SPT recall are the motor cues provided

by the performance of the task items.

No studies have compared young and older adults on

recall of SPT items with objects present and SPT items

without objects present in order to determine if motor

cues alone, or the combination of motor and object cues

accounts for the elimination of age differences in SPT

recall. SPT items with objects and items without

objects have been compared using both young samples and

young and older subjects with the data collapsed over

age groups. In most studies, no differences in recall

of SPT items with and without objects have been found

(Backman & Nilsson, 1984; R. L. Cohen, 1988; R. L.

Cohen, Peterson, & Mantini-Atkinson, 1987, Experiment

1). Others have found that the presence of objects

sometimes facilitates recall (Helstrup, 1986, 1987);

yet other data showed that objects may impair recall

(R. L. Cohen, Peterson & Mantini-Atkinson, 1987,

Experiment 2). These studies did not compare young and

older subjects, and therefore, does not address the

question of whether object cues are important in

eliminating age differences in SPT recall.











Age differences in SPT recall have been found when

items were not present (Guttentag & Hunt, 1988; Knopf &

Neidhardt, 1989). In one study, items were typical of

most SPT studies; however, no objects were presented

and subjects pantomimed the SPT actions (Knopf &

Neidhardt, 1989). For example, no objects were

presented for items such as "put on a cap" and "throw a

lasso" and subjects pretended to perform the actions.

Young adults recalled more items than older adults. In

a second study, subjects were given actions to perform

which did not require the use of objects, such as

"shake your head yes" and "touch your knee" (Guttentag

& Hunt, 1988). Again, an age-related deficit in recall

was found. These studies suggest that motor cues alone

may not be sufficient to eliminate age differences in

SPT recall. As predicted by the multimodal hypothesis,

perhaps the combination of motor and object cues are

important in eliminating age differences in SPT recall.

In summary, object cues alone do not appear to

eliminate age differences in recall. When object cues

are combined with motor cues in SPT recall, there is

usually no increase in recall for young and older

adults collapsed over age groups. However, these

studies do not address the role of object cues in SPT











recall as a function of age. Support for the

multimodal hypothesis comes from studies which found

that age differences in SPT recall remain when items

are enacted but do not provide object cues. There are

no studies which have varied the presence of objects

and compared young and older adults' recall

performance. Experiment 2 compared young and older

adults for recall of SPT items with objects and items

without objects in order to address the role of objects

in eliminating age differences in SPT recall.


The Influence of Strategic Processing on Age
Differences in SPT Recall

It has been argued that age differences in SPT

recall are eliminated because SPT recall is

nonstrategic (R. L. Cohen, 1981, 1983). In contrast,

verbal recall is thought to be strategic, and thus

favors young adults who are able to spontaneously use

organization techniques, visual imagery, verbal

association, etc. Several pieces of evidence have been

cited to support the notion that SPT recall is

nonstrategic: 1) a primacy effect which implicates the

use of rehearsal is not observed in SPT recall (Backman

& Nilsson, 1984, 1985; R. L. Cohen, 1981, 1983), 2)

subjects report no attempt to memorize during SPT











presentation (R. L. Cohen, 1981, 1983), and

3) populations who are presumed to have strategy

deficits such as young children, older adults, and

educably mentally retarded adults are not penalized in

tests of SPT recall (Backman, 1985; Backman & Nilsson,

1984, 1985; R. L. Cohen & Bean, 1983; R. L. Cohen &

Stewart, 1982; Dick & Kean, 1989).

The strategic processes which the SPT literature

has focused on are rehearsal and organization.

Rehearsal memorization strategies are indicated by a

primacy effect in which early items on the list are

recalled at a higher rate (Rundus & Atkinson, 1970;

Atkinson & Shiffrin, 1971). There is agreement that a

primacy effect is not seen in SPT recall (Backman &

Nilsson, 1984, 1985; R. L. Cohen, 1981, 1983). This is

strong evidence that SPTs are encoded without

rehearsal.

The use of organization in SPT recall has also

been investigated in the literature but without such

clear findings. Evidence for the use of organization

has cast doubt on the notion that SPT recall is

nonstrategic in all respects. A comparison of two

lists of SPTs and sentences, one list which could be

organized into five superordinate categories and one











list which could not be organized, showed that the

unorganized lists caused greater impairment in SPT

recall than sentence recall (Backman, Nilsson, &

Chalom, 1986). They conclude that SPT recall is

dependent on organization, and therefore, is not

entirely nonstrategic.

Backman and Nilsson (1984, 1985) suggest that the

multidimensional task properties of SPTs enhance the

detection of superordinate categories for older adults

and this may account for the lack of age differences in

SPT recall. These multidimensional task properties

allow SPTs to be organized by whether or not objects

are present, by the imperatives of the acts (e.g.

"name", "put", etc.), or by semantic categories (e.g.

parts of body, toys, etc.). Using ARC clustering

scores, it was found that SPTs were organized to a

greater extent than comparable verbal tasks, providing

further evidence that SPTs are not entirely

nonstrategic (Backman & Nilsson, 1984, 1985). Although

there were no age differences in ARC scores, the age by

task interaction found for recall was not found for

cluster scores. This suggests that the elimination of

age differences in SPT recall is not due to older

adults' enhanced detection of superordinate categories











in SPTs. However, the detection of superordinate

categories may have been reduced because ARC scores

were based only on two superordinate categories, items

which used objects versus those that did not. The

absence of a relationship between organization and

recall was also found when cluster scores were based on

motor versus cognitive activities (Lichty et al.,

1986). It is possible that the use of only two

categories does not provided enough opportunity for

organization to show age differences.

There is some evidence that age differences in

recall are eliminated in memory for events which have

an organized structure (Padgett & Ratner, 1987).

However, this study did not analyze the data using a

cluster score as a dependent measure. Therefore, it

cannot be determined whether the elimination of age

differences in recall was due to the older adults'

benefiting from clustering.

The effect of organization on event memory was

also explored in a study which varied verbal and motor

processing at both encoding and retrieval (Norris &

West, 1988). The young adults made greater use of

organization in all conditions except the motor

encoding/verbal retrieval condition. This was also the











only condition which did not show age differences in

recall. These results suggest that under certain

conditions young adults make greater use of

organization than older adults, and this may result in

age differences in recall. The finding that young and

older adults were equivalent both in recall of SPTs and

organization of SPTs (Backman & Nilsson, 1984, 1985)

may be specific only to the motor encoding/verbal

retrieval condition. However, Norris and West (1988)

tested for recall of event memory, not SPTs, and their

results may not be replicated using SPT recall as the

dependent measure.

In summary, it is possible that age differences in

SPT recall are eliminated because SPT recall is

nonstrategic. The superior performance of young adults

on verbal recall tasks is thought to be due to their

spontaneous use of strategies such as rehearsal and

organization. The absence of a primacy effect in SPT

recall is evidence that rehearsal is not important in

SPT recall. On the other hand, evidence for the use of

organization in SPT recall has suggested that SPT

recall is not entirely nonstrategic. Although

organization appears to improve SPT recall more than

sentence recall, the nonsignificant age by task











interaction provides evidence that older adults do not

differentially benefit from organizational strategies

in SPT recall. Organization may not account for the

elimination of age differences in SPT recall although

these studies may not have included enough

superordinate categories to elicit organization

strategies. It has also been suggested in the

literature that the absence of age differences in

organization and recall varies across conditions.

Experiments 1 and 2 examine age differences in the use

of organization in SPT recall with items from four

superordinate categories and under conditions which

vary motor cues during both encoding and retrieval.


Age Differences in SPT Recall and Memory for Cognitive
Activities

As previously reviewed, age differences have been

eliminated in SPT recall (Backman, 1985; Backman &

Nilsson, 1984, 1985; Dick & Kean, 1989). In contrast,

experiments in memory for cognitive activities have

consistently found an age-related deficit in recall

(Kausler & Hakami, 1983; Kausler, Lichty, & Davis,

1985; Kausler, Lichty, & Freund, 1985; Kausler et al.,

1986; Lichty et al., 1986). This differential outcome

has occurred in spite of many similarities between the











two types of tasks (Kausler & Lichty, 1988). There are

several possible reasons for the discrepant findings.

The first obvious distinction between the SPT and

cognitive activity studies concerns different

characteristics in the type of item or task employed.

In addition, different procedures used in the studies

such as the rate of presentation of items may account

for the discrepant results.

There are important distinctions between SPT and

cognitive items which may account for the discrepant

results concerning age differences in recall. SPTs are

discrete one-step actions which are typically

motorically performed. In contrast, Kausler and his

colleagues have tested for recall of cognitive

activities such as word-picture naming, serial

learning, and WAIS Information and Arithmetic items.

Unlike SPTs, these activities are continuous actions in

which motor involvement is mostly limited to paper and

pen activity, and as such are primarily cognitive in

nature. A few cognitive activity tasks provide motor

cues, such as object assembly and card sorting, but

these are not typical. It is possible that the

discrepant results from the SPT and cognitive activity

memory studies occur because SPT actions provide older











adults with motor cues that they would not otherwise

spontaneously generate.

Age differences in recall of motor and cognitive

activities were compared to address the role of motor

cues in eliminating age differences in activity recall

(Lichty et al., 1986). Age differences favoring the

young adults were significant for both motor and

cognitive activities suggesting that the discrepant

results in the literature is not due to motor cues. As

previously mentioned, the age deficit found for recall

of motor activities may be accounted for by other

factors including task familiarity, task labels, and

rate of presentation.

Activity memory was also investigated by Norris

and West (1988). In agreement with the SPT literature

(Backman, 1985; Backman & Nilsson, 1984,1985; Dick &

Kean, 1989), age differences were eliminated in the

motor encoding/verbal retrieval condition. On the

other hand, in agreement with Kausler and his

colleagues, Norris and West (1988) found that age

differences persisted in the other two conditions in

which motor cues were provided during retrieval. This

supports the conclusion drawn by Lichty et al. (1986)











that motor cues are not critical in eliminating age

differences in activity memory.

In addition to the different items or tasks used

in the SPT and cognitive activity memory studies,

another important difference which may account for the

contradictory results is the rate at which items are

presented. In the SPT literature, items have been

presented at 5-8 second intervals. In contrast, the

cognitive activity memory studies have had subjects

perform activities for intervals of 45-180 seconds. It

is possible that rate of presentation, rather than item

type differences, accounts for the contradictory

results found in the SPT and cognitive activity

studies.

Rate of presentation was explored in a study in

which subjects engaged in activities for 45, 90, and

180 seconds (Kausler et al., 1986). Young adults

recalled more than older adults at all three rates of

presentation. They concluded that activity duration

has a negligible effect on the magnitude of age

deficits in recall. However, this study did not

address the lack of age differences found in the SPT

studies in which items are presented at 5-8 second











intervals, a significantly briefer period than tested

by Kausler et al. (1986).

The elimination of age differences in recall under

fast presentation conditions is counter-indicated by

Salthouse's (1985) speed hypothesis which states that

older adults' cognitive deficits are due to slower

processing. The slowing hypothesis is supported by

evidence that older adults perform better when testing

procedures allow them to perform at their own pace.

This data would suggest that older adults should be

penalized by the fast presentation rates of SPTs and

that age differences might be reduced in cognitive

activity studies which present tasks for longer

periods; in fact, the reverse is found.

It is possible that the very brief presentation

rate of SPTs prohibits elaborate processing which would

otherwise benefit young adults. Support for this comes

from the following: Young adults' recall was reduced

to a level comparable to that of elderly adults when

encoding time was restricted (Simon, 1979). In

addition, Craik and Rabinowitz (1985) found that the

magnitude of the age-related memory deficit was

increased with longer presentation rates. They argue

that longer encoding times are beneficial to young











adults because this gives them the additional time

needed to carry out effective processing operations

which older adults fail to do when not guided by

environmental support. Hence, the contradictory

results in the SPT and cognitive activity literature

may be explained by the differences in the rate of

presentation.

In summary, age differences have been eliminated

in SPT recall, but remain in cognitive activity memory.

Different factors may account for the discrepant

results. First, there are important item distinctions

such as the prevalence of motor cues in SPTs.

Alternatively, the discrepant results may be due to

important procedural differences such as the rate of

presentation in the SPT and cognitive activity studies.

Item type and rate of presentation were varied in

Experiment 3 in order to identify what variables may

account for the discrepant results in the SPT and

cognitive activity memory literature.



The models of memory and aging reviewed earlier

provide principles which may explain why age

differences in recall are eliminated when motor,

object, and organization cues are provided. A











neuropsychological explanation for the elimination of

age differences in SPT recall specifically addresses

the benefit of motor cues. As previously reviewed,

preserved learning in amnestics may be accounted for by

two distinct neuroanatomically memory systems which

govern declarative and procedural knowledge (N. J.

Cohen, 1982). The elimination of age differences in

the motor encoding/verbal retrieval condition is

arguably an example of declarative knowledge. Hence,

the absence of an age-related memory deficit in SPT

recall cannot be accounted for by defining it as a

procedural memory task. Nevertheless, there may be a

distinction between SPT and verbal recall tasks based

on neuroanatomical differences. Older adults' improved

memory performance when actions are performed may be

due to the relative sparing of the cortical motor areas

in normal aging (Winbald, Hardy, Backman, & Nilsson,

1985). A neurochemical basis for the preservation of

memory when motor cues are provided has also been

implicated. Research has demonstrated that

neurotransmitter changes associated with aging are not

anatomically uniform but vary in different brain

regions (Allen et al., 1983; Strong, 1985). Winbald et

al. (1985) have also proposed that older adults may











recall performed actions as well as young adults

because concentrations of dopamine, a critical

neurotransmitter for motor functions, are not reduced

in the hippocampus with age. It is possible that the

elimination of age differences in SPT recall is due to

the sparing of structural areas and neurochemical

processes associated with motor activity.

Alternatively, the contextual model suggests that

the motor, object, or organization cues, or their

combination, benefits older adults by providing them

with information that is not otherwise spontaneously

processed, as occurs with young adults. According to

the contextual model of memory and aging, an age-

related memory deficit is found when the specific task

demands and the material to-be-remembered is not

appropriately familiar and meaningful to older adults.

According to Backman & Nilsson (1984, 1985), older

adults are less likely to spontaneously encode less

contextually rich material such as verbal tasks. In

contrast, young adults are able to take only the

semantic cues provided by verbal recall tasks and

"spontaneously recode" the task in terms of all the

potential contextual cues (Backman & Nilsson, 1984,

1985). This is supported by the finding that young











adults' recall of SPTs is not greater than their recall

of comparable verbal tasks (Backman, 1985; Backman &

Nilsson, 1984, 1985). The motor performance of SPTs,

the object cues, and the enhanced detection of

superordinate categories may provide additional

contextual information which enhances recall for older

adults, whereas the additional information provided by

these cues would be redundant to young adults.

In conclusion, this research project investigated

some of the variables which may contribute to the lack

of age differences in SPT recall. Motor cues, an

important distinction between SPT and traditional

verbal recall tasks, was varied during both the

encoding and retrieval stages. In order to investigate

the multimodal hypothesis the benefit of object cues to

older in activity recall was also analyzed. In

addition, the use of organization in SPT recall was

examined in order to investigate the hypothesis that

the lack of age differences in SPT recall is due to

nonstrategic processing. Finally, contradictory

results in the SPT and cognitive activity memory

literature were examined by varying item type and the

rate of presentation.














EXPERIMENT 1

Introduction

Extensive research on memory and aging has shown

that there is an age-related decline in secondary

memory (Craik, 1977; Kausler, 1982; Poon, 1985).

Recent research has suggested that this observed age

deficit in recall for newly acquired information may

need to be qualified. In contrast to verbal recall

tasks, age differences have often been eliminated in

the recall of Subject Performed Tasks (SPTs) (Backman &

Nilsson, 1984, 1985; G. Cohen & Faulkner, 1989; Dick &

Kean, 1989). A critical distinction between SPTs and

traditional verbal recall tasks is that verbal memory

tasks are typically limited to auditory or visual

registration of the material, whereas SPTs provide

multimodal cues. SPT encoding may provide auditory,

visual, motor, tactual, and even olfactory and

gustatory registration of the material. The purpose of

this study is to examine the role of cues and strategic

processing in eliminating age differences in SPT

recall.











Age differences in verbal recall tasks have been

attributed to young adults' ability to spontaneously

generate cues and strategies for remembering

information. In contrast, older adults are less able

to generate cues on their own and may need to rely on

cues provided by either the examiner or the test

conditions (Hultsch & Deutsch, 1981). Age differences

in SPT recall may be eliminated because older adults

benefit from the additional contextual cues which are

provided by the test conditions. In contrast to verbal

recall tasks, SPTs provide multimodal cues such as

motor, visual, auditory, and tactile cues. Hence, age

differences in recall are eliminated because older

adults do not have to rely on generating cues for

themselves (Backman & Nilsson, 1984, 1985).

The cues which have been emphasized in the SPT

literature are motor cues provided by the enactment of

the task and visual cues provided by the presence of

objects. Motor cues are a particularly important

distinction between SPTs and verbal recall tasks

because motor cues are almost always present in SPTs

and presumably never present in verbal recall tasks.

Evidence for the importance of motor cues in

eliminating age differences in SPT recall comes from











studies which have controlled for motor cues under

various SPT conditions. Age differences were observed

in the recall of sentence descriptions of SPTs,

sentences with the objects present, sentences with

instructions to imagine performing the action, and

examiner performed actions. In contrast, age

differences were eliminated in the recall of SPTs when

subjects performed the action during encoding (Backman,

1985; Backman & Nilsson, 1984, 1985; G. Cohen &

Faulkner, 1989; Dick & Kean, 1989). These results

suggest that the motor cues provided in SPT recall are

a critical variable in eliminating age differences in

recall.

SPT studies have emphasized the importance of

motor cues provided during encoding but motor cues

could also enhance retrieval processes. The relative

influence of motor cues during encoding and retrieval

is an important aspect of investigating the elimination

of age differences in SPT recall. Various

investigators have attributed the age-related decline

in verbal recall to either encoding or retrieval

deficits in the elderly. Within the memory for

activity literature, R. L. Cohen (1981, 1983) has

emphasized the importance of retrieval mechanisms in











activity recall, Backman and Nilsson (1984, 1985) have

focused on encoding processes as critical in activity

recall, and Kausler and his colleagues have implicated

both encoding and retrieval operations in influencing

activity recall (Kausler & Hakami, 1983; Kausler et

al., 1986). SPTs are a useful way of examining both

encoding and retrieval processes because motor cues can

be varied at both stages.

It is possible that age differences are eliminated

in SPT recall for reasons other than the presence of

motor cues. It has also been argued that age

differences in SPT recall are eliminated because SPT

recall is nonstrategic (R. L. Cohen, 1981, 1983).

Verbal recall is enhanced by strategies such as

rehearsal, organization, verbal association, and visual

imagery. Age differences in the recall of verbal tasks

have been attributed to young adults' spontaneous use

of these strategies. In contrast, older adults are

thought to be less efficient at spontaneously using

such strategies (Hultsch & Deutsch, 1981).

Two strategies, rehearsal and organization, have

been addressed in the activity literature. There is

evidence that subjects do not engage in rehearsal

during SPT learning because early items from a SPT list











are not recalled at a higher rate than middle items, as

is typically seen in verbal recall. When early items

are recalled at a higher rate, called a primacy effect,

it is interpreted as evidence for rehearsal (Atkinson &

Shiffrin, 1971). This absence of a primacy effect

provides evidence for the nonstrategic nature of SPT

encoding (Backman & Nilsson, 1984, 1985; R. L. Cohen,

1981, 1983).

In addition to rehearsal, organization in SPT

recall has been examined in order to investigate the

role of strategic processing in eliminating age

differences in SPT recall. It has been suggested that

the multidimensional task properties of SPTs enhance

the detection of categories by older adults and that

this may account for the lack of age differences in SPT

recall (Backman & Nilsson, 1984, 1985). Older adults

may improve their performance in SPT recall because

they organize the list into semantic categories (e.g.

parts of body, toys), items with and without objects,

or types of actions (e.g. "name", "put").

The literature has produced contradictory results

concerning organization. Some studies suggest that the

elimination of age differences can not be attributed to

enhanced organizational processing by older adults.











The age (young vs older adults) by condition (verbal vs

SPT tasks) interaction found for SPT recall was not

found using clustering scores (ARC) based on the

grouping of items with and without objects (Backman &

Nilsson, 1984, 1985). This indicates that the degree

of age differences in clustering does not vary across

conditions and suggests that older adults' improved

performance in SPT recall is not due to clustering. In

addition, age differences were not eliminated for the

recall of a highly organized tour-taking event (Padgett

& Ratner, 1987). In contrast, other studies suggest

that older adults may benefit from highly organized

material and bring their level of recall up to that of

the young adults. Age differences were eliminated in

the recall of an organized clay-making event (Padgett &

Ratner, 1987). In addition, age differences in event

memory were eliminated only in the condition (motor

encoding/verbal retrieval) in which age differences in

organization were also eliminated (Norris and West,

1988).

In the SPT literature, the preponderance of

evidence suggests that both young and older adults's

recall improves with organized material (Backman &

Nilsson, 1984, 1985). However, the age by task











interaction found for recall has not been found when

cluster scores are used as the dependent measure

suggesting. The pattern of age differences in

clustering appears to be the same for SPTs and verbal

recall. Older adults do not appear to differentially

benefit from organizational strategies in activity

recall as compared to verbal recall (Backman & Nilsson,

1984, 1985; Lichty et al., 1986). Hence, organization

may not account for the elimination of age differences

in SPT recall.

There was a limitation, however, in the earlier

research. These studies restricted organization to

only two categories. This may not have provided

sufficient organizational cues for older adults to

improve their recall. When multiple categories for

organization were provided, age differences in recall

were eliminated in the condition which also eliminated

age differences in organization (Norris & West, 1988).

In summary, the activity literature suggests that

age differences in recall of SPTs may be eliminated

because older adults may benefit from motor cues and/or

organizational cues. This study varied motor cues

during the encoding and retrieval stages in SPT recall,

and varied list organization. The following four











conditions were compared: verbal encoding/verbal

retrieval, verbal encoding/motor retrieval, motor

encoding/verbal retrieval, and motor encoding/motor

retrieval. In addition, two lists were used to test

for recall, one list in which the items could be

organized into four categories, and a second list in

which items could not be readily organized.

Four hypotheses concerning motor cues were

examined. First, it was hypothesized that the young

adults would recall more than the older adults in the

verbal encoding/verbal retrieval condition (Backman,

1985; Backman & Nilsson, 1984, 1985; G. Cohen &

Faulkner, 1989; Dick & Kean, 1989; Norris & West,

1988). This outcome was expected because this

condition resembles traditional verbal recall tasks,

such as word list recall, in which age differences are

commonly observed (Kausler, 1982). The second

hypothesis predicted that age differences would be

observed in the verbal encoding/motor retrieval

condition (Norris & West, 1988). This condition

required subjects to translate verbally processed

information into motor actions for recall. Young

adults were expected to be advantaged in this condition

because they are familiar with processing verbal











information and spontaneously recoding this information

into potential contextual cues which are not directly

provided by the task. Older adults, on the other hand,

were expected to be disadvantaged in this condition

because they are less efficient at translating and

transforming the verbal information. In contrast, the

translation from motor encoding to verbal retrieval has

not produced age differences in recall (Backman, 1985;

Backman & Nilsson, 1984, 1985) perhaps because the

young and old adults are both unfamiliar with

processing lists of unrelated actions at encoding. The

third hypothesis predicted that age differences in

recall would be eliminated in the motor encoding/verbal

retrieval condition (Backman, 1985; Backman & Nilsson,

1984, 1985; G. Cohen & Faulkner, 1989; Dick & Kean,

1989; Norris & West, 1988). Older adults were expected

to perform at the same level as young adults in this

condition because the motor encoding provides the

subjects with multimodal cues. By processing these

available multimodal cues, the older adults were

expected to use the same cues for remembering as young

adults. Finally, age differences were expected to be

eliminated in the motor encoding/motor retrieval

condition (Saltz, 1988). Older adults were expected to











perform as well as young adults in this condition

because contextual cues were provided during both the

encoding and retrieval stages. It has been

demonstrated that age differences in verbal recall are

eliminated when cues are provided during both encoding

and retrieval (Smith, 1977). A similar outcome was

expected for the SPTs.

This study also compared recall of SPT items which

could be organized into four semantic categories with

SPT items which could not be organized in order to

examine the influence of strategic processing on

eliminating age differences in SPT recall. There is no

previous SPT study which has examined age differences

in organization when motor cues are varied during both

encoding and retrieval. This study explored age

differences in the organization of SPTs under those

conditions which vary verbal and motor processing

during both the encoding and retrieval stages.

Previous SPT literature has compared clustering in the

verbal encoding/verbal retrieval condition to

clustering in the motor encoding/verbal retrieval

condition. These results were mixed as indicated

earlier. The additional conditions in this study

(verbal encoding/motor retrieval and motor











encoding/motor retrieval) have not been used

previously. It was hypothesized that age differences

in recall would be eliminated only in those conditions

which also eliminated age differences in organization,

suggesting that the lack of age differences in SPT

recall may be related to strategic processing (Norris &

West, 1988).

Method

Subjects

The subjects included 80 older adults (60-79

years, M = 68.7, SD = 4.9) and 80 young adults (18-26

years, M = 19.7, SD = 1.4). The older adults were

recruited from the community by newspaper advertisement

and solicitation at community organizations and senior

citizen organizations and received $10.00 for their

participation. The young adults were recruited from

university classes and received course credit for their

participation. The older and young adults did not

differ in terms of education, p > .05 (older adult

range = 8-21 years, M = 14.3, SD = 3.4; young adult

range = 12-17, M = 13.5, SD = 1.0), or WAIS-R

Information subtest scores, p > .05 (older adult range

= 8-18, M = 12.1, SD = 2.3; young adult range = 6-17,

M = 11.5, SD = 2.4). The older adults were screened











for current health status to eliminate persons with

cardiovascular disease or other chronic health problems

that may affect memory performance (see Appendix A).

Materials

The SPT items were primarily obtained from R. L.

Cohen's (1981) SPT item list (see Appendix B). In

order to fulfill the requirements of variables tested

in this experiment, additional SPT items were added to

R. L. Cohen's original list.

Two 16-item lists were used to test recall. The

items on one list could be organized into four semantic

categories including movements with 1) hands, 2) legs,

3) face, and 4) torso. The items on the second list

were also body movements, but the items did not

represent distinct semantic categories. Pilot testing

confirmed that seven out of ten subjects were able to

identify all four categories in the organized list,

whereas no subjects were able to identify consistent

categories in the unorganized list.

Procedure

The design was a 4 x 2 x 2 mixed factorial in

which there were two between subjects factors, test

condition and age group, and one within subjects












factor, list organization. The between-subjects factor

included the following four test conditions: 1) verbal

encoding/verbal retrieval, 2) verbal encoding/motor

retrieval, 3) motor encoding/verbal retrieval, and 4)

motor encoding/motor retrieval. Twenty subjects from

each age group were randomly assigned to each

condition. The following diagram illustrates the

between subjects design used:


Verbal Encoding

Examiner reads SPT list

Verbal recall Motor recall

Verbal encoding/ Verbal encoding/
Verbal retrieval Motor retrieval



Motor Encoding

Examiner reads SPT list
as the subjects perform the SPTs

Verbal recall Motor recall

Verbal encoding/ Verbal encoding/
Verbal retrieval Motor retrieval



The within-subjects factor, list organization, was

examined by testing all subjects on recall of two

lists, one which could be organized into four semantic











categories and one which could not be organized into

semantic categories.

Subjects were first given general instructions

which were specific to the encoding and retrieval

condition under which they were tested (see Appendix

C). A practice task was then demonstrated to the

subjects. The practice task was followed by the two

SPT lists. The two SPT lists were part of a larger

battery including two other SPT lists, the WAIS-R

Information subtest, and a demographic questionnaire.

SPT list order was counterbalanced such that each list

was placed in a different position for each of four

list orders. In order to reduce interference effects

between lists, the Information subtest and the

demographic questionnaire were administered following

the second list.

The SPT items were read to all of the subjects at

a rate of one item every six seconds. Subjects in the

verbal encoding conditions were instructed only to

listen to the items. Subjects in the motor encoding

conditions were instructed to perform the actions

specified by each SPT item as the items were read to

them by the experimenter.

Immediate free recall was obtained after each list

was presented. Verbal retrieval required written











verbal recall of the items. Motor retrieval required a

re-enactment of the SPT items.

Subjects in the motor retrieval conditions were

videotaped in order to have a record of their

performance to test for inter-rater reliability. The

record sheet from the subjects in the verbal retrieval

conditions was also available to test for inter-rater

reliability.

Results

Twenty-one protocols were independently scored by

two examiners. The scores obtained from the two

examiners were within one point of each other on 95% of

the protocols, and produced a correlation of .96. This

indicates that the scoring method was highly reliable.

The data analysis was conducted on the scores obtained

by the primary investigator.

For both age groups there were no significant

differences in the recall of items obtained from R. L.

Cohen's (1981) list and the new SPT items which were

added to fulfill the experimental requirements, R >.05

(p-value corrected for two t-tests).

The repeated measures analysis was a 2 (age) x 4

(test condition) x 2 (list organization) x 4 (list

order) mixed factorial in which age, test condition,













and list order were between-subject factors and list

organization was a within-subjects factor. All apriori

comparisons were conducted with Tukey's test (p < .05).

There were significant main effects for age group,

F(1, 128) = 53.83, p < .0001, w2 = .21; condition,

F(3, 128) = 14.10, p < .0001, w2 = .16; and

organization, F(l, 152) = 24.10, p < .0001, w2 = .11

(see Table 1-1). The age by condition interaction was

not significant. Young adults recalled more than older

adults. The verbal encoding/verbal retrieval condition

produced significantly less recall than all other

conditions. Finally, more items were recalled on the

organized list than the unorganized list.

Main effects for list order were not significant;

however, the order by condition by list interaction was

significant, F(9, 128) = 2.07, p < .05, w2 = .01.

Posthoc tests using Scheffe's (R < .05) showed that

condition differences varied as a function of list and

order. In every case, recall was significantly lower

in the verbal/verbal condition than the other three

conditions. In addition, list differences varied as a













function of condition and order. The organized list

was recalled at a higher rate than the unorganized list

only under some order and condition combinations. No

clear pattern emerged.

The group by list organization interaction was

also significant, F(1,152) = 8.23, p < .005, w2 = .04.

For the older adults, more items were recalled on the

organized list than on the unorganized list; however,

there was no organization effect for young adults (see

Figure 1-1).

Apriori analyses of the hypotheses concerning

conditions produced the following results: young adults

recalled significantly more than the older adults in

the verbal/verbal condition, F(1,152) = 13.97, p < .005

and the motor/verbal condition F(1,152) = 9.41,

P < .005. Age differences in recall were not

significant in the verbal/motor and the motor/motor

conditions (see Figure 1-2). This confirmed the

hypotheses that age differences in recall would remain

in the verbal/verbal condition and age differences

would be eliminated in the motor/motor condition. The













hypotheses concerning the verbal/motor and motor/verbal

conditions were not confirmed.

In order to investigate the role of encoding and

retrieval cues on the elimination of age differences,

the data were collapsed in order to compare verbal and

motor encoding, as well as verbal and motor retrieval

as a function of age. Encoding effects were examined

by combining the verbal/verbal and verbal/motor

conditions into one verbal encoding condition, and this

was compared to the motor encoding condition which was

the combined motor/verbal and motor/motor conditions.

Likewise, verbal retrieval (the combined verbal/verbal

and the motor/verbal conditions) were compared to motor

retrieval (the combined verbal/motor and the

motor/motor conditions) to analyze retrieval effects.

The group by retrieval interaction was significant,

F(1,152) = 4.29, p < .05, w2 = .01. For older adults,

posthoc comparisons showed that more items were

recalled with motor retrieval than with verbal

retrieval; however, there was no retrieval effect for











young adults. The group by encoding interaction was

not significant.

The data from the organized list were also

analyzed using ARC clustering scores as the dependent

measure to observe organization effects. The ARC

measure was chosen because it is does not vary with the

number of items or categories recalled (Roenker,

Thompson, & Brown, 1971). Main effects for group and

condition, as well as their interaction, were not

significant. In all four conditions, age differences

in the ARC clustering score were not significant. In

contrast, age differences in recall were eliminated in

only the verbal/motor and the motor/motor conditions,

R <.005. Due to very large standard deviations in ARC

scores (M = .23, SD = .34), the data was analyzed again

using an arcsine transformation of the ARC scores.

This produced the same results as above. The

hypothesis that age differences in recall would be

eliminated in those conditions which also eliminated

age differences in organization was not supported;

however, interpretation of the data is made with

caution due to the large variability in ARC scores.

Large standard deviations for ARC scores were also

reported by Lichty et al. (1986).











Because of the limited usefulness of the ARC

scores, post hoc analyses with Scheffe's test (p < .05)

were conducted to compare age differences in recall as

a function of list organization and test conditions.

Age differences in recall of the organized list were

eliminated in the three conditions with motor cues:

verbal/motor, motor/verbal, and motor/motor. There

were significant age differences in recall only in the

verbal/verbal condition. In contrast, for the

unorganized list, there were significant age

differences in the verbal/verbal, verbal/motor, and

motor/verbal conditions. There were no age differences

in recall of the unorganized list only in the

motor/motor condition. This comparison suggests that

the combination of organization and motor cues was

important in eliminating age differences in SPT recall.

Age differences in the recall of an organized list

were eliminated when motor cues were available during

either encoding or retrieval.

A serial position analysis was also performed in

order to examine the influence of rehearsal on SPT

recall (see Figures 1-3 to 1-6). Consistent with other

SPT studies, there was a distinct primacy effect in the

verbal/verbal condition in contrast to the absence of a














primacy effect in the motor/verbal condition (Backman &

Nilsson, 1984; 1985; R. L. Cohen, 1981). Serial

position effects were also analyzed in conditions which

have not yet been investigated in the literature:

verbal/motor and motor/motor. In addition to the

motor/verbal condition, the primacy effect was also

reduced in the motor/motor condition.

The data also produced unexpected results in terms

of recency effects. There was an absence of recency

effects in both the verbal/motor and the motor/motor

conditions. Together, these results suggest that motor

encoding reduces primacy effects and motor retrieval

reduces recency effects relative to those observed in

the corresponding verbal conditions.

The serial position analyses were very consistent

across age groups. Both older and young adults showed

the same pattern of an absence of primacy effects with

motor encoding and an absence of recency effects with

motor retrieval. This suggests that individual item

characteristics and the specific encoding and retrieval











conditions influenced the serial position curves for

both ages.


Discussion

Geropsychologists have attributed the age-related

decline in memory to encoding and retrieval deficits in

the elderly (Kausler, 1982), and their differential

influence has been debated (Poon, Walsh-Sweeney, &

Fozard, 1980). In this study, the manipulation of

motor cues at both stages of processing resulted in an

elimination of age differences when motor cues were

provided during retrieval, while an age deficit in

recall remained when motor cues were provided only

during encoding. These results lend support to the

notion that older adults do not spontaneously generate

retrieval cues, and indicated that older adults can

bring their recall performance up to the level of young

adults when the task provides motor cues during

retrieval. Further support for the retrieval deficit

hypothesis comes from another study in which an age

deficit occurred in recall but not for recognition of

SPTs (Knopf & Neidhardt, 1989).

It has been argued that the reinstatement of

internal and external contextual information during

recall can compensate for older adults' retrieval











deficits, thus eliminating age differences in recall

(Burke & Light, 1981). Hence, age differences in

recall may be eliminated in the verbal/motor and

motor/motor conditions because the motor enactment at

retrieval elicits contextual information which was

processed during encoding either in terms of internal

images of enactment or via the enactment itself.

Kausler and Lichty (1988) have also argued that the age

deficit which is sometimes found in activity memory is

a result of older adults' deficit in retrieving

contextual information at recall. When contextual cues

are provided at retrieval, as occurs with motor

retrieval, older adults improve their level of recall.

Unlike young adults, older adults must rely on

contextual cuing at retrieval. In contrast to the

motor retrieval conditions, age differences in the

motor/verbal condition are seen because the nonverbal

contextual information which was processed during

encoding is of no benefit for the older adults when

they are faced with a verbal recall task.

Many previous studies have found that age

differences in SPT recall are eliminated in the

motor/verbal condition (Backman, 1985; Backman &

Nilsson, 1984, 1985; G. Cohen & Faulkner, 1989; Dick &











Kean, 1989). This finding was not replicated in this

study. Other experimenters have also found significant

age differences in SPT recall with the motor/verbal

condition (R. L. Cohen, Sandler, & Schroeder, 1987;

Craik, 1989; Knopf & Neidhardt, 1989; Nilsson &

Backman, 1989; Guttentag & Hunt, 1988; Lichty et al.,

1986). Several explanations have been offered to

account for these discrepant findings including item

characteristics, the use of objects, and list

characteristics (Norris & West, in press).

The magnitude of age differences can vary greatly

from item to item (Kausler & Lichty, 1988). As noted

in a recent review, age differences in recall were

significant in approximately 15% of the items (Kausler

& Lichty, 1988). The age differences were not

significant for the remaining items, and on several

items, the older adults recalled more than the young

adults. The present study also suggests that there may

be important item characteristics which predict the

magnitude of age differences in recall. Age

differences in recall were significant for 34% of the

items, and the older adults occasionally recalled items

at a higher rate than young adults (see Table 1-2).

This data clearly points out that age differences in












recall vary substantially from item to item. Further

exploratory work is needed to identify item

characteristics which predict age differences in

recall. Possible item characteristics which may

influence the magnitude of age differences include item

familiarity, vividness, distinctiveness, complexity of

movement, and others.

Earlier studies have investigated some of these

variables that may reflect item characteristics

predictive of SPT recall in young adults. Familiarity

and vividness did not predict recall (R. L. Cohen,

Peterson, & Mantini-Atkinson, 1987; Knopf & Neidhardt

1989). On the other hand, there is also data which

suggests that the goal of the action rather than the

kinesthetic cue is more important in predicting recall.

That is, the movement sequence (e.g. a circular hand

motion) is less predictive of recall than the action

goal (e.g. drawing a circle) (R. L. Cohen & Heath,

1988). This implies that item recallability is less

influenced by sensory information such as kinesthetic

cues than by variables related to the meaningfulness of











the action. These variables might concern how

frequently items are typically performed, how easy they

are to imagine, how simple versus complex they are, and

other predictive variables of item recallability.

These earlier studies have laid a foundation for

identifying item characteristics that influence SPT

recall; however, they do not address the variability of

the magnitude of age differences across items.

Research is still needed in order to identify item

characteristics that affect the pattern of age

differences. In the verbal memory literature,

experimental control has been maintained over variables

such as word frequency and concreteness. During word

list acquisition, subjects are accurate in predicting

which items will be subsequently recalled; whereas

subjects (and presumably experimenters) are highly

inaccurate in predicting recall of SPTs (R. L. Cohen,

1988). Further research in this area may shed light on

those item characteristics which predict SPT recall,

and more specifically, those item characteristics and

stimulus properties which enhance the memory

performance of older adults, and hence, eliminate age

differences in recall.











Perhaps item characteristics which predict age

differences in recall are important in terms of their

value for cuing during retrieval. Kausler and Hakami

(1983) argue that the retrievability of memory traces

varies directly with the distinctiveness of those

traces. Hence, age differences in activity recall will

be greater for those items which are not highly

distinctive because older adults will have greater

difficulty in retrieving those items. Presumably,

motor retrieval cues, as well as other yet unknown item

characteristics, provide distinctive contextual

information which aids older adults at retrieval.

It has also been suggested that age differences in

SPT recall may be predicted by list characteristics

such as organization. The hypothesis that age

differences in recall would be eliminated only in those

conditions which also eliminated age differences in

organization could not be sufficiently addressed

because of large variability in ARC scores. However,

support for the organization hypothesis came from the

observed list by group interaction. Older adults

obtained a higher level of recall on the organized list

than the unorganized list; whereas there was no

organization effect for young adults. It appears that











age differences in SPT recall were related to list

organization. Age differences were eliminated in the

recall of an organized SPT list when motor cues were

provided either during encoding or retrieval.

Organization cues in SPT recall appear to

differentially benefit older adults. In contrast, age

differences were found in the recall of an unorganized

SPT list except when motor cues were provided at both

encoding and retrieval. This suggests that older

adults perform poorly on SPT recall when list items can

not be readily organized unless the task provides cues

during retrieval which were also provided during

encoding. As has been observed in the verbal memory

literature, minimal age differences are seen when cues

are provided at both encoding and retrieval (Poon,

1985).

Another variable which has been reviewed in the

aging literature concerns the use of objects in the

performance of the task items (Norris & West, in

press). SPT studies which have not found significant

age differences in recall in the motor/verbal condition

have typically used lists in which some, if not all,

items required the use of objects (Backman & Nilsson,

1984, 1985; Dick & Kean, 1989; G. Cohen & Faulkner,











1989). No items in the present study required the use

of objects. It is possible that age differences in SPT

recall were magnified because the lists did not include

object cues. Those studies in which age differences in

SPT recall were found included lists in which all items

required the use of objects (Lichty et al., 1986),

lists in which only some items required the use of

objects (R. L. Cohen, Sandler, & Schroeder, 1987), and

lists in which no items used objects (Guttentag & Hunt,

1988; Knopf & Neidhardt, 1989). It is likely that the

presence of objects interacts with other important

effects such as rate of presentation (see Experiment

#3). Perhaps object cues, in interaction with other

variables, are particularly important in predicting age

differences in the motor/verbal condition because the

objects are processed during encoding, but do not serve

as cues during retrieval. It is possible that age

differences in the motor/verbal condition used here

were increased because there were no cues provided by

the objects during encoding. This issue will be

addressed in Experiment 2.

Finally, the manipulation of motor cues during

both encoding and retrieval produced interesting

results concerning serial position. Primacy effects











were eliminated when items were motorically encoded,

and recency effects were reduced when items were

motorically retrieved. This pattern was evident for

both young and older adults. It has been argued that

primacy effects are not found in SPT recall because SPT

recall is nonstrategic with respect to rehearsal (R. L.

Cohen, 1981). The absence of rehearsal for both young

and older adults may be a partial explanation for why

age differences are sometimes eliminated in SPT recall.

Age differences in verbal recall have been attributed

to young adults' spontaneous use of strategies such as

rehearsal (Hultsch & Deutsch, 1981). Contrary to

verbal recall tasks, in SPT recall young adults do not

appear to use rehearsal to a greater extent than older

adults. Hence, age differences in SPT recall may be

reduced in some cases because young adults are not

making greater use of rehearsal in SPT recall than

older adults.

The negative recency effect observed with motor

retrieval was unexpected. Two possible explanations

may be considered. First, it is possible that a

recency effect is not seen with motor retrieval because

the re-enactment of other items serve as interference,

thus reducing recall of the terminal items on the list.











Interference effects from competing motor acts have

been observed with the motor re-enactment of sentences

(Zimmer & Engelkamp, 1985; Saltz & Donnenworth-Nolan,

1981). This explanation is tenuous. A recency effect

results from recalling the last items on the list

first. Hence, there are no intervening items to create

an interference effect. Alternatively, it has been

argued that recency effects are due to a highly

accessible verbal memory trace (R. L. Cohen, 1981).

Atkinson and Shiffrin (1968) have also stressed the

verbal nature of short-term store. Hence, motor

retrieval may not produce recency effects because the

action recalled is not strongly influenced by a verbal

memory trace.

The serial position analysis underscores the

notion that verbal recall tasks and motor recall tasks

are governed by different memory laws. A comparison of

the serial position results from the verbally processed

information (i.e. the verbal/verbal condition) and the

motorically processed information (i.e. the motor/motor

condition) illustrates that recall of motorically

processed items is not influenced by item position.

When items are enacted at both encoding and retrieval a

jagged serial position graph is produced, suggesting











that some items have greater recall probability than

other items, regardless of their positions on the list.

This supports the notion that there are important item

characteristics which predict recall of motorically

processed information.

In conclusion, this study suggests that age-

related memory declines may not be pervasive. Age

differences in SPT recall were eliminated when motor

cues were provided during retrieval. However, the

influence of motor retrieval cues does not stand in

isolation. The benefits of organization cues may be

enhanced by motor processing, or may be reduced when

object cues are absent during encoding. Furthermore,

the magnitude of age differences in recall is item-

dependent. Individual action items vary along numerous

dimensions including the number and kind of body

movements required, the types of multimodal cuing, the

familiarity of the goal of the action, and so forth.

These dimensions may together predict the

distinctiveness of individual items which may in turn

predict the retrievability of the memory trace. The

relevant stimulus properties are not the same for word

recall and SPT recall. More systematic analysis of








80


item characteristics that predict age differences in

SPT recall are needed.











Table 1-1

Means (and Standard Deviations) for Number of Items
Recalled


Group (32a)


Younger
21.1 (3.6)


Older
17.1 (3.9)


Condition(32a)

Verbal/Verbal Verbal/Motor Motor/Verbal Motor/Motor
16.1 (4.7) 19.8(3.5) 20.7(4.0) 19.8 (3.4)


Organization(16a)


Organized list
10.0 (2.3)


Unorganized list
9.2 (2.5)


Note. aMaximum Possible Score
For each main effect, differences were
significant at R < .0001












Table 1-2

Percent of Young and Older Subjects Recalling Items


Organized List


Item
Snap your fingers
Cross your legs
Blow in the air
Lean to the left
Tap your foot
Stick out your tongue
Tilt back in chair
Clasp your hands together
Bend forward at the waist
Raise one leg
Count finger on hand
Laugh
Stretch your legs
Shrug your shoulders
Cross your fingers
Yawn


Young
87
87
52
92
62
46
71
69
73
53
37
74
73
41
69
89


Old
60
87
34
74
57
44
64
56
70
44
34
54
66
43
56
87


Difference
27*
0
18*
18*
5
2
7
13
3
9
3
20*
7
-2
13
2


Unorganized List


Item
Shake head back and forth
Pretend to be asleep
Name street you live on
Knock on the table
Clap your hands
Cup hand over ear
Name four colors
Point to your mouth
Pretend to play the piano
Nudge arm of the chair
Clear your throat
Scratch your nose
Smooth your hair
Blink three times
Flex your arm
Wave good-bye


Young
69
62
79
74
36
59
72
44
37
85
49
73
77
74
81
71


Old
52
37
56
55
22
51
37
25
30
72
34
84
67
62
60
65


Difference
17*
25*
23*
19*
14
8
35*
19*
7
13
15
-11
10
12
20*
6


Note. *R < .05, chi-square test.
















25


C 20-
0
R
R
E 15-
C YOUNG
T "

R OLDER
E 10-
C
A
L
L 6-


0


Figure 1-1: Recall by Age and Condition




















0
0
R
R 8
E
C 6 YOUNG

R OLDER
E
C 4-
A
L
L
2-


0
ORGANIZED UNORGANIZED
LISTS


Figure 1-2: Recall by Age and Organization (Exp 1)



























YOUNG

-+- OLDER


40 -


SI I I I I I I
1-2 3-4 5-6 7-8 9-10 11-12 13-14 15-16
ITEM POSITION




Figure 1-3: Serial Position--Verbal/Verbal


100 r


80o


60 -


























-- YOUNG
-I- OLDER


20 -


Figure 1-4: Serial Position--Motor/Verbal


100 r


80 -


60 -


40 -


I I I I I I 1 1
1-2 3-4 5-6 7-8 9-10 11-12 13-14 15-16
ITEM POSITION



























YOUNG

-+- OLD


40 h


20 F


Figure 1-5: Serial Position--Verbal/Motor


100 r


80 -


60 -


I I I I I I I I
1-2 3-4 5-6 7-8 9-10 11-12 13-14 15-16
ITEM POSITION



























- YOUNG

--- OLDER


40 1


I I I I I I I I


1-2 3-4


5-6 7-8 9-10 11-12 13-14 15-16
ITEM POSITION


Figure 1-6: Serial Position--Motor/Motor


100 r


80 F


60 F














EXPERIMENT 2

Introduction

Age differences in the recall of SPTs have often

been eliminated in studies in which all, or most, list

items required the use of an object (Backman, 1985;

Backman & Nilsson, 1984, 1985; Dick & Kean, 1989). In

contrast, age differences in SPT recall have been found

when few or no items required the use of an object (R.

L. Cohen, Sandler, & Schroeder, 1987; Guttentag & Hunt,

1988; Knopf & Neidhardt, 1989). It is possible that

age differences in SPT recall were not eliminated in

Experiment 1 because no items required the use of

objects.

It has been argued that age differences in SPT

recall are eliminated because SPTs involve multiple

sensory stores including visual, auditory, motor, and

tactual (Backman & Nilsson, 1984, 1985). It is

possible that motor cues alone are not sufficient to

eliminate age differences in recall. These multimodal

properties of SPTs, often present when objects are

used, may be as important as motor cues per se in

eliminating age differences in recall. An age deficit











may be found in SPT recall when no objects are used to

carry out the actions because fewer stimulus properties

are provided.

The multimodal hypothesis suggests that the

combination of visual and motor cues are necessary to

eliminate age differences in SPT recall. Evidence for

the multimodal hypothesis comes from studies that

demonstrate that neither motor cues nor object cues

alone are sufficient to eliminate age differences in

activity memory. Young adults recalled significantly

more performed actions than older adults when no

objects were used (e.g., items like "shake your head

yes" were used), suggesting that motor cues alone do

not eliminate age differences in SPT recall (Guttentag

& Hunt, 1988; Knopf & Neidhardt, 1989). In addition,

there is evidence that object cues alone are

insufficient to bring older adults' recall up to the

level of young adults. Young adults recalled more than

older adults when SPTs were presented in the form of

sentences with objects present, thereby eliminating the

motor cues (Backman, 1985). Furthermore, age

differences in recall have also been found when

subjects watched the experimenter perform the task,

thereby providing visual but not motor cues (Dick &




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 ETN6IGJ7K_GL51ND INGEST_TIME 2012-02-20T22:29:12Z PACKAGE AA00009074_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES