Egg (WWWIJHIHIIWIW“(HUIUH“UHHWHIWWI

H.

MICHIGAN STATE UNIVERSITY

Hill/II lfill/NI!I!!!llIll/Illlllllllll/l/lll

3 1293 01020 5932

ll

 

 

 

 

This is to certify that the

dissertation entitled

USE“ OF'EXECUTIVE CONTROL IN ACCESSING

EPISODIC AND SEMANTIC MEMORY
IN PATIENTS WITH ALZI-IEIMER‘S DEMENTIA

presented by

Daniel Malcolm Spica

has been accepted towards fulfillment
of the requirements for

PhD degree in m

MOM m

Major professor

 

Date loll”! '75‘

MS U i: an Affirmative Action/Equal Opportunity Institution

 

 

 

49—“

 

 

LIBRARY
Michigan State
University

 

 

 

PLACE ll RETURN BOXtomnavothbMMm youttocotd.
TOAVOID FINES Mun onotbdonddoduo.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU I'M Affltmdlvo MOVE“ Opportunly ltltkwon

 

USE OF EXECUTIVE CONTROL IN ACCESSING
EPISODIC AND SEMANTIC MEMORY
IN PATIENTS WITH ALZHEIMER’S DEMENTIA

BY

Daniel Malcolm Spica

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Psychology

1994

ABSTRACT
USE OF EXECUTIVE CONTROL IN ACCESSING

EPISODIC AND SEMANTIC MEMORY
IN PATIENTS WITH ALZHEIMER’S DEMENTIA

BY

Daniel Malcolm Spica

Executive control functions are believed to mediate,
among other processes, the organization of material to be
stored into or retrieved from declarative memory stores.
The construct of declarative memory typically divides the
memory stores into two subsystems, episodic (memory of
autobiographical information and events), and semantic
memory (generalized knowledge including meanings of words).

Both executive control and memory deficits have been
documented in Alzheimer’s disease (AD) dementia; however, it
has not been clear to what extent executive control
dysfunction may differentially affect episodic and semantic
memory in AD patients. Similarly, while it has been
hypothesized that normal elderly males exhibit more
executive dyscontrol than elderly females, it is not known
if these relative weaknesses are more prominent in the
episodic or semantic memory systems.

Executive control deficits in 26 Alzheimer's disease
(AD) patients were examined in episodic and semantic memory
performances through comparisons with performances of 26
normal control subjects. The two declarative memory

subsystems were tested by the administration of the

California Verbal Learning Test to examine episodic memory,
and Category and Letter Fluency tasks to examine semantic
memory. Executive control functions during the memory
performances were evaluated through the performance
characteristics of a) semantic clustering (grouping words by
meaning), b) phonemic clustering (grouping words by sound),
c) Perseveration errors, and d) Intrusion errors.

While the AD patients exhibited significant numbers of
Intrusion and Perseveration errors compared to normal
controls, these deficits in executive control were not
specific to a single memory system (semantic or episodic).
No differences in semantic and phonemic clustering were
significant. In regards to memory impairments, the AD
patients demonstrated the greatest memory deficits on the
Category Fluency Test, a task believed to maximally tax the
semantic network, providing support for semantic network
degradation in AD. Finally, analysis of the normal control
sample performances did not provide statistical support for
the prediction that elderly males would exhibit greater
decline in executive controls than females; male and female
normal control subjects demonstrated general equality in
levels of performance. Possible future directions for

research are discussed.

ACKNOWLEDGEMENTS

I wish to thank Dr. Abeles and the members of my
committee, Drs. Robert Caldwell, Gordon Wood, and A.S.
Aniskiewicz for their patience and guidance. I also wish to
express my deep appreciation to Dr. Anne Bogat for her
unending encouragement and support during the process of my
graduate education. Finally, I am grateful to Lisa Marie
Bell for her many contributions to the writing of this work,

and more generally for her gift of hope.

iii

TABLE OF CONTENTS

Page

List of Tables ............................................. v

List of Figures ........................................... vi

Introduction ............................................... 1

Memory Deficits in Alzheimer’s Disease ................ 1
Divisions of Long-Term Memory; Procedural And

Declarative Memory ................................... 3

Subsystems in Declarative Memory......................5

A Measure of Episodic Memory .......................... 7

Verbal Fluency Tasks as a Measure of Semantic Memory..8

Role of Executive Control; An Overview ........ . ....... 8

Executive Control Deficits in Alzheimer’s Disease....14
Dissociation of Semantic and Episodic Memory

in Alzheimer's Disease ........ . ......... ... .......... 15
Hypotheses ................................................ 17
Method .................................................... 20

Participants ................................ . ..... ...20

Procedure ............................................ 21
Results ................................................... 24

Hypothesis 1 Analyses ................................ 26

Hypothesis 2 Analyses ............................. ...29

Hypothesis 3 Analyses ................................ 31
Discussion ................................................ 38

Future Directions .................................... 46
List of References ........................................ 53

iv

LIST OF TABLES

Table Page
1 Subject Characteristics ................................ 21
2 Summary of Correlation Coefficients from Independent

Raters .................... . .................... . ..... 23
3 Summary of Raw Scores on Memory tasks. ................. 24
4 Summary of Multivariate Test Statistics ................ 26
5 Summary of Univariate F-Tests.. ........... . ...... . ..... 26
6 Summary of T-Test Comparisons of AD Patient and Normal

Control Performance Characteristics .................. 28
7 Summary of Standardized AD Patient Performances

(Based on Normal Control Levels of Performance) ...... 30
8 Normal Subject Characteristics ......................... 31
9 Summary of Raw Scores of Normal Control Subjects

on Memory Tasks ...................................... 32

10 Summary of Multivariate Test Statistics

Comparing Normal Male and Normal Female Performance..34

11 Summary of Univariate F-Tests Comparing Normal Male

and Normal Female Performance ........................ 34

12 Summary of Significant Findings ........................ 38

LIST OF FIGURES

Figure Page

1

Performance Characteristics on Episodic
and Semantic Memory Tasks ...... . ............... ......25

Performance Characteristics of Alzheimer’s
Patients and Normal Controls (Standard Scores) ....... 28

Mean Performance Characteristic Levels of Normal
Controls on Episodic and Semantic Memory Tasks
(Standard Scores).... ..... . ..... ............. ........ 32

Performance Characteristics of Normal Controls
on Episodic Memory Tasks (Standard Scores). .......... 35

Performance Characteristics of Normal Controls
on Semantic Memory Tasks (Standard Scores) ........... 36

Sample Semantic Network "Vehicles" ..................... 42

vi

INTRODUCTION

In recent years there has been a surge of interest in
Alzheimer's disease (AD) both in the medical community and
the general public. The prevalence of AD in this country is
typically estimated around 6 percent in persons over 65
(Schneck, Reisberg, & Ferris, 1982), and the tragic nature
of AD's symptoms and prognosis has surely helped in making
it so prominent. At present, AD is incurable and its
symptoms untreatable. However, major advances have been
made in the study of AD, primarily in diagnostic techniques
differentiating it from other dementing disorders, some of
which have a greater likelihood of remission than AD (Wells,
1984). Memory loss, reputed to be among the best early
indicators of dementia (Lezak, 1983), has served as a
valuable area of study for the differentiation of AD victims
both from people with other dementing conditions, and from

otherwise healthy forgetful elderly persons.

Memory Deficits in Alzheimer’s Disease

 

Diagnostic techniques involving memory functions have
helped to provide the greatest diagnostic gains in

discriminating between AD victims and normal aged persons.

2
While decline in memory functions is a major component of
both normal aging and, to a greater extent, the pathological
process of AD (Poon, 1985), evidence is mounting that the
specific nature of the memory decline is quite different in
these two aged conditions. While routinized, "overlearned"
tasks and skills have been found to be well preserved in
normal aging as well as in at least the early and middle
stages of the AD process (Wilson, Kaszniak, & Fox, 1981),
new acquisition of factual information has been found by
various researchers to be particularly difficult for AD
victims (Joynt & Shoulson, 1985; Mesulam, 1985).

It is not yet clear if AD victims are deficient in
learning all types of new information. On tasks which
require recall and recognition of novel verbal or visual
spatial information, such as learning a list of words or set
of drawings, AD patients are consistently found to be
impaired beyond the decline associated with normal aging
(Flicker, Ferris, Crook, Bartus, & Reisberg, 1986).
However, only inconsistent evidence is available showing AD
patients as impaired on tasks requiring acquisition of new
cognitive-motor processes, such as reading words or images
in a mirror (Eslinger and Damasio, 1985; Gordon, 1984).

The possible dissociation of such "types" of memory
functions has led researchers to devise multi-system memory
theories (see Tulving, 1987, for discussion). Certainly
both types of learning tasks mentioned above (list learning

and motor process learning) require access to long-term

3
memory stores for good performance, but the nature of the
information needed is clearly different; indeed, the
technical distinction between these types of cognitive

functions is that of "Procedural" and "Declarative" memory.

Divisions of Long-Term Memory; Procedural and Declarative
MEEQEX

The necessity of dividing memory into general realms of
"knowing how" and "knowing what" has been recognized both in
psychology (e.g., Bergson, 1911; Bruner, 1969) and in
philosophy (Ryle, 1949). The direct distinction between
procedural and declarative memory was first made in the
field of artificial intelligence to delineate the knowledge
of skills and procedures from the knowledge about which we
(or computers) are able to make declarations (Winograd,
1975). This particular terminology has held fast in the
neurosciences, helping to accommodate experimental and
clinical findings that some amnesic conditions can result in
individuals with preserved abilities in learning, while at
the same time the subject does not have access to the
conscious experience of acquiring the particular skill. For
example, the now famous case of H.M. illustrates such a
dissociation; H.M. suffered a severe amnesic syndrome
following bilateral medial temporal lobe resection intended
to curtail his frequent epileptic seizures. Subsequent to
the temporal lobotomy H.M. was able to show improvement on a

mirror-tracing task (new skill learning) after repeated

4
daily testing, but had no recollection from one session to
the next of prior exposure to the task (Milner, 1962). H.M.
similarly showed improved performances on other
perceptual-motor tasks, such as bimanual tracking and rotary
pursuit tests, but was unable to recollect ever having seen
the tasks before (Corkin, 1968).

Other amnesic conditions have also been shown to
differentially affect skill performances from the memory of
acquiring the skill, such as amnesia caused by
electroconvulsive shock therapy. Squire, Cohen, and
Zouzounis (1984) taught a mirror-reading skill to depressed
psychiatric patients in anticipation of each patient’s
course of electroconvulsive shock treatments. Despite
marked retrograde amnesia following the treatments,
including an inability to remember being taught the skill,
the patients’ post-therapy mirror-reading performances
remained at normal levels. That is, their knowledge of the
procedure survived the amnesic process while the conscious
knowledge of acquiring the skill was lost (Squire, Cohen, &
Zouzounis, 1984).

Data collected with H.M. and with numerous other
patients has repeatedly demonstrated preserved learning
capacity during amnesia in skills as varied as
perceptual-motor (working jigsaw puzzles; Brooks & Baddeley,
1976), conceptual reasoning (tower of Hanoi task; N. Cohen,
1984), application of numerical rules (Wood, Ebert, &

Kinsbourne. 1982), and learning through priming effects

5
(Graf, Squire, & Mandler, 1984). Priming refers to the
facilitation of performance by prior exposure to material,
such as target words. For example in the Graf et al. (1984)
study, the amnesic subjects were shown a number of words
(they were thus "primed"). Then the subjects were provided
with a stimulus of three-letter word stems, and asked to
complete the word. The findings revealed a priming effect
of biasing the subject to complete the stimuli stems using
words to which they had just been exposed, rather than other
words sharing the same first three letters. Further, the
amnesics exhibited the priming effect to the same extent as
normal controls, despite the amnesics’ marked deficiencies
in recalling or even recognizing the target priming words
(Graf, Squire, & Mandler, 1984).

Work investigating priming effects for AD patients has
not shown the same facilitation evidenced by other types of
amnesics. So called "implicit" memory for verbal material,
as measured by lexical priming tasks has been found to be
severely impaired in AD victims compared to age-matched
controls (Butters, Heindel, & Salmon, 1990; Salmon,
Shimamura, Butters, & Smith, 1988). Likewise, implicit
memory for pictorial information, as measured by picture
priming, has also been demonstrated to decline rapidly in AD

samples (Heindel, Salmon, & Butters, 1990).

Subsystems in Declarative Memory

In further distinguishing long-term memory functions,

6

Tulving (1972) proposed that there exists two subsystems
within "propositional" (declarative) memory which he called
"episodic" and "semantic" memory. In his 1983 book
"Elements of Episodic Memory" which reviews the fund of
research generated after his 1972 proposal, Tulving
describes episodic memory as essentially the
autobiographical and subjectively experienced information
(memory of an event or episode), and semantic memory as
roughly the generalized knowledge we possess and draw upon
(including our mastery of concepts such as semantic meanings
of words). Some researchers have suggested that the
episodic/semantic dichotomy does not fully describe human
memory due to overlap in the systems (see McKoon, 1985,
1986; Ratcliff, 1986); however, Tulving (1983) stated:
"Episodic memory is concerned with unique, concrete,
personal experiences dated in the rememberer’s past;
semantic memory refers to a person’s abstract, timeless
knowledge of the world that he shares with others" (p. v).

While research into these memory systems has spanned
into animal models and experiments, for the purposes of this
study we will concentrate on the assessment of memory
abilities in humans. Over the years, human episodic memory
tasks have been developed in various forms, such as recall
of pictures (Nebes, Martin & Horn, 1984), picture
recognition (Damasio, Eslinger, Damasio, Van Hoesen, &
Cornell, 1985; Tulving, Schacter, McLachlin, & Moscovitch,

1988), story recall (Butters et al., 1987; Ostergaard,

7
1987), and numerous tests of list learning (e.g., Bushke &
Fuld, 1974; Delis, Kramer, Kaplan, & Ober, 1987; Ostergaard,
1987). Similarly, various versions of fluency tasks have
been developed to tap semantic memory, as discussed below.
In addition, there have been some semantic memory
investigations using measures requiring the recognition of
celebrity faces (Damasio et al., 1985), memory for
historical events (Cermak & O’Connor, 1983), and even
knowledge of the rules, jargon, and etiquette of golf

(Schacter, 1983).

A Measure of Episodic Memory

The recent advent of the California Verbal Learning
Test seems especially promising since the measure's design
was theory-driven, and it may be the only test to date that
quantifies (and provides normative data based on 273
subjects) the multifactorial ways in which an examinee
attempts to organize, encode, consolidate, and retrieve
verbal material processed through the episodic memory system
(Delis et al., 1987). The routine scoring procedures also
include indices of semantic clustering, serial clustering,
primacy-middle-recency effects, perseverations and
intrusions, discriminability, response bias, free and cued
recall, and many other dimensions pertinent to the concept
of executive control in memory processing (Delis et al.,

1987; Lezak, 1983).

8

Verbal Fluency Tasks as a Measureyof Semantic Memory

Tests of verbal fluency, which require subjects to
generate a list of words belonging to some prescribed group,
have been used in clinical neuropsychology since the
mid-19603 when they were found to be sensitive to cerebral
dysfunction, particularly of the left frontal and left
temporal cortical regions. Of the methods researched to
assess verbal fluency, the two most commonly employed are
(a) asking the subject to produce as many words as possible
in one minute beginning with a certain letter (letter
fluency), and (b) requiring as many words as possible within
a certain category, such as animals, fruits, or vegetables,
within a one minute time limit (category fluency). The most
popular representative procedure for letter fluency testing
is probably that of the Controlled Oral Word Association
Test, a component of the Multilingual Aphasia Examination
(Benton & Hamsher, 1976), which uses the letters "C", "F" or
"L". Likewise, procedures for category fluency tests are
found as part of the Boston Diagnostic Aphasia Exam
(Goodglass & Kaplan, 1983), the Western Aphasia Battery
(Kertez, 1980) and the Mattis Dementia Rating Scale (Mattis,

1976).

Role of Executive Control; An Overview
The concept of some controlling mechanism that mediates
and modulates our attention to stimuli and dictates the

process through which we are to receive, decode, and/or

9
respond to that stimuli has taken on many related forms over
the years in the scientific literature (see Pylyshyn, 1990).
Ashcraft’s (1989) description states: "This component has
the responsibility of parcelling out the mental resources of
attention in reaction to the three memory components'
[sensory, long-term, and short-term/working] changing
demands for processing resources" (p.66).

Among the most detailed conceptions of executive
control is that from Gordon D. Logan, who has demonstrated
that four major executive functions can be distinguished
(Logan, 1985):

1) Choice among alternative strategies to approach a

particular task.

2) Construction or instantiation of the strategy chosen

to enable performance of the task.

3) Execution and maintenance of the strategy to perform

the task in real-time.

4) Inhibition or disablement of the strategy in

response to changes in the goals or environment
that make the strategy irrelevant or

inappropriate.

Through a series of experiments, Logan showed that each
of his proposed functions have empirical consequences in
tasks employing common cognitive psychology techniques. For
instance, he presented the words ABOVE and BELOW to subjects

on a screen either above or below a fixation point (i.e.,

10

spatial Stroop task). The subjects were asked to identify
each presented word by pressing specified keys, and their
performances were evaluated by latencies and error rates.
In addition, Logan varied the tasks by manipulating the
frequencies at which the words identities and positions were
compatible or conflicting; e.g., the word Amnm positioned
above the fixation point = compatible, Amnm presented below
the fixation point = conflicting. As Logan describes
(1985), numerous strategies are possible for good
performance as the relative frequencies of compatible/
conflicting word positions vary, and his experiments were
interpreted as demonstrating that subjects will actively
choose the strategies enabling them to best achieve their
performance goals (Logan, 1980; Logan & Zbrodoff, 1979).

Similarly, using the same spatial Stroop task, Logan
added the dimension of cuing; before the presentation of the
word, an X would appear on the screen signalling where the
word will appear (above or below fixation). By varying the
delay intervals between the appearance of the cue X and
presentation of the target word, Logan and Zbrodoff found
that the subjects' response times began to decrease when a
400 to 600 msec delay was provided. That is, the subjects
decreased reaction times indicate that they were able to
construct a response strategy if given 400 to 600 msec to do
so (Logan & Zbrodoff, 1982).

Logan's work was less definitive on the nature of

strategy execution and maintenance function, except to say

11
that the procedures do occur. He presented two potential
scenarios and provided research evidence for each: a) a
strategy may be engaged reflexively and ballistically to
similar stimuli with very little modification or
intervention (Logan 1978), or b) a strategy may require
continual monitoring and executive control for its
maintenance (Schneider & Shiffrin, 1977). Logan resolved
these competing scenarios and lines of evidence by
suggesting that different cognitive tasks and strategies
will require different levels of monitoring throughout their
execution.

Finally, in discussing inhibition or abandonment of
strategies as the environment changes, Logan (1985)
recounted evidence from stop-signal studies which require a
subject to cease an activity as soon as possible once a
"stop" signal has been given. For instance, Ladefoged and
colleagues (1973) found that subjects required only 200 msec
to stop speaking when signalled, and Logan (1982)
demonstrated that subjects stopped typing only 300 msec
after the stop signal. Further, Logan and Cowan (1984)
pointed out that the response to a stop signal appears to
take very similar amounts of time irrespective of task
complexity. For example, in comparing performances for
simple reaction times (responding to a light display) and
choice reaction times (responding only to a specified color
of light display), a choice reaction time response took on

average only 17 msec longer to stop than a simple reaction

12
time response (Logan, 1984).

As seen above, subjects can disengage from tasks very
rapidly, even when the tasks are relatively ballistic, such
as talking or typing. Therefore, Logan reasoned that some
functional entity that serves as an executive controller is
monitoring and, when necessary, discontinuing responses.
That is, an executive process may determine the significance
of the incoming stimuli, then either respond to it quickly
if the goals of the task have changed or an error is
detected, or place them in a line with the other responses
if they require the conventional response (Logan, 1985).

Inhibition probably constitutes the single most
widely-recognized function of executive control. For
instance, Martin, Caplan, and Hier (1984) associated verbal
perseverations with a lack of inhibition of inappropriate
response by executive control. Santo Pietro and Rigrodsky
(1986) later focused on a combination on what corresponds to
Logan's first (choice) and fourth (inhibition) executive
functions to interpret oral-verbal errors in adult aphasics.
They attempted to account for errors of perseveration by
looking to an inability to release information from working
memory, thereby causing the subject, during a list learning
task, to draw on a limited pool of words "trapped" in a
cycle between working and short-term memory (Santo Pietro &
Rigrodsky, 1986). They suggested that once a word had been
retrieved from the long-term memory store, it was retained

in working memory through non-volitional rehearsal, which in

13
turn reduced the capacity of working memory and impaired its
ability to add new words or resume the search of long-term
memory. Considering Logan's proposed executive functions,
it is the responsibility of the executive control system to
release the information from working memory (choose to or
choose not to rehearse specific information in working
memory), as well as to inhibit uttering the wrong response.

Other putative errors made by the executive control
system include failure to "chunk" or cluster information in
meaningful arrays to facilitate retrieval or manipulation
(Delis et al., 1987; Lezak, 1983). Whereas it has been
demonstrated that the use of contextual cues for
organization of material to be memorized varies with age,
socioeconomic class, and other organismic dimensions (Craik,
Byrd, & Swanson, 1987), profound impairment of such
functioning is the hallmark of AD. As Moscovitch and Umilta
(1990) state, "A malfunctioning central executive [in
working memory] is typically an early symptom of dementia"
(p. 33).

As noted above, one view of executive control is that
of "...parcelling out the mental resources of attention..."
(Ashcraft, 1989, p. 66). A recent review of attentional
processes in older individuals reports advances in measuring
attention in a multitude of categories: divided attention,
attention switching, sustained attention, and selective
attention (McDowd & Birren, 1990). The equipment required

to test attention performances involve technology such as

14
electroencephalograms (Thompson & Marsh, 1973), visual
tracking equipment (Hartley, Kieley, & McKenzie, 1987) and
other tests of chemical and electrical activity of the
brain, and, although such methods are beyond the scope of
this study, they may prove useful in the future for
determining to what extent general arousal is necessary for
utilization of executive control functions. Lowered general
levels of central nervous system arousal has been noted in
older as compared to younger normal adults (e.g., Woodruff,
1975); conversely, over-arousal has also been postulated to
explain attentional performance differences in older adults.
As noted by McDowd and Birren (1990), further investigation
is needed to determine the role arousal may play in the
various cognitive performances of older individuals.

For the present purposes, the analysis of intrusion
errors (lack of inhibition), perseveration errors (lack of
inhibition and possible lack of release from working memory
cycle), and inefficient production or recall grouping (lack
of clustering) will be used in this study to assess the
relative use of executive control in accessing episodic and

semantic memory stores.

Executive Control Deficits in Alzheimer’s Disease
James Becker (1987, 1988) has proposed a multiple

factor hypothesis of cognitive decline in Alzheimer’s

dementia. After analyzing data from 71 AD patients in his

1988 study, Becker discovered a two-component pattern of

15
dissociable impairments, comprising functions in (a)
secondary memory and (b) the central executive system as
described by Baddeley (Baddeley, 1986; Baddeley, Bressi,
Della Sala, Logie, & Spinnler, 1986).

Secondary memory refers to the episodic long-term
memory operations discussed above. The central executive
system oversees the actions of short-term memory processes
by parcelling out attentional resources to the memory
stores, also discussed above. Becker found these two
components to be dissociable in his sample even to extent
that two of his subjects demonstrated unique deficits for
one of each component. That is, one patient was impaired on
secondary memory tasks while scoring within normal limits on
executive system tasks, and the other patient had the
reverse pattern of normal secondary memory scores and
impaired executive system performances (Becker, 1988). This
two-component hypothesis may prove useful in adding to our
conception of the previously observed inconsistent
dissociation between the two declarative memory systems in

AD patients.

Dissociation of Semantic and Episodic Memory in Alzheimer's
melee

While comparing performances of demented, Huntington
disease, and alcoholic-Korsakoff patients on tasks of
semantic and episodic memory, Butters and his colleagues

found the three groups to demonstrate different patterns of

16
impairment (Butters, Granholm, Salmon, Grant, & Wolfe,
1987). To investigate the subdivisions of long-term memory,
they used recall of narrative passages to test episodic
memory, and both category and letter fluency to test
semantic retrieval. While all three patient groups were
profoundly impaired on memory for passages, their
performances on fluency tasks differentiated the AD group
from the Huntington disease and alcoholic-Korsokoff
patients. Specifically, on both the letter and category
fluency tasks, the Huntington disease patients exhibited
moderate impairment, and the alcoholic-Korsokoff patients
were found to be severely impaired. Conversely, the AD
patients were found to have deficits only in category naming
fluency, while showing preserved abilities in letter fluency
(Butters et al., 1987).

These findings are in partial conflict with earlier
studies of AD performances on letter fluency tasks (e.g.
Ober, Dronkers, Koss, Delis, & Friedland, 1986; Rosen,
1980). Because of this inconsistency in the literature
pertaining to fluency performances in AD samples, it was of
interest to replicate the procedures with additional
analysis of the strategies used (phonemic and semantic
clustering) and errors made by the patients. Perhaps as
Becker (1988) suggested a differentiating factor between AD
subjects and their age peers is their faulty use of
executive control when attempting to recall information from

their long-term memory store.

Hypotheses

The purpose of this study was to investigate
similarities and differences in how AD patients access their
two declarative (episodic and semantic) long-term memory
stores. To that end, this study compared the utilization of
specific executive control functions by AD subjects on an
episodic memory test relative to their utilization of such
functions during tests of semantic memory. To determine the
relative sufficiency of the AD access strategies, patterns
of performance of the AD group were compared to those
exhibited by age matched normal controls on the same tasks.

In light of the preceding discussion, executive control
access strategies were examined by analyzing performances
during a list-learning task (episodic memory) and two verbal
fluency tasks (semantic memory). Four parameters of
executive control during information access were identified.
These parameters are performance characteristics of a) the
response errors made, and b) the extent and types of
clustering used.

Analyzable errors were of two classes, comprising:

1. Perseveration of response within a task trial

2. Intrusion errors of inappropriate responses
As noted above, efficiency of executive strategy during
retrieval of information from long-term memory stores can be
assessed by examining the use of clustering of information
retrieved. A lack of clustering is believed indicative of

compromised executive control. Clustering of responses was

17

18
operationally defined by two related parameters, comprising:
1. Extent of semantic clustering (word meanings)
2. Extent of phonemic clustering (word sounds)

As described above, these performance characteristics
reflect the quality and extent of executive control
exercised by both participant groups during episodic and
semantic tasks. A number of contrasts were examined within
and between groups, and were intended to help elucidate the
nature of executive processing in memory access through a
variety of comparisons; normals vs. AD, females vs. males,
semantic vs. episodic functions, perseverative errors vs.
intrusion errors, semantic vs. phonemic clustering, etc.
Although the existing literature did not lend itself to
hypotheses about all of these contrasts, a number of

predictions were possible, discussed below.

Hypothesis 1:

In light of the above discussion, executive control of
AD patients was expected to be more compromised during tasks
of episodic memory as opposed to semantic memory tasks.
Therefore, the AD group was expected to exhibit greater
impairment of performance on each of the parameters (errors
made, and extent and type of clustering used) during an
episodic memory task (California Verbal Learning Test) than
during semantic memory tasks (verbal fluency). The levels
of impairment on these parameters were determined relative

to the age-matched controls. That is, it was predicted that

19
the AD subjects would utilize less semantic and phonemic
clustering and to exhibit more perseveration and intrusion
errors during the episodic task than during the semantic

tasks.

Hypothesis 2:

Consonant with existing AD literature, the AD subjects
were expected to show inferior performances on all tasks in
comparison with normal controls. Further, based on the
previous studies noted above, the rank order of task
performance discrepancy from normal levels was expected to
be as follows, beginning with the task predicted to yield
the lowest performance:

1. List-learning task (California Verbal Learning Test)

2. Category fluency task

3. Letter fluency task

Hypothesis 3:

In light of the fact that normal aged males have been
fcnind to exhibit more signs of frontal lobe pathology than
aged normal females (Albert & Kaplan, 1980; Veroff, 1980)
and one such system is involved in executive control (Lezak,
1983; Milner, 1964; Ramier & Hecaen, 1970; Walsh, 1978), the
Same indicators of faulty executive control discussed in
IhYpotheses 1 and 2 were expected to be more prevalent in the

normal male subjects than the normal female subjects.

METHOD

Participants

Twenty-six individuals who were involved in the ongoing
Michigan State University (MSU) Psychological Clinic Aging
Research Project volunteered to participate in the current
study. The MSU Psychology Clinic project participants
ranged in age from 61 to 83 years old. They had been deemed
through history and interview to be relatively healthy and
free from significant problems of mood state, and were
participating with informed consent in a longitudinal study
of cognition. Twenty-six individuals diagnosed with AD
involved in a similar ongoing study in the MSU Department of
Psychiatry Neurobehavioral Clinic and Research Center also
volunteered to take part in the current study. For those
participants deemed to be too cognitively impaired to
appreciate all aspects of his or her research consent,
consent was also obtained from the appropriate guardians.
Each of the 26 participants in the AD sample were
outpatients who had been previously diagnosed with AD by
their attending neurologist and met the criteria for
"primary degenerative dementia" outlined in the Diagnostic
and Statistical Manual of Mental Disorders, Third Edition
(American Psychiatric Association, 1987) and met the
criteria for "possible Alzheimer’s disease" developed by the
National Institute of Neurological and Communicative
Disorders and Stroke and The Alzheimer's Disease and Related
Disorder Association (McKhann, Drachman, Folstein, Katzman,

20

21
Price, & Stadlan, 1984).

Descriptive data for the two groups of participants are
shown in Table 1. Matching was successful in that the age
of the normal control group (Mean [M] = 71.27, standard
deviation [SD] = 6.40) did not differ significantly from
that of the AD group (M = 71.65, SD = 5.62): t(50) = 0.230,
p = .819. The groups did differ significantly in years of
education, as the healthy normal control group averaged
13.19 years (SD = 2.89) and the AD grouped averaged 11.54
years (SD = 1.68): t(50) = -2.527, p = .015. Each group
consisted of 10 males and 16 females, combining to a total
sample of 52 participants.

TABLE 1

SUBJECT CHARACTERISTICS

 

 

Normal Alzheimer’s
Controls Patients Total prob
N = 26 N = 26
Age
(MeantSD) 71.27 i 6.40 71.65 i 5.62 71.46 i 5.97 ns
Range 61 - 83 60 - 81 60 - 83
Years of Education
(MeantSD) 13.19 t 2.89 11.54 i 1.68 12.37 i 2.48 .015
Range 8 - 20 8 - l6 8 - 20
Gender (Frequency)
Males 10 10 20 ns
Females 16 16 32 ns

 

Procedure

Participants volunteering for either of the two MSU
projects were tested with a battery of cognitive tests and
interviewed as part of the ongoing studies. Each

participant was tested individually in a single session.

22

Included in the comprehensive battery were the following
measures: The Category Fluency task (Goodglass & Kaplan,
1983), the Letter Fluency Task (Benton & Hamsher, 1976), and
the California Verbal Learning Test (Delis et al., 1987).
The Category Fluency task requires the participant to name
as many animals as possible in one minute. The Letter
Fluency task requires the participant to provide as many
words as possible that begin with the letters "C", "F", and
"L" (separate trials of 60 seconds each). The California
Verbal Learning Test (CVLT) is a word-list learning task in
which the participant is aurally presented a list of 16
shopping items and required to recall as many as possible,
in any order (repeated for five trials total). All
participants were administered each measure according to the
exact instructions provided in the respective manuals.

Scoring the four parameters of performance
characteristics (Semantic clustering, Phonemic clustering,
Perseveration errors, Intrusions errors) was carried out on
the three measures according to the procedure prescribed in
the CVLT manual (Delis et al., 1987). The CVLT scoring
procedures were well suited for the fluency tasks as they
involved simply tallying either the raw number of
perseverations, intrusions, or the occurrence of consecutive
correct responses from the same category (semantic or
phonemic) depending on the dimension being scored (Delis et
al., 1987). The tallied raw scores for the four parameters

were divided by the total production of words for that test

23
to control for varying number of words produced.

Because the fluency tasks (Category and Letter) do not
involve finite constrained sets of words from which the
subjects may respond, a standard method was devised to
determine which responses were semantically and phonemically
related during these tasks. This was accomplished through
three independent raters who, blind to the subjects' group
membership and the hypotheses, tallied the occurrences of
semantic and phonemic clustering for every protocol. [Each
rater was a licensed practicing neuropsychologist.] Rater
disagreements were resolved by majority ratings.
Inter-rater reliabilities, as measured by Pearson
correlation coefficients between the three raters, ranged
from 0.701 to 0.942. Specific correlations of semantic and
phonemic clustering ratings are listed in Table 2.

TABLE 2

SUMMARY OF CORRELATION COEFFICIENTS
FROM INDEPENDENT RATERS

 

Ratings of Performance Characteristics During Category Fluency:

Semantic Clustering Phonemic Clustering
RATER 1 RATER 2 RATER 1 RATER 2
RATER 2 0.924 0.701
RATER 3 0.926 0.929 0.818 0.712

 

Ratings of Performance Characteristics During Letter Fluency:

Semantic Clustering Phonemic Clustering
RATER 1 RATER 2 RATER 1 RATER 2
RATER 2 0.853 0.932
RATER 3 0.889 0.942 0.931 0.947

 

RESULTS
Table 3 lists the means and standard deviations for
each of the three memory tasks, including indices of

performance characteristics.

TABLE 3

SUMMARY OF RAW SCORES ON MEMORY TASKS

 

Alzheimer’s Patients Normal Controls
Mean Standard Deviation Mean Standard Deviation

 

California Verbal Learning Test:

(Total Trials 1-5) 18.35 9.60 46.39 12.75
Semantic Clustering 0.243 0.324 0.372 0.162
Phonemic Clustering 0.071 0.093 0.050 0.031
Perseveration Errors 0.166 0.384 0.106 0.080
Intrusion Errors 0.697 0.993 0.069 0.108
Category Fluency Test 9.23 5.00 18.69 4.07

Semantic Clustering 0.579 0.319 0.551 0.127
Phonemic Clustering 0.075 0.185 0.008 0.020
Perseveration Errors 0.209 0.297 0.016 0.030
Intrusion Errors 0.083 0.317 0.000 0.000
Letter Fluency Test 22.39 11.51 39.31 12.56
Semantic Clustering 0.104 0.133 0.084 0.066
Phonemic Clustering 0.189 0.123 0.231 0.128
Perseveration Errors 0.096 0.153 0.033 0.051
Intrusion Errors 0.059 0.122 0.003 0.011

 

To make the various tests statistically comparable, all
performance data were transformed to Z-scores based on the
entire sample. A MANOVA was performed with one between-
subject factor (Diagnosis: AD or Normal Control) and two
within-subject factors: a) task type (Episodic memory task
[CVLT], and semantic memory tasks [Category Fluency and
Letter Fluency]) and b) Performance Characteristics

[Semantic clustering, Phonemic clustering, Perseveration

24

F1

The

See

25
errors, and Intrusion errors]. In this analysis, the
Category Fluency and Letter Fluency tests were averaged to
provide overall scores of semantic memory functioning for
each performance characteristic. Mean Z-score values for
the Episodic (CVLT) and Semantic (Category and Letter

Fluency tests) memory performance characteristics are

depicted in Figure 1.

Z—score

 

 

 

 

 

O .6
* AD Episodic
O .4 .
+ AD Semantic
0.2 * Normal Episodic
*‘ Normal Semantic
O
-O .2
-O .4
-O .6

 

 

 

Performance Characteristic

Figure 1: Performance Characteristics on Episodic and
Semantic Memory Tasks

The 3-way interaction of the MANOVA was not significant, as
seen on Table 4. Table 5 lists the related univariate

F-tests for the three factors.

26
TABLE 4

SUMMARY OF MULTIVARIATE TEST STATISTICS

 

 

Source Multivariate F df p

Performance Characteristics* [Perf C] 0.000 3, 48 1.000
Diagnosis x Perf C 7.992 3, 48 <0.001
Task Type x Perf C 0.000 3, 48 1.000
Diagnosis x Task Type x Perf C 2.388 3, 48 0.080

 

*Performance Characteristics [Perf C] include a) Semantic clustering, b)
Phonemic clustering, c) Perseveration errors, and d) Intrusion errors.

TABLE 5

SUMMARY OF UNIVARIATE F-TESTS

 

 

Source SS df MS F p
Between Subjects:

Diagnosis 10.421 1 10.421 3.407 0.071
Error 152.955 50 3.059

Within Subjects:

Task Type 0.000 1 0.000 0.000 1.000
Diagnosis x Task Type 1.157 1 1.157 1.138 0.291
Error 50.833 50 1.017

Perf C* 0.000 3 0.000 0.000 1.000
Diagnosis x Perf C 11.167 3 3.722 6.067 0.001
Error 92.035 150 0.614

Task Type x Perf C 0.000 3 0.000 0.000 1.000
Diagnosis x

Task Type x Perf C 4.665 3 1.555 2.752 0.045
Error 84.767 150 0.565

 

*Performance Characteristics [Perf C] include a) Semantic clustering, b)
Phonemic clustering, c) Perseveration errors, and d) Intrusion errors.

Hypothesis 1
With regard to the three hypotheses stated above,

Hypothesis 1 predicted that Alzheimer's patients would
exhibit more prominent executive dysfunction on the episodic

memory task (CVLT) than on the semantic memory tasks

27
(Category Fluency and Letter Fluency). That is, the AD
patients would exhibit greater differences from the normal
controls (strictly in terms of executive control performance
characteristics) while performing the CVLT than when
performing the Fluency tasks; these differences would be
evidenced as utilizing semantic clustering and phonemic
clustering less frequently, and providing more perseveration
and intrusion errors. While the data in Table 3 demonstrate
the overall performance superiority of the normal controls
over the AD subjects on the three memory tasks, Hypothesis 1
is restricted to executive controls alone, and to only the
AD subjects; the AD subjects did not show a significant
differential in their use of executive control relative to
the type of memory task. That is, the results pertaining to
executive function variables alone indicate no significant
Diagnosis x Task Type interaction (Table 5). Consequently,
Hypothesis 1 was not supported; AD subjects did not
consistently perform significantly lower on the CVLT than on
Fluency measures. However, a significant Diagnosis x
Performance Characteristic interaction was noted and
explored further. Mean Z-score values for the four
performance characteristics (collapsed across task type) are

depicted in Figure 2.

28

Z-score

 

"' AD Patients

7” Normal Controls

 

 

 

 

 

 

 

 

Performance Characteristic

Figure 2: Performance Characteristics of Alzheimer’s
Patients and Normal Controls (Standard Scores)

To further elucidate the nature of the interaction,
comparisons between AD patients and normal controls on the
individual Performance Characteristics were conducted,
revealing significant differences in AD vs. normal control
performances in Perseveration error rates and Intrusion

error rates across task types, as tabulated below (Table 6).

TABLE 6

SUMMARY OF T-TEST COMPARISONS OF AD PATIENT
AND NORMAL CONTROL PERFORMANCE CHARACTERISTICS

 

 

T df p
Semantic Clustering -0.706 50 0.484
Phonemic Clustering 1.019 50 0.313
Perseveration Errors 2.327 50 0.028

Intrusion Errors 3.515 50 0.001

 

29

In light of the fact that the AD and normal control
groups differed significantly in years of education (11.54
years and 13.19 years, respectively), the MANOVA was
repeated entering years of education as a covariate. The
results revealed that the education variable did not
significantly contribute to the Diagnosis x Performance
Characteristic interaction: Multivariate F(3,47) = 0.088, p

= 0.966.

Hypothesis 2
The first level of Hypothesis 2, which predicted that

the AD participants in this study would show inferior
performances on all tasks in comparison to normal controls,
was supported as suggested by the means and standard

deviations in Table 3. All AD vs. Normal comparisons were

significant: CVLT t(50) -8.96, p = <.001; Category
Fluency t(50) = -7.49, p = <.001; Letter Fluency t(50) =
-5.06, p = <.001.

To address the second level of Hypothesis 2 (predicting
that the AD patient’s performance discrepancy from normal
levels would be greatest for the CVLT, and least for the
Letter fluency task), additional analyses were conducted.
Standard scores were computed for the AD patients on all
three tasks (Z-scores based on the performances of the

normal controls). The resultant means and standard

deviations for the three tasks are listed in Table 7.

30
TABLE 7

SUMMARY OF STANDARDIZED AD PATIENT PERFORMANCES
(BASED ON NORMAL CONTROL LEVELS OF PERFORMANCE)

 

Mean Standard Deviation Range

 

California Verbal Learning Test

(Total Trials 1-5) -2.20 0.75 -3.56 - -0.66
Category Fluency Test -2.32 1.23 -3.86 - 0.08
Letter Fluency Test -1.35 0.92 -2.73 - 0.77

 

Comparisons of the scores in Table 7 revealed that
while the AD patients' CVLT performances were not
significantly more impaired than their Category Fluency
performances (t[25] = 0.614, p = .545), their Category
Fluency performances were significantly more impaired than
their Letter Fluency performances (t[25] = -4.295, p =
<.001). The CVLT performances were also significantly more
impaired than the Letter Fluency performances (t[25] =
-4.691, p = <.001). That is, Hypothesis 2 predicted the
rank ordering of the tasks, progressing from yielding the
greatest impairment to least impairment in AD patient
performance, would be as follows:

1. List-learning task (California Verbal Learning Test)

2. Category Fluency task

3. Letter Fluency task
The actual results found the following order:

1. Category Fluency task

2. List-learning task (California Verbal Learning Test)
3. Letter Fluency task

31

Hypothesis 3

Hypothesis 3 predicted that the normal male
participants would exhibit more executive dysfunction on the
three tasks than the normal female participants. That is,
the normal males would utilize semantic clustering and
phonemic clustering less frequently and make more
perseveration and intrusion errors than the normal females
in the sample. Demographics for the normal controls are
listed in Table 8. The normal male and female participants
did not differ significantly in age (t[24] = 0.760, p =
0.449) or levels of education (t[24] = 0.562, p = 0.579).
The means and standard deviations for the male and female

performances are provided in Table 9.

TABLE 8

NORMAL SUBJECT CHARACTERISTICS

 

 

MALES FEMALES prob
N = 10 N = 26
Age (MeaniSD) 72.50 i 5.97 70.54 t 6.72 ns
Range 62 - 80 61 - 83
Years of Education (MeaniSD) 13.6 i 2.37 12.94 i 3.21 ns

Range 9 - 20 8 - 16

 

~33

32

TABLE 9

SUMMARY OF RAW SCORES OF NORMAL CONTROL SUBJECTS
ON MEMORY TASKS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

tumfis quuss
Mean Standard Deviation Mean Standard Deviation
California Verbal Learning Test
(Total Trials 1-5) 42.00 14.13 49.13 11.42
Semantic Clustering 0.355 0.164 0.383 0.052
Phonemic Clustering 0.047 0.038 0.052 0.028
Perseveration Errors 0.071 0.062 0.128 0.084
Intrusion Errors 0.065 0.139 0.072 0.087
Category Fluency Test 9.23 5.00 18.69 4.07
Semantic Clustering 0.585 0.124 0.530 0.128
Phonemic Clustering 0.000 0.000 0.013 0.024
Perseveration Errors 0.021 0.035 0.013 0.028
Intrusion Errors 0.000 0.000 0.000 0.000
Letter Fluency Test 22.39 11.51 39.31 12.56
Semantic Clustering 0.089 0.058 0.081 0.073
Phonemic Clustering 0.283 0.149 0.198 0.105
Perseveration Errors 0.070 0.065 0.010 0.019
Intrusion Errors 0.000 0.000 0.005 0.015
ZAScor.
0.8 _.1__
I 1 Standard Devration
0.6 """"""""""""""""
DMean __l
0.4 ““1"" _ i I 7
. 2 , .‘i .
. F
_o.2 ‘ ‘ L ' '
_LL_
_o_4 ~~~~~~~~~~~~
o . ..... 1L ............
-o.a ‘b
to q, c Q’.
(<59 &c V53,\0 \3’ § ¢~§§O
*iEEP <€:§§P' ‘Sifép 42959

Figure 3: Mean Performance Characteristic Levels of Normal
Controls on Episodic and Semantic Memory Tasks
(Standard Scores)

33

Mean Z-score values for the four performance
characteristics are depicted in Figure 3.

Examination of Table 8 reveals that the performance
superiority of the normal female subjects over the normal
males was inconsistent. Statistical analyses were
conducted, and as data on only 10 male and 16 female normal
control participants were obtained for the study, caution
was warranted in interpreting the following results.

All performance data were transformed to Z-scores based
on the normal control sample. A MANOVA was again performed
with one between-subject factor (Sex) and two within-subject
factors: a) task type (Episodic memory task [CVLT], and
semantic memory tasks [Category Fluency and Letter Fluency])
and b) Performance Characteristics [Semantic clustering,
Phonemic clustering, Perseveration errors, and Intrusion
errors]. As in the previous MANOVA (investigating
Hypothesis 1), the Category Fluency and Letter Fluency tests
were averaged to provide overall scores of semantic memory
functioning. Whereas this combining of the two semantic
memory tasks was required by Hypothesis 1 in the previous
analysis, it was necessary to combine the scores in the
current analysis due to the lack of variation on the
Category Fluency Intrusion Errors variable (no normal
control subjects of either sex made an intrusion error
during the Category Fluency test). The MANOVA results
indicated the 3-way interaction was not significant, as seen

on Table 10. Table 12 lists the related univariate F-tests

 

34

for the three factors.

TABLE 10

SUMMARY OF MULTIVARIATE TEST STATISTICS COMPARING
NORMAL MALE AND NORMAL FEMALE PERFORMANCE

 

 

Source Multivariate F df p

Performance Characteristics* [Perf C] 0.025 3, 22 0.995
Sex x Perf C 0.464 3, 22 0.711
Task Type x Perf C 0.133 3, 22 0.939
Sex x Task Type x Perf C 2.495 3, 22 0.086

 

*Performance Characteristics [Perf C] include a) Semantic clustering, b)
Phonemic clustering, c) Perseveration errors, and d) Intrusion errors.

TABLE 11

SUMMARY OF UNIVARIATE F-TESTS COMPARING
NORMAL MALE AND NORMAL FEMALE PERFORMANCE

 

 

Source SS df MS F p
Between Subjects

Sex 0.178 1 0.178 0.150 0.702
Error 28.444 24 1.185

Within Subjects

Task Type 0.303 1 0.303 0.418 0.524
Sex x Task Type 5.694 1 5.694 7.850 0.010
Error 17.410 24 0.725

Perf C * 0.079 3 0.026 0.026 0.994
Sex x 1.490 3 0.497 0.492 0.689
Error 72.689 72 1.010

Task Type x Perf C 0.375 3 0.125 0.134 0.939
Sex x Task Type x Perf C 7.040 3 2.347 2.520 0.065
Error 67.055 72 0.931

 

*Performance Characteristics [Perf C] include a) Semantic clustering, b)
Phonemic clustering, c) Perseveration errors, and d) Intrusion errors.

A significant main effect for Sex was not found, nor
was a Sex x Performance Characteristic interaction (Table

11). Hypothesis 3 was not supported as males did not yield

35
11). Hypothesis 3 was not supported as males did not yield
significantly different Performances Characteristics than
the females. However, a significant Sex x Task Type
interaction was evident (Table 11). Mean Z-score
performance characteristics values for the Episodic (CVLT)
and Semantic (Category and Letter Fluency tests) tasks of
the male and female normal control subjects are depicted in

Figure 4 and Figure 5.

Z-score

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

(334
El Male Episodic
02 ...... ..
D Female EpiSOdIC
C) E I I I C“?
l_Tl
_Q_2 ................ . . . .......
-0.4 .................
4343
0\\<\Q Q§®Q \xo‘f’ \xo‘g
004a. Gog}: \°°Qz é\o°©
235° Q~<~°° 0‘90

Performance Characteristic

Figure 4: Performance Characteristics of Normal Controls on
Episodic Memory Tasks (Standard Scores)

36

Z-score

 

B Male Semantic

I Female Semantic

 

 

 

 

 

 

 

 

 

 

_\¢Q - 09 0‘9 ‘9
c?) c>° ~ 0“ - 0°
- o - 0 (8° ‘3‘
(S 6° Q°\ ‘0
65566 $00.22. 639 \°
Q Q

Performance Characteristic

Figure 5: Performance Characteristics of Normal Controls on
Semantic Memory Tasks (Standard Scores)
Examination of the raw data (Table 9) and of the
Z-scores (Figures 4 and 5) suggests males made more errors

on the semantic memory tasks (Fluency Tests) than the
females. In fact, 9 of the 10 normal control males made at
least one perseveration error, while 5 of the 16 females
made such errors. Conversely, only two control subjects
(both women) made intrusion errors (both on the Letter
Fluency test). However, post hoc comparisons of the
individual Task Types were conducted, revealing no clear
significant differences between normal male and female
participant performances on either their episodic (F[1, 24]

= 2.428, p = 0.132) or semantic task performances (F[1, 24]

37

0.132) or semantic task performances (F[1, 24]

2.428, p

3.533, p .072). Although Hypothesis 3 was not fully
supported, a trend for males to make more perseverative
errors on the semantic memory tasks was noted, particularly

the Letter Fluency Test.

DISCUSSION

The significant results of the analyses are summarized in

Table 12.

TABLE 12

SUMMARY OF SIGNIFICANT FINDINGS

 

Hypothesis 1 Analyses:
Diagnosis x Performance Characteristic Interaction
- AD more Intrusions than controls
- AD more Perseverations than controls

 

Hypothesis 2 Analysis:
Rank Order of AD Patients’ Test Performances

Greatest Impairment
i Category Fluency
> Not Significant
i CVLT
> Significant
i Letter Fluency
Least Impairment

 

Hypothesis 3 Analysis:
Sex x Task Type Interaction

- Males more errors on Fluency Tasks than females

 

The Hypothesis 1 analysis was intended to determine if
executive control in the AD sample was more compromised
during either the episodic or semantic memory tasks. The
findings suggest no Diagnosis x Task interaction; although
the AD patients did not perform as well as the normal
controls on any of the memory tasks and they exhibited
inferior executive control abilities, the AD subjects did
not demonstrate a differential failure of executive controls

(relative to the normal controls) on the episodic or

38

39
semantic tasks. Consistent with the findings of Becker
(1988), the more general failure of executive control --
across task types-- was found in the AD performances.
Specifically, the AD patients made significantly more
Intrusion and Perseveration errors than the normal control
subjects across tasks.

Overall, the Hypothesis 1 analyses (i.e., no
significant differences between execution function mediation
in episodic and semantic systems) are consistent with
Logan’s and Cowan's (1984) suggestion that executive
functions work with similar efficiency irrespective of task
type. That is, the data presented here suggest that the
executive control failure seen in AD is not system-specific,
but that the executive functions, though compromised in AD,
retain relative independence from the episodic and semantic
memory systems. This independence is in further agreement
with neuropsychological theories of attention developed by
Pribram and McGuiness (1975), McGuinness and Pribram (1980),
Tucker and Williamson (1984), and G. Goldberg (1987), which
hold the executive controls as separate from, if not
superordinate over, the other cognitive processes they

mediate, such as memory.

Hypothesis 2 predicted the episodic memory task (CVLT)
would reveal the greatest impairment for the AD sample.
This was not supported. Category Fluency, one of the

semantic memory tasks, revealed the greatest impairment in

40
the AD sample. This performance gradient is likely not due
to executive control deficits, as Hypothesis 1 analyses
indicated no significant differences in the use of executive
control between episodic and semantic memory systems.
Alternatively, the significant impairment on the Category
Fluency task revealed in the Hypothesis 2 results may be
more reflective of a deterioration in the underlying
knowledge system in AD patients.

Although the Category and Letter Fluency tasks were
grouped together in the analyses to form an index of
semantic memory, hindsight suggests the two tasks may differ
substantively. There is some evidence suggesting that the
Category and Letter Fluency tasks both enlist frontal and
temporal territories in the brain (Frith et al., 1991);
however, evidence from brain-lesioned patients suggests a
relatively greater reliance on left frontal regions during
the Letter fluency task than the Category Fluency task
(Benton, 1968; Milner, 1967; Perret, 1974). In addition,
naming deficits for words within specific categories have
been reported from temporal lobe damage (Hart, Berndt, &
Caramazza, 1985; McKenna & Warrington, 1980).

Irrespective of mediating brain structures, the
Category and Letter Fluency tasks share the requirements of
initiation and maintenance of search and retrieval
processes. The tasks differ behaviorally in the type of
strategy used during the search and retrieval processes;

reports suggest Letter Fluency is guided by phonemic and/or

41
lexical cues, while production in Category Fluency tasks is
more dependent on the integrity of the underlying structure
of semantic knowledge systems (Bayles, Salmon, Tomoeda,
Jacobs, Kaszniak, & Troster, 1989)

It has recently been suggested that the structure of
semantic knowledge --the 'semantic network'-- is maximally
affected in the cognitive degeneration of AD (e.g., Ober &
Shenaut, 1988; Schwartz, Marin, Saffran, 1979; Smith,
Murdoch, & Chenery, 1989). The 'semantic network' refers to
the knowledge base of an individual and is assumed to be
organized hierarchically (Warrington, 1975). General
concepts and attributes of information occupy superordinate
‘positions, while more specific attributes and exemplars are
positioned subordinately (see Figure 6). It has been
proposed that access to a specific category member begins
with accessing a superordinate category in the hierarchical
network, with progressive activation downward, resulting in
increasingly specific semantic representation (Smith et al.,
1989), although at least one case study has been reported
revealing exceptions to this general-to-specific rule (Rapp
& Caramazza, 1989) .

One way in which the degradation of the semantic
network in AD patients has been observed is in their
decreased ’depth' of search; that is, when engaged in memory
search for meaningful words (e.g., during a Letter Fluency
task) normal subjects tend to produce specific exemplars of

categories (Rosch, Mervis, Gray, Johnson, & Boyes-Bream,

 

42
1976), while AD patients have been found to provide more
general classes of exemplars (Martin & Fedio, 1983). For
example, whereas a normal control subject when asked to list
words beginning with the letter 'F' may produce the words
"flapjacks", "Fritos", and "frankfurters", AD patients tend
to produce less specific exemplars such as "food." Such a
pattern of semantic network degradation is described as
'bottom up’ decay, as the lower end of the network (the most
specific exemplar of a category) apparently becomes hardest

to access first in AD.

 

Vehicles

 

 

 

 

 

 

 

 

 

 

 

 

 

Lend
Vehicles

 

 

 

 

 

 

 

 

Passenger

 

 

 

 

 

 

 

 

 

Non-
passenger

 

 

 

 

 

 

 

Water
Vehicles

 

 

 

 

 

 

 

Motorized

 

 

 

 

 

 

 

 

 

Non-
motorized

 

 

 

 

 

 

 

 

 

 

 

 

Automobile

 

Delivery

Semi-truck

Speedboet

Ferryboet

Seilboet

Canoe

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Ven

 

 

 

 

Figure 6: Sample Semantic Network "Vehicles"

The rank ordering of the tasks on which the AD patients

43
differed from the controls in the Hypothesis 2 analyses (see
Table 12) is consistent with the AD semantic network
degradation hypothesis in that the Category Fluency test
yielded the greatest impairment in overall word production
relative to normal controls (Table 3). Moreover, the Letter
Fluency task data suggest that in the AD group the use of
phonemic cues remained relatively intact; for AD subjects
Letter Fluency was least impaired of the three tasks
administered, and in addition, the AD group’s use of
phonemic clustering on any of the three tasks was not
significantly different from normal controls. The fact that
the Category Fluency task in the present study yielded the
greatest AD decrements of semantic and episodic memory
measures is also consistent with a recent report by Monsch,
Bondi, Butters, Paulsen, Salmon, Brugger and Swenson (1994)

which yielded similar results.

The analysis of Hypothesis 3 did not provide
statistical support for the prediction that normal elderly
males would exhibit greater decline in executive controls
(i.e., less clustering and more errors) than the normal
elderly females. Male control subjects did produce more
perseveration errors than female control subjects, although
this difference was not statistically significant.

General equality of cognitive performance of male and
female control subjects has been reported by Kaufman and

colleagues, who found that while "fluid intelligence" (as

 

44

defined by Horn, 1978) declines as a function of age in late
adulthood, males and females do not differ in their rates of
decline (Kaufman, Kaufman-Packer, McLean, & Reynolds, 1991).
Decline in fluid intelligence is presumably related to
errors of perseveration in that perseveration demonstrates
difficulty making appropriate transitions in mental set.
Perseveration has been found to increase in later life; Ruff
and colleagues reported significant increases in
perseveration with age on a figural fluency test analogous
to the verbal fluency tasks used in the present study (Ruff,
Light & Evans, 1987). Similar findings were presented by
Daigneault, Bruan, and Whitiaker (1992) who found
significant increases in perseverative errors across a
variety of neuropsychological tasks in 58 adults aged 45-65
years when compared to 70 adults aged 25-35 years. Those
authors interpreted the results as suggesting significant,
measurable deterioration in brain function in the prefrontal
regions may occur in healthy normal persons prior to age 65.

Perseverative errors are routinely measured by the
Wisconsin Card Sorting Test (Heaton, 1981). This task
requires the subject to sort cards consistently to one of
three conceptual categories (color, form, or number); once
the subject demonstrates consistent set maintenance, the
conceptual categorizing rule of the task is shifted (e.g.,
from color of design to number of design on the card), to
which the subject must adapt through corrective feedback

from the examiner. Perseveration errors occur when the

 

45
subject is unable to switch conceptual sets to meet the new
categorizing rule.

The Wisconsin Card Sorting Test has repeatedly been
found sensitive to frontal lobe function (Bornstein, 1986;
Drewe, 1974; Hermann, Wyler, & Richey, 1988; Milner, 1963).
However, Heaton (1981) found no sex differences on rates of
perseverative errors for a sample of 150 adults aged below
65 years. When an older sample was studied (35 males and 56
females, aged 45-83 years), significant sex differences
emerged for the Wisconsin Card Sorting Test, with males
committing significantly more perseveration errors on the
task than females, as well as scoring lower on four other
standard indices from the same test believed reflective of
frontal lobe efficiency: Number of Categories Achieved,
Total Errors, Percent Conceptual Level Responses, and Number
of Trials to First Category (Boone, Ghaffarian, Lesser,
Hill-Gutierrez, & Berman, 1993).

Veroff (1980) also reported a significant positive
correlation in a healthy normal aged sample between age and
findings on neuropsychological tests suggestive of frontal
lobe dysfunction. In that study men exhibited more signs of
frontal dysfunction than women. Recent neuro-imaging
studies have revealed cerebral blood flow changes in both
sexes with advanced age, and that decreases in blood flow to
the frontal regions is more pronounced in older men than
women (Mathew, Wilson & Tant, 1986).

The normal control sample in the current study did not

 

cc
di
We
th
fl
fu
hi.
Va:

3f;

46
exhibit significant sex differences in executive function
performances. The males did exhibit the more perseverative
errors than females on the Fluency tasks, but the finding
was not statistically significant. Perhaps this tendency
for the normal males to commit more perseverative errors may
have been more robust with a larger sample size, or a sample
of more advanced age. Nevertheless, while it is tempting to
speculate about frontal lobe dysfunction in the males of the
sample, the remaining performance characteristic data did
not show tendencies toward sex differences, and, consistent
with Kaufman et al. (1991), provided no evidence of sex
associated executive dyscontrol or frontal lobe dysfunction.
The results do reveal, however, the importance of
sex-specific normative data for clinical applications of the

Verbal Fluency tasks.

Future Directions

The general findings of the three analyses are
consistent with much of the existing literature: no
differential use of executive controls across tasks types
was evidenced in the AD patients; the AD patients showed
their greatest memory performance decrements on semantic
fluency tasks; normal control males exhibited more executive
function errors than females on fluency tasks. However,
hindsight suggests that the study could be modified in
various ways to minimize potential confounds that may have

affected the data. As in most experimental designs, two

 

47
general areas of potential confound could be addressed in
future efforts to replicate the current study: a)
difficulties in experimental samples or measures, and b)

shortcomings in the related constructs.

Experimental Sample:

The primary concern with the experimental sampling in
the current study is that the relatively small sample sizes
(26 AD, 26 normal) may have led to Type II statistical
errors. Larger cell sizes may have also benefitted this
study by increasing the occurrence of intrusion errors,
allowing more meaningful analysis of this behavior in both
sample populations. The mean number of intrusion errors
during either Fluency task was less than one error for both
the AD patients and normal controls (with no normal control
subjects making intrusion errors during Category Fluency).
Such small numbers raise the potential for floor effects in
the analyses.

It should be noted, however, that while more subjects
is generally better than less, the logistics of clinical
studies often make amassing a greater N size prohibitive to
the completion of the work. On this point Kraemer (1981)
has suggested that any experimental sample size be
determined by the four principle factors of a) acceptability
in the field of research, b) feasibility, c) power, and d)
cost. In addition, she states that for most designs in the

field of clinical psychiatry a sample size of 20 subjects

re

CU

48
per cell frequently strikes the proper balance between these
four factors (Kraemer, 1981).

Others aspects of amassing the sample could be
improved. Specifically, the normal control subjects had
significantly more years of education than the AD patients
(13.19 and 11.54 years, respectively). Higher levels of
education have been associated with greater efficiency in
some executive control functions; e.g., the use of
contextual cues for guiding memorization (Craik, Byrd, &
Swanson, 1987). However, the influence this education
disparity had on the present data appears insubstantial; the
data pertaining to Hypothesis 1 was re-analyzed with
education as a covariate and did not produce different
results than without the covariate. Nevertheless, the
possibility exists that the disparity in education has
influenced the data in an unforeseen manner, and an

education-matched sample is always more desirable.

Experimental Measures:

In addition to improving sampling procedures, it
appears possible to refine the devices used to more
optimally measure executive control in the samples. The
CVLT, Category Fluency, and Letter Fluency tasks had the
advantage of clinical relevance as these devices are widely
used in patient evaluations of dementia. However, in
retrospect, one of the experimental computer measures

currently under development to measure executive controls

49
may be more suited to our analyses due to their greater
precision, such as those presented by Bilder, Turkel,
Lipschutz-Brock, and Lieberman (1993). These tasks have the
advantage of collecting reaction time data, providing more
subtle measures of executive processing. In addition, the
tasks are designed to separate executive control processes

from other mental operations, such as memory, when desired.

Revisions in Constructs:

A remaining possibility exists that the
episodic-semantic dichotomy itself does not optimally
describe human memory systems. In their 1994 chapter on
this topic, neuropsychologists E. Goldberg and Barr
questioned the sufficiency of the dichotomy based on cases
of medial temporal lobe and diencephalic amnesia. Rather
than conceiving knowledge as divided into procedural and
declarative systems (with the declarative system further
divided into semantic and episodic memory, as discussed
above), E. Goldberg and Barr proposed a modified taxonomy
dividing knowledge into Generic and Singular types. In this
scheme, "Singular" knowledge describes knowledge of specific
facts that pertains to single, unique entities (E. Goldberg
& Barr, 1994). Singular knowledge can be void of temporal
significance or association, e.g., "Rome is the capital of
Italy", or can maintain an intrinsic temporal quality, such
as knowledge for personal events or information: e.g., "I

met my spouse in Rome in 1950." Singular knowledge also

 

50
includes finite sets of information intentionally encoded
for later retrieval (i.e., episodic memory) such as shopping
lists and specific scenes from one’s experience. The
processes used during the CVLT task correspond to the
Singular-episodic system.

"Generic" knowledge, on the other hand, is void of
intrinsic temporal quality, and pertains to entire classes
of entities: e.g., "Mature frogs usually have four legs",
"Both apples and bananas have peels." Linguistic
(definitions, uses of words, grammatical rules, etc.)
information is also within the Generic knowledge system.
The Fluency tasks in the present study correspond to the
Generic-semantic system. E. Goldberg and Barr make the
additional point that Generic knowledge is not limited to
lexical information, as it includes knowledge for proper
procedures (e.g., sequence for cleaning a pipe), as well as
peripheral aspects of meanings (e.g., "Camels do not have
antlers"). E. Goldberg and Barr (1994) relate the original
semantic-declarative taxonomy to their own by proposing,
"Singular knowledge, by definition, includes all episodic
and singular semantic knowledge. Generic knowledge includes
all procedural and generic semantic knowledge" (p. 73).
They use evidence from the amnesia syndromes to suggest
impairment of Singular knowledge or impairment of Generic
knowledge reflect two phenomenally and neuroanatomically
dissociable conditions.

Applied to the Goldberg/Barr model, the current study

51
compares a portion of the Generic knowledge system, with a
portion of the Singular knowledge system. As such, our
findings suggest no clear disparity in use of executive
controls in the two systems, but a more informative design
would encompass all four components of the model, including
the Singular-semantic and procedural aspects of knowledge.
Future investigations may determine the utility of the
Goldberg/Barr model in understanding degradation of memory

systems in AD.

In summary, the current investigation suggests that
patient's with Alzheimer’s dementia exhibit significant
executive function deficits in the form of intrusion and
perseveration errors compared to normal controls, and that
this decline in executive control is not specific to a
single memory system (semantic or episodic). In addition,
similar to recent findings pertaining to semantic network
degradation in AD, the patients in this sample demonstrated
the greatest memory deficits on a Category Fluency Test, a
task believed to maximally tax the semantic network.
Finally, analysis of the normal control sample performances
did not provide statistical support for the prediction that
elderly males would exhibit greater decline in executive
controls, but the data instead demonstrated an equality of
performance levels between the sexes.

It is a common hope among neuropsychology researchers

to learn something new about the human brain through the

52
study of clinical conditions such as Alzheimer's disease.
It is generally believed that executive controls operate in
superordinate positions relative to other cognitive
processes. Because the study presented here did not uncover
a dimension of system-specificity in executive functions, it
may provide further suggestion that the field is proceeding

on the proper course.

 

LI ST OF REFERENCES

LIST OF REFERENCES

Albert, M.L., & Kaplan, E. (1980). Organic implications of
neuropsychological deficits in the elderly. In L.W.
Poon, J.L. Fozard, L.S. Cermak, D. Arenberg, & L.W.
Thompson, (Eds.), New directions in memory and aging:
Proceedings of the George A. Talland memorial conference
(pp. 253-259). Hillsdale, N.J.: Lawrence Erlbaum.

American Psychiatric Association. (1987). Diagnostic and
statistical manual of mental disorders (3rd ed.). a
Washington DC: Author.

 

Ashcraft, M.H. (1989). Human memory and cognition.
Glenview: Scott, Flersman and Company.

Baddeley, A.D. (1986). Working memory. New York: Oxford
University Press.

Baddeley, A.D., Bressi, S., Della Sala, S., Logie, R., &
Spinnler, H. (1986). Dementia and working memory.
Quarterly Journal of Experimental Psychology, 38A,
603-618.

Bayles, K.B., Salmon, D.P., Tomoeda, C.K., Jacobs, D.,
Kaszniak, A.W. & Troster, A.I. (1989). Semantic and
letter category naming in Alzheimer's patients: A
Predictable Difference. Developmental Neuropsychology,
e, 335-347.

Becker, J.T. (1987). A two-component model of the memory
deficit in Alzheimer’s disease. In R.J. Wurtman, S.H.
Corkin, & J.H. Growden, (Eds.), Alzheimer's disease;_
Advances in basic research and therapies (pp. 343-348).
Cambridge: Center for Brain Sciences and Metabolism
Charitable Trust.

Becker, J.T. (1988). Working memory and secondary memory
deficits in Alzheimer's disease. Journal of Clinical
and Experimental Neuropsychology, 19, 479-497.

Benton, A.L. (1968). Differential behavioral effects in
frontal lobe disease. Neuro s cholo ia, p, 53-60.

Benton, A.L., & Hamsher, K. de S. (1976). Multilingual

53

54
aphasia examination. Iowa City: University of Iowa.

Bergson, H. (1911). Matter and memory. London: Allen and
Unwin.

Bilder, R.M., Turkel, E., Lipschutz-Brock, L., Lieberman,
J.A. (1993). Antipsychotic medication effects on
neuropsychological functions. Psychopharmacology
Bulletin, 28, 353-366.

Boone, K.B., Ghaffarian, S., Lesser, I.M., Hill-Gutierrez,
E. & Berman, N.G. (1993). Wisconsin Card Sorting Test
performance in healthy, older adults: Relationship to
age, sex, education, and IQ. Journal of Clinical

Psychology, 32, 54-60.

Bornstein, R.A. (1986). Contribution of various
neuropsychological measures to detection of frontal lobe
impairment. International Journal of Clinical

 

Neuropsychology, e, 18-22.

Brooks, D.N., & Baddeley, A. (1976). What can amnesic
patients learn? Neuropsychologia, 1e, 111-122.

Bruner, J.S. (1969). Modalities of memory. In G.A.

Talland, & N.C. Waugh, (Eds), The pathology of memory
(pp. 253-259). New York: Academic Press.

Butters, N., Grandholm, E., Salmon, D.P., Grant, 1., &
Wolfe, J. (1987). Episodic and semantic memory. A
comparison of amnesia and demented patients. Journal of

Clinical and Experimental Neuropsychology, 2, 479-497.

Butters, N., Heindel, W.C., & Salmon, D.P., (1990).
Dissociation of implicit memory in dementia:
Neurological implications. Bulletin of the Psyhconomic
Society, ee, 359-366.

Cermak, L.S., & O'Connor, M. (1983). The anterograde and
retrograde retrieval ability of a patient with amnesia
due to encephalitis. Neuropsychologia, g1, 213-234.

Cohen, N.J. (1984). Preserved learning capacity in amnesia:
Evidence for multiple memory systems. In L.R. Squire, &
N. Butters, (Eds.), Neuropsychology of memory
(pp.83-103) New York: Guilford Press.

Corkin, S. (1968). Acquisition of motor skill after
bilateral medial temporal excision. Neuropsychologia,
6, 255-265.

Daigneault, S., Bruan, C.M., & Whitiaker H.A. (1992). Early
effects of normal aging on perseverative and

55

non-perseverative prefrontal measures. Developmental
Neuropsychology, e, 99-114.

Damasio, A.R., Eslinger, P.J., Damasio, H., Van Hoesen,
G.W., & Cornell, S. (1985). Multimodal amnesic syndrome
following bilateral temporal and basal forebrain damage.
Archives of Neurology, 32, 252-259.

Drewe, E.T. (1974). The effect of type and area of brain
lesion on Wisconsin Card Sorting Test performance.
Cortex, 1Q, 159-170.

Flicker, C., Ferris, S.H., Crook, T., Bartus, R.T., &
Reisberg, B. (1986). Cognitive decline in advanced
age: Future directions for the psychometric
differentiation of normal and pathological age changes

in cognitive function. Developmental Neuropsychology,
2, 309-322.

Frith, C.D., Friaston, K.J., Liddle, P.F., & Frackowiak,
R.S.J. (1991). A PET study of word finding.
Neuropschologia, 22, 1137-1148.

Goldberg, E. & Barr, W.B. (1994). Selective knowledge Loss
in Activational and representational amnesias. In L.R.

Squire, & N. Butters, (Eds.). Neuropsychology of memory
(2nd ed.) (pp. 72-80) New York: Guilford Press.

Goldberg, G. (1987). From intent to action: Evolution and
function of the premotor systems of the frontal lobe. In
E. Perecman: The Frontal Lobes Revisited, (pp. 273-306)
New York: IRBN Press.

Goodglass, H., & Kaplan, E. (1983). Assessment of aphasia
and related disorders. Philadelphia: Lea and Febiger.

Graf, P., Squire, L.R., & Mandler, G. (1984). The
information that amnesic patients do not forget.
Journal of Experimental Psychology: Learning. Memory

and Cognition, mp, 164-178.

Hart, J., Berndt, R.S., & Caramazza, A. (1985).
Category-specific naming deficit following cerebral
infarction. Nature, 316, 439-440.

Heaton, R.K. (1981). Wisconsin Card Sorting Test Manual.
Odessa, FL: Psychological Assessment Resources.

Heindel, W.C., Salmon, D.P., & Butters, N. (1990).
Pictorial priming and cued recall in Alzheimer’s and
Huntington's disease. Brain and Cognition, 1;, 282-295.

Hermann, B.P., Wyler, A.R., & Richey, E.T. (1988).

56

Wisconsin Card Sorting Test performance in patients with
complex partial seizures of temporal-lobe origin.

Journal of Clinical and Experimental Neuropsychology,
e9, 467-476.

Horn, J.L. (1978). Human ability systems. In P.B. Baltes
(Ed.), Life-span development and behavior (pp.211-256).
New York: Academic Press.

Kaufman, A.S., Kaufman-Packer, J.L., McLean, J.E., &
Reynolds, C.R. (1991). Is the pattern of intellectual
growth and decline across the adult life span different

for men and women? Journal of Clinical Psychology, e1,
801-812.

Kertez, A. (1980). Western aphasia battery. London,
Ontario: University of Western Ontario.

Kraemer, H.C. (1981). Coping strategies in psychiatric

clinical research. Journal of Consulting and Clinical
Psychology, 32, 309-319.

Ladefoged, P., Silverstein, R., & Papcun, G. (1973).
Interruptibility of speech. Journal of the Acoustical
Society of America, pp, 1105-1108.

Lezak, M.D. (1983). Neuropsychological assessment. New
York: Oxford University Press.

Logan, G.D. (1978). Attention in character classifications:
Evidence for the automaticity of component stages.
Stop-signal studies of repetition priming. Journal of
Experimental Psychology: General, 107, 32-63.

Logan, G.D. (1980). Short-term memory demands of
reaction-time tasks that differ in complexity. Journal
of Experimental Psychology: Human Perception and
Performance, p, 375-389.

Logan, G.D. (1982). On the ability to inhibit complex
movements: A stop-signal study of typewriting. Journal

of Experimental Psychology: Human Perception and
Performance, e, 778-792.

Logan, G.D. & Cowan, W.B. (1984). On the ability to inhibit
simple thought and action: A theory of an act of
control. Psychological Review, 21, 295-327.

Logan, G.D. & Zbrodoff, N.J. (1979). When it helps to be
misled: Facilitative effects of increasing the
frequency of conflicting stimuli in a stroop-like task.

Memopy and Cognition, 1, 166-174.

57

Logan, G.D. & Zbrodoff, N.J. (1982). Constraints on
strategy construction in a speeded discrimination task.

Journal of Experimental Psychology: Human Perception

and Performance, e, 502-520.

 

Logan, G.D. (1985). On the ability to inhibit simple
thoughts and actions: II. Stop-signal studies of

repetition priming. Journal of Experimental Psychology:

LearningyyMemory and Cognition, 11, 675-691.

Martin, A., & Fedio, P. (1983). Word production and
comprehension in Alzheimer's disease: The breakdown of
semantic knowledge. Brain and Lan ua e, 12, 124-141.

Mattis, S. (1976). Mental status examination for organic
mental syndrome in the elderly patient. In L. Bellack &

T. Karasu (Eds.). Geriatric psychiatpy: A handbook for

psychiatrists and primapy care physicians (pp. 313-334).
New York: McGraw-Hill.

Mathew, R.J., Wilson W.H., & Tant, S.R. (1986).
Determinants of resting regional cerebral blood flow in
normal subjects. Biolo ical Ps chiat , 21, 907-914.

McGuinness, D. and Pribram, K. (1980). The neuropsychology
of attention: Emotional and motivational controls. In:
The brain and psychology, (pp. 95-139). New York:
Academic Press.

McKenna, P., & Warrington, E.K. (1980). Testing for normal
dysphasia. Journal of Neurology. Neurosurgery. and

Psychiatpy, 3;, 781-788.

McKhann, G., Drachman, D., Folstein, M.F., Katzman, R.,
Price, & D., Stadlan, E.M. (1984). Clinical diagnosis of
Alzheimer's disease: report of the NINCDS-ADRDA work
group under the auspices of Department of Health and
Human Services Task Force on Alzheimer’s disease.
Neurology, 33, 939-944.

McKoon, G. (1985). The role of semantic information in
episodic retrieval. Journal of experimental Psychology:
Learning, Memory, and Cognition, 11, 742-751.

McKoon, G. (1986). A critical evaluation of the semantic-
episodic distinction. Journal of experimental

Psychology: Learning, Memopy, and Cognition, 1;, 295-
306.

Mesulam, M. (1985). Principles of behavior neurology.
Philadelphia: F.A. Davis Company.

Milner, B. (1962). Les troubles de la memoire accompagnant

58

des lesions hippocampiques bilaterales. In Physiologie
de l'hippocampe (pp.257-272). Paris: Centre National
de la Recherche Scientifique.

Milner, B. (1963). Effects of different brain lesions on
card sorting. Archives of Neurology, 2, 90-100.

Milner, B. (1964). Some effects of frontal lobectomy in
man. In J.M. Warren, & K. Akert (Eds.). The frontal
granular cortex and behavior (pp. 77-121). New York:
Grune and Straton.

Milner, B. (1967). Brain mechanisms suggested by studies of
the temporal lobes. In F.L. Darley (Ed.), Brain
mechanisms underlying speech and language (pp. 122-145).
New York: Grune & Stratton.

Monsch, A.U., Bondi, M.W., Butters, N., Paulsen, J.S.,
Salmon, D.P., Brugger, P., & Swenson, M.R. (1994). A
comparison of category and letter fluency in Alzheimer’s
disease and Huntington disease. Neuropsychology, e,
25-30.

Nebes, R.D., Martin, D.C., & Horn, L.C. (1984). Sparing of
semantic memory in Alzheimer’s disease. Journal of

Abnormal Psychology, 2;, 321-330.

Ober, B.A., & Shenaut, G.K. (1988). Lexical decision and
priming in Alzheimer's disease. Neuropsychologia, ee,
273-286.

Ober, B.A., Dronkers, N.F., Koss, E., Delis, D.C., &
Friedland, R.P., (1986). Retrieval from semantic memory
in Alzheimer’s-type dementia. Journal of Clinical and
Experimental Neuropsychology, 8, 75-92.

Ostergaard, A.L. (1987). Episodic, semantic and procedural
memory in a case of amnesia at an early age.

Neuropsychologia, 2;, 341-357.

Perret, E. (1974). The left frontal lobe of man and the
suppression of habitual responses in verbal categorical
behaviour. Neuropsychologia, 1;, 323-330.

Poon, W. (1985). Differences in human memory with aging:
Nature, causes, and clinical implication. In J.E.
Birren & K.W. Schaie (Eds.), Handbook of the psychology
of aging (2nd ed., pp. 427-462). New York: Van
Nostrand Reinhold.

Pribram, K.H. and McGuinness, D. (1975). Arousal,
activation, and effort in the control of attention.

Psychological Review, ee, 116-149.

59

Ramier, A., & Hecaen, H. (1970). Role respectif des
atteintes frontales et de la lateralisation lesionnelle
dans les deficits de la "fluence verbale." Revue

Neurologigue, 123, 12-22.

Ratcliff, R. (1986). More on the distinction between
episodic and semantic memories. Journal of experimental
Psychology: Learning. Memory. and Cognition, 22, 295-
306.

Rosen, W.G., Terry, R.D., Fuld, P.A., Katzman, R., & Peck,
A. (1980). Pathological verification of ischemic
score in differentiation of dementias. Annals of

Neurology, 7, 486-488.

Ruff, R.M., Light, R.H., & Evans, R.W. (1987). The Ruff
Figural Fluency Test: a normative study with adults.

Developmental Neuropsychology, 2, 37-51.

Ryle, G. (1949). The concept of mind. San Francisco:
Hutchinson.

Santo Pietro, M.J. & Rigrodsky, S. (1986). Patterns of
oral-verbal perseveration in adults. Brain and

Language, 22, 1-17.

Schacter, D.L. (1983). Amnesia observed: Remembering and
forgetting in a natural environment. Journal of
Abnormal Psychology, 22, 236-242.

Schneck, M.R., Reisberg, B., & Ferris, S.H. (1982). An
overview of current concepts of Alzheimer’s disease.
American Journal of Psychiatry, 139, 165-173.

Schwartz, M.F., Marin, O.S.M., & Saffran, E.M., (1979).
Dissociations of language function in dementia: A case
study. Brain and Lan ua e, 1, 277-306.

Smith, S.R., Murdoch, B.E., & Chenery, H.J. (1989).
Semantic abilities in dementia of the Alzheimer type: I.
Lexical semantics. Brain and Language, 36, 314-324.

Squire, L.R., Cohen, N.J., & Zouzounis, J.A. (1984).
Preserved memory in retrograde amnesia: Sparing of a
recently acquired skill. Neuropsychologia, 22,
145-152.

Tucker, D.M. & Williamson, P.A. (1984). Asymmetric neural
control systems in human self-regulation. Psychological
Review, 22, 185-215.

Tulving, E. (1972). Episodic and semantic memory. In E.
Tulving, & W. Donaldson (Eds.), Organization of

60
memory (pp.381-403). New York: Academic Press.

Tulving, E. (1983). Elements of semantic memory. Oxford:
Clarendon Press.

Tulving, E. (1987). Multiple memory systems and
consciousness. Human Neurobiology, e, 67-70.

Tulving, E., Schacter, D.L., McLachlan, D., & Moscovitch, M.
(1988). Priming of semantic autobiographical knowledge:
A case study of retrograde amnesia. Brain and
Cognition, 2, 3-20.

Veroff, A.E. (1980). The neuropsychology of aging:
Quantitative analysis of visual reproductions.
Psychology Research, 22, 259-268.

Warrington, E.K. (1975). The selective impairment of
semantic memory. Quarterly Journal of Experimental

Psychiatpy, 21, 635-657.

Wells, C.E. (1984). Diagnosis of dementia: A reassessment.
Psychosomatics, 22, 183-190.

Wilson, R.S., Kaszniak, A.W., & Fox, J.H. (1981). Remote
memory in senile dementia. Cortex, 21, 41-48.

Winograd, T. (1975). Frame representations and the
declarative-procedural controversy. In D. Bobrow, & A.
Collins (Eds.), Representation and understanding
(pp.185-210). New York: Academic Press.

Wood, F., Ebert, V., & Kinsbourne, M. (1982) The
episodic-semantic memory distinction in amnesia and
memory: Clinical and experimental observations. In L.

Cermak (Ed.), Human memory and amnesia (pp. 167-194).
Hillsdale, NJ: Erlbaum.

IGQN STATE UNIV

lllllllllll Ill llll lllllllllllllllllllillllll