Wechsler Adult Intelligence Scale - Wechsler Memory Scale Differenc

Western Michigan University
ScholarWorks at WMU
Dissertations Graduate College
8-1984
Wechsler Adult Intelligence Scale/Wechsler Memory Scale

Difference Scores: Their Relationship to Brain Dysfunction and
Closed Head Injury
Don A. Boyd
Follow this and additional works at: https://scholarworks.wmich.edu/dissertations
Part of the Psychoanalysis and Psychotherapy Commons
Recommended Citation
Boyd, Don A., "Wechsler Adult Intelligence Scale/Wechsler Memory Scale Difference Scores: Their
Relationship to Brain Dysfunction and Closed Head Injury" (1984). Dissertations. 2387.
https://scholarworks.wmich.edu/dissertations/2387
This Dissertation-Open Access is brought to you for free

and open access by the Graduate College at
ScholarWorks at WMU. It has been accepted for inclusion
in Dissertations by an authorized administrator of
ScholarWorks at WMU. For more information, please
contact [email protected].
WECHSLER ADULT INTELLIGENCE SCALE/WECHSLER MEMORY SCALE
DIFFERENCE SCORES: THEIR RELATIONSHIP TO BRAIN
DYSFUNCTION AND CLOSED HEAD INJURY
by
Don A. Boyd
A Dissertation
Submitted to the
Faculty of The Graduate College
in partial fulfillment of the
requirements for the
Degree of Doctor of Education
Department of Counseling and Personnel

Kalamazoo, Michigan
August 1984
R eproduced with permission of the copyright owner. Further reproduction prohibited without permission.
WECHSLER ADULT INTELLIGENCE SCALE/WECHSLER MEMORY SCALE
DIFFERENCE SCORES: THEIR RELATIONSHIP TO BRAIN
DYSFUNCTION AND CLOSED HEAD INJURY
Don A. Boyd, Ed.D.
Western Michigan University, 1984
Performance differences between the Wechlser Adult Intelligence
Scale (WAIS) and the Wechsler Memory Scale (WMS) were studied. Dif
ferences in performance between the WAIS Full Scale IQ Score and the
WMS Memory Quotient (WMS discrepancy score) were compared across
three groups consisting of a closed head injury group (N = 45), a
localized lesion group (N^ = 25), and a psychiatric group (N = 45)
who were unimpaired on the Halstead-Reitan Neuropsychological Bat
tery. Also, WMS discrepancy scores were compared across a long-coma
group and a short-coma group and the correlation coefficient between
the WMS discrepancy score and neuropsychological impairment rating
was obtained.
The WMS discrepancy score was hypothesized to be a marker of
diffuse closed head injury; and it was predicted that WMS discrep
ancy scores would be greater in the closed head injury group, that
the long-coma patients would show greater WMS discrepancy scores
than a short-coma group, and that WMS discrepancy scores would be
related to severity as defined by impairment on the Halstead-Reitan
Neuropsychological Battery. Results confirmed the expected differ
ences between groups. The closed head injury group was significantly
different from both the localized lesion group and the unimpaired
psychiatric group on the dimension of WMS discrepancy score. The
long-coma group had significantly larger WMS discrepancy scores.
The correlation between WMS discrepancy score and neuropsychological
impairment rating was only weakly supported. The closed head injury
group showed greater WMS discrepancy scores despite the fact that
they showed the lowest WAIS scores. Results were interpreted as
being supportive of the idea that the WMS discrepancy score may be a
marker of diffuse closed head injury.
Implications for the use of the WMS discrepancy score and
neuropsychological ratings were discussed. The WMS discrepancy
score may be more descriptive of severity in closed head injury than
neuropsychological impairment ratings. Despite weaknesses in the
use of the WMS as a comprehensive test of memory, it may be of clini
cal value when used with other tests to highlight specific diffi
culties in the area of fluid verbal skills or information processing.
The systematic inclusion of the WMS discrepancy score in the evalua
tion of closed head injuries seems warranted. Recommendations for
further research were offered.
INFORMATION TO USERS
T his re p ro d u c tio n was made fro m a copy o f a d o c u m e n t sent to us fo r m ic ro film in g .

W hile the m ost advanced technology has been used to p ho tograph and reproduce
this d o c u m e n t, the q u a lity o f the re p ro d uctio n is heavily dependent upon the
q u a lity o f the m aterial su b m itted .
T h e fo llo w in g e xp lan atio n o f techniques is provided to help c la rify m arkings o r

n otatio n s w hich m ay appear 011 this rep ro d u ctio n .
1 .T h e sign or “ targ et” fo r pages apparently lacking from the docum ent
photographed is “ Missing Page(s)” . I f it was possible to o btain the missing
page(s) o r section, they are spliced in to the film along w ith adjacent pages. This
m ay have necessitated c u ttin g through an image and d up licatin g adjacent pages
to assure com plete c o n tin u ity .
2. W hen an image on the film is o b lite ra te d w ith a ro u nd black m a rk , it is an

in d ic a tio n o f e ith e r b lu rred copy because o f m o v e m e n t during exposure,
d up licate co p y, o r co p yrig h ted m aterials th a t should n o t have been film e d . F o r
b lu rred pages, a good image o f the page can be fo u n d in the adjacent fram e. I f
co pyrig h ted m aterials w ere d eleted, a target n ote w ill appear listing the pages in
the adjacent fram e.
3. W hen a m ap, draw ing o r c h a rt, e tc ., is p a rt o f the m a te ria l being pho to g rap h ed,
a d e fin ite m etho d o f “ sectioning” the m aterial has been fo llo w e d . It is
custom ary to begin film in g at the upper le ft hand corner o f a large sheet and to
co ntin u e from le ft to right in equal sections w ith small overlaps. I f necessary,
sectioning is co ntinued a g a in -b e g in n in g b elo w the first ro w and co ntin u in g on
u n til com plete.
4. For illustrations th a t c an no t be satisfactorily reproduced by xerographic

means, photographic p rin ts can be purchased at a d d itio n a l cost and inserted
in to y o u r xerographic co p y. These prints are available upon request fro m the
D issertations C ustom er Services D e p a rtm e n t.
5. Some pages in any d o c u m e n t may have in d is tin c t p rin t. In all cases the best
available copy has been film e d .
University/
Microfilms
International
300 N. Zeeb Road
Ann Arbor, Ml 4 8106
R eproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8425228
B o yd , Don Allen
WECHSLER ADULT INTELLIGENCE SC ALE/W ECHSLER MEMORY SCALE

DIFFERENCE SCORES: THEIR RELATIONSHIP TO BRAIN DYSFUNCTION
AND CLOSED HEAD INJURY
Western Michigan University Ed.D. 1984
University
Microfilms
International 3 0 0 N. Z e e b R oad, Ann Arbor, Ml 48106
Copyright 1984
by
Boyd, Don Allen
All Rights Reserved
R eproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
Copyright by
Don A. Boyd
1984
ACKNOWLEDGMENTS
I would like to acknowledge the support and effort of my
chairman, Dr. Robert Hopkins, in helping complete this important
but urgent project. Also I wish to thank Dr. Uldis Smidchens and
Dr. Malcolm Robertson for giving time when it was most difficult.
I wish to express appreciation to Dr. Richard Gay, friend and
mentor, who was most helpful in providing direction and creating a
clinical foundation.
I want to thank my parents, Francis and Alice Boyd, for their
lifelong ability to understand the importance of things that are
fun and interesting.
I wish to thank Becky, Joe, and Karen, who are still too young
to understand how much more meaning they give to the completion of
this project.
Thanks to my extraordinary wife, Deborah Jo. Her talent,
effort, and love were essential to the completion of the disserta
tion. Her character has contributed much to the happiness of
others around her.
Don A. Boyd
ii
TABLE OF CONTENTS
ACKNOWLEDGMENTS .......................................... ii
LIST OF T A BLES .............................................. v
CHAPTER
I. THE PROBLEM AND THE BACKGROUND..................... 1
The Problem.................................... 1
Review of Literature ........................... 4
Definition of Closed Head Injury .............. 5
Definition of Memory Processes ............... 7
Closed Head Injury and Memory Impairment . . . . 9
The WMS and Its Relationship to Brain

T r auma ...................................... 13
The WAIS and Its Relationship to Brain

Trauma ............................. . . . . . 25
WAIS and WMS Relationships................... 38
The Halstead-Reitan Neuropsychological

B a t t e r y .................................... 43
Summary of Literature Review ................. 52
General Research Hypothesis ..................... 56
Limitations of the S t u d y ....................... 57
II. DESIGN AND METHODOLOGY............................. 58
S u b j e c t s ...................................... 58
The Unimpaired Control Group ................. 58
The Closed Head Injury G r o u p ................. 60
The Localized Trauma Group ................... 61
Subject Selection and Group Assignment ........ 63
iii
Table of Contents— Continued
CHAPTER
Criteria Instruments ........................... 66
Research Design ................................. 69
Data Collection Procedure ................... 69
Design of the S t u d y ......................... 70
Research Hypotheses (Operational Hypotheses) ... 72
III. DATA A N A L Y S I S ..................................... Ik
Analysis of Population D a t a ..................... Ik

Additional Statistical Analysis ............. , . 78
D i s c u s s i o n ..................................... 83
IV. FINDINGS, CONCLUSIONS, AND RECOMMENDATIONS .......... 89
Summary........................................ 89
F i n d i n g s ...................................... 91
Conclusions.................................... 93
Recommendations ................................. 100
APPENDIX.................................................. 108
BIBLIOGRAPHY .............................................. Ill
iv
LIST OF TABLES
1. Comparison of Closed Head Injury Group, Localized

Lesion Group, and Psychiatric Unimpaired Group by
Sex, Mean Age, and Mean Neuropsychology Impairment
R a ting................. 63
2. Means, Ranges, and Standard Deviations for the WAIS,

WMS, Halstead-Reitan Neuropsychological Impairment
Rating, and WMS Discrepancy for the Entire Research
Population............................................ 75
3. ANOVA for Three Ranges of WMS Discrepancy Scores

(Closed Head Injury Group, Localized Lesion Group,
and Psychiatric UnimpairedGroup) ...................... 77
4. ANOVA for Two Ranges of WMS Discrepancy Scores by

Length of C o m a ........................................ 77
5. ANOVA of WAIS Full Scale Scores for Closed Head

Injury Group, Localized Lesion Group, and a
Psychiatric Unimpaired Group ............................ 79
6. ANOVA of WMS Quotients for Closed Head Injury Group,

Localized Lesion Group, and a Psychiatric Unimpaired
G r o u p ................................................ 80
7. ANOVA of WAIS Full Scale Scores for Long and Short

Coma G r o u p s .......................................... 81
8. ANOVA of WMS Quotients for Long and Short Coma

G r o u p s................................................ 82
CHAPTER I
THE PROBLEM AND THE BACKGROUND
The Problem
This research was initiated to gain further knowledge regarding
the cooperative utilization of the Wechsler Memory Scale and the
Wechsler Adult Intelligence Scale. More specifically, it is con
cerned with the degree of difference between the Wechsler Adult
Intelligence Scale (WAIS) full scale intelligence quotient and the
Wechsler Memory Scale (WMS) memory quotient. For the remainder of
the present study this difference score will be more conveniently
referred to as the "WMS discrepancy score."
This WMS discrepancy score, when it does exist, is thought to
be important due to the fact that Wechsler (1945) constructed the
memory scale to be equivalent to the WAIS full scale intelligence
quotient score. Therefore, a Wechsler memory quotient significantly
below the WAIS intelligence quotient should logically predict diffi
culties in information processing or memory as measured by the WMS.
The validity of the WMS as a screening device for memory dys
function has been disputed however (Prigatano, 1978). He suggested
that the WMS was not sufficient as a test of memory owing to its
emphasis on verbal and language types of memory and its lack of
capability to measure functions other than short-term memory pro
cesses. Prigatano (1977) also suggested that it is a poor screening
test for head injury.

1
2
Despite its published shortcomings as a memory test, this re
searcher has observed deficits in WMS performance in populations of
closed head injury patients. In an outpatient rehabilitation set
ting, a number of closed head injury patients were examined on an
outpatient basis who were 2 years or more posttrauma. They seemed
to perform poorly on the WMS in relation to the WAIS; a fact that
appeared to be confirmed by the subjective impression that the
rehabilitation of these individuals was often most hindered by an
inability to show the capabilities for short-term memory and reten
tion. Such subjective opinions of memory dysfunction have been
formerly documented (Benton, 1979; Black, 1973). This study
attempts to show that a comparison of the WMS discrepancy score
between closed head injury and nonclosed head injury patients might
help to determine if it could be used as a marker or indicator of
specific types of short-term memory deficits; particularly those
short-term memory processes that may be peculiar to thespecific
diffuse processes of closed head injury.
Neuropsychological batteries are constructed to be sensitive to
the type and severity of head injury. However, these neuro
psychological batteries are not satisfactory as brief screening
devices for closed head injuries and they do not afford a comparison
between memory and nonmemory processes. The WMS has beenused in
conjunction with the Halstead-Reitan Neuropsychological Battery for
years and it has recently been added to the Michigan Neuro
psychological Battery (Smith, 1983). Therefore, with the possibil
ity that it may show different sensitivity to different clinical
syndromes, the use of the WMS in neuropsychological batteries needs
to be further defined. This would be particularly true in respect
to the meaning of the WMS discrepancy score. Logically, the
Wechsler memory quotient and WAIS full scale intelligence quotient
should show similar sensitivity to those types of brain dysfunction
where memory disturbance is not an expected complaint. Conversely,
in brain trauma where the damage is diffuse, there should be greater
memory impairment of the type measured by the WMS discrepancy score.
Also, if the WMS score is truly sensitive to short-term memory dis
ruption following closed head injury, then the more severe closed
head injuries should be associated with greater WMS discrepancy
scores. Also, since severity of head injury is measured by impair
ment ratings on neuropsychological tests, then the WMS discrepancy
score should increase along with those severity ratings among the
closed bead injured. Lastly, it is suggested that the closed head
injury patients will show greater WMS discrepancy scores than will
patients with localized lesions or psychiatric patients who are un
impaired.
This study then proposes to examine the WMS discrepancy score
in relation to other severity indicators and also to carry out a
comparison of the WMS discrepancy score with brain damaged and non
brain-damaged populations. Such comparisons might greatly extend
the usefulness of the WMS as it is coming to be used as an addi
tional neuropsychological technique. Due to its cited difficulties
in validity as a memory task or an individual screening technique
(Prigatano, 1977), its use within the context of other assessment
measures might prove more useful. Additionally, the early detection
and assessment of the information processing and memory deficits of
the closed head injured patient might be more meaningful and more
simply assessed by a comparison of WMS and WAIS scores than with
other phases of neuropsychological assessment. Thus, this investi
gation may help in the efficient use of measurement techniques
already present in neuropsychological batteries. The ultimate de
tection and rehabilitation of cognitive deficits in the closed head
injury patient requires this type of specific description at the
earliest diagnostic opportunity.
In summary, the problem addressed by this research involves a
greater understanding of the possible short-term memory or informa
tion processing deficits in closed head injury patients by the use
of a WMS and WAIS comparison score called the WMS discrepancy score.
It is attempted in this study to elucidate the meaning or importance
of the WMS discrepancy score by correlative or comparative tech
niques and that the specific diagnostic capability of these deficits
in instances of closed head injury could be carried out through
already existing neuropsychological techniques which utilizes the
WAIS in combination with the WMS.
Review of Literature
The review of the literature will be organized in the following
manner: (a) definition and explanation of the diagnosis closed head
injury to clarify this trauma as a type of injury process; (b) the
definition and explanation of memory process; (c) a review of the
5
literature with reference to the impact of head injury upon memory
processes; (d) the WMS will be described and a review will be pre
sented of its relationship to brain trauma; (e) the WAIS will be
reviewed along with its relationship to brain trauma; (f) the WAIS
and WMS relationships will be described with emphasis upon their
differential sensitivity to brain impairment; and (g) the Halstead-
Reitan Neuropsychological Battery will be described with an emphasis
upon its sensitivity to brain dysfunction. A brief summary of the
literature which attempts to highlight the major findings will be
provided at the end of these reviews.
Definition of Closed Head Injury
Closed head injury results from the impact of a blunt object
striking the head. It is differentiated from a penetrating or
missile wound trauma by the absence of penetration of the brain case.
In penetrating head wounds the penetrating object itself may destroy
brain tissue and constitute an immediate cause of localized brain
lesion. Beyond the superficial fact that closed head Injuries do
not involve penetration and localized disruption of brain tissue,
closed head injuries involve a set of occurrences which qualify it
as a type of trauma quite distinct from other brain lesion producing
processes.
Ommaya and Gennarelli (1974) have explicated the actual physi
cal process of damage. The brain, as a semigelatinous mass, is seen
to distort and move within the brain case under conditions of rota
tion, sudden acceleration, or sudden deceleration. These movements,
6
following blows to the head, distort the physical shape of the brain
particularly at the frontal and temporal poles. Shearing forces
disrupt the longer axon processes within the brain causing micro
scopic, rather than mass, lesions. Additionally, the force of the
brain being pressed against the surfaces of the brain case may pro
duce "bruising." Ommaya and Gennarelli (1974) attributed traumatic
unconsciousness also to shear strain which extends to lower brain
centers causing a "disconnection of the alerting system of the
brain."
Levin, Benton, and Grossman (1982) have cataloged additional
features. They described primary traumatic effects which are simi
lar to those just described. These include both contusions and
microscopic shearing and stretching of nerve fibers across communi
cating brain areas. They also cite the presence of intracranial
hemorrhage. Raised intracranial pressure due to swelling of damaged
tissues and the presence of mass lesions due to subdural bleeds and
hematomas are additional effects.
Closed head injury by definition is fundamentally, therefore, a
diffuse brain injury in which long axon processes connecting lower
brain centers with the cortex, and long axons within the cortex, are
disrupted. It is important for the purposes of the present study to
understand closed head injury, therefore, as a mostly nonlocalized
phenomena which may have superimposed localized affects due to
hematomas or hemorrhage. The major aspect of acceleration and de
celeration of the brain within the skull and shearing forces are its
major discriminating features. These processes differentiate it
from a more local or focalized process resulting from cerebral
vascular accidents, penetrating head wounds, tumors, or other patho
logical process.
Sequelae of closed head injury have been reviewed by Benton
(1979). He noted the following symptom picture: (a) impairment
of concentration, (b) fatigability, (c) disturbances in memory,
(d) emotional instability and lowered tolerance for frustration,
(e) personality alteration including either depression or dis-
inhibition and euphoria, (f) some aphasic deficits, and (g) mixed,
inconsistent sensory deficits. For the entire range of closed head
injuries it would appear that the first three factors are the most
prevalent and serious in the report of closed head injuries and that
these symptom complexes, which appear to disrupt a client's momen
tary capacity to attend to stimuli in the environment and to memorize
them, create long standing disability. An estimated 40% suffering
these disabilities fail to make a satisfactory long-term social and
economic adjustment following those injuries (Benton, 1979). The
importance of developing an assessment technique to alert clinicians
to the presence of disabling deficits in short-term memory is quite
obvious given this prevalence.
Definition of Memory Processes
Memory is a very broad concept that encompasses a number of
different processes and functions. E. W. Russell (1981) has re
viewed the experimental literature and described types of memory
8
functioning which have applications to clinical neuropsychological
use.
E. W. Russell's (1981) description of memory function includes
a sensory store which is tied to immediate sensory detection capac
ity and which fades extremely quickly. Of greater importance is
short-term memory also known as "immediate memory." This memory
function lasts for 20 to 40 seconds or slightly more. It is deter
mined by specific capacity and is usually related to about seven
items; plus or minus two. Short-term memory is considered a "work
ing memory" in that it is strongly associated with ongoing conscious
ness and the momentary organization of awareness. Therefore, short
term memory is intimately associated with the ongoing process of
mental organization as well as being an indicator of momentary
"storage." Thirdly, long-term memory has three recognized stages
consisting of consolidation, storage, and retrieval. The consolida
tion stage is said to relate to the period when short-term memory is
being transferred to long-term memory storage. During this stage of
consolidation, long-term memory processes are particularly unstable
and this unstable period may last for minutes to hours. The word
recent memory is often used to describe that period of transfer
which is longer than actual short-term memory but it is not a true
long-term memory phenomenon. Long-term memory therefore extends
from the recent memory of a few minutes, to years, depending upon
the completeness and the organization of the memory trace.
Future discussions of memory in this paper will be primarily
concerned with memory under the above descriptions of short-term
9
memory or long-term with recent memory being specifically designated
as an intermediate and transitional term between the two.
Closed Head Injury and Memory Impairment
Despite the broad range of memory impairments associated with
closed head injuries, the actual pattern of recovery of memory func
tions following closed head injury and the long-term effects are
less clearly agreed upon. Conkey (1938) compared closed head injury
and control patients on a variety of cognitive tasks starting at
2 weeks post-injury and extending to 50 weeks post-injury. By the
time of the fourth test, the performance of the head injured group
approximated that of the control group. Conkey claimed, however, in
this group that memory performance fell behind the recovery of other
intellectual functions, particularly verbal functions. Brooks (1972)
tested 27 closed head injury patients on many tests of short-term
memory and, when compared with a normal control group, documented
difficulties on all measures used. He used a test very similar to
the verbal learning and visual reproduction subtest of the WMS and
interpreted his results to indicate that the head injury patients
showed severe memory deficits by acquiring less information ini
tially and also showed increased forgetting through the process of
interference. In a later study, Brooks (1975) studied 30 closed
head injury patients whom he classified as "very severe" with most
having coma periods of 24 hours or more. He interpreted those re
sults to suggest that long-term memory is more impaired than short
term memory in closed head injury and that it is related to overall
10
severity as measured by the length of coma. It should be noted,
however, that Brooks utilized a simple digit span subtest as one
measure of short-term memory which does not require organization of
the stimulus material and requires only very brief retention periods.
An inability to organize novel stimuli may contribute to diffi
culties in recent memory as suggested by Brooks. Impairment of
memory functioning in 27 closed head injury patients in a task of
extremely short recognition memory (10 to 20 seconds) was described
by Levin, Grossman, and Kelly (1976). Persistent impairment on this
type of task was uncovered in this group several months or more
after trauma. They noted that many of the patients tested more than
a year following injury were among the most impaired, suggesting
rather permanent deficits. Since the present study is concerned
with the long-term residuals in memory functioning, this nicely
corroborates the presence of the problem with the results of the
early study by Conkey (1938) cited earlier, which showed some memory
impairments to persist even after recovery of many cognitive func
tions .
Brooks (1975) specifically attempted to objectify whether short
term memory or long-term memory processes are most affected ii closed
head injury patients. The 30 patients of his study were alert and
well recovered from any recent posttraumatic or acute memory diffi
culties. The presentation of high frequency words and lists wich
recall periods ranging from zero seconds to 20 seconds was used in
addition to digit span. Brooks noted that head injured patients
were near normal in terms of digit span subtests but were
11
significantly poorer on even very short delayed recall. He inter
preted these results to suggest a "long-term" memory deficit. Again
these results could be an artifact of the overall complexity of the
memory task, with digit span subtests representing the most simple
and immediate form of short-term memory.
In another study attempting to find the nature of head injury
impairments, Silverstein, Rosenbaum, and Rennick (1979) attempted to
determine whether memory impairments were the results of decay in
unstable memory traces or whether brain damaged individuals were
more subject to greater interference which erased previous memories.
Their sample included a mixed group of head injuries, including
closed head injuries as well as neurosurgical patients. The results
indicated that decay, and impairment of consolidation at the level
of immediate memory, was greatest for this group. This implies
deficits in the area of short-term memory and favors the hypothesis
of rapid decay.
The previous studies suggest that while there is agreement
regarding the presence of memory impairment in head injury patients,
speculation about the type of memory impaired varies from the short
term memory processes to more long-term memory processes. The
actual incidence of memory dysfunction following closed head injury
was probably first investigated by W. R. Russell (1932). He cal
culated that about 36% of the cases following closed head injury
develop some type residual memory problem. However, methodological
problems with the study were present in that the degree of memory
impairment was not reported and the method arriving at "severity"
12
was not reported. More recently, Lidvall, Linderoth, and Norlin
(cited in Schacter & Crovitz, 1977) depended upon the patient's sub
jective report to determine the incidence of memory impairment and
found reported memory difficulties in about 16% of closed head in
jury cases they reviewed. These cases range from only 2 to 90 days
posttrauma, however, suggesting that much of their sample was biased
in the direction of very mild postconcussive syndromes. Both of the
above studies lack comparative psychometric data or any controls
upon the severity of head injury, such as length of coma.
In a very comprehensive review of the literature regarding
closed head injuries, Schacter and Crovitz (1977) concluded that
most studies of closed head injury substantiate correlations between
posttraumatic amnesia, which is the period of absence of continuous
awareness following injury, and length of coma to a broad range of
memory difficulties. However, they noted that the literature con
cerning the recovery process ranges from studies that suggest no
remaining memory impairment 2 years after closed head injury to
those who were able to find correlations between coma and memory
deficits as much as 5 years after injury. There is a general tend
ency for a variety of memory effects to be found in head injuries,
especially with longer periods of coma or posttraumatic amnesia.
Differences in methods of measurement as well as rather important
differences in agreement about what is considered long or short
periods of coma prevent meaningful consistencies in conclusions
drawn about the overall impact of closed head injury upon later
memory functioning. Early as well as later studies, as we have seen,
however, suggest that memory impairment is a remaining deficit of
considerable importance.
A major lack of specific comparisons between types of head
trauma is conspicuous in the early literature. The next section
which more specifically reviews the use of the WMS to measure head
injury effects allows examination of more consistent results of a
single psychometric instrument.
The WMS and Its Relationship to Brain Trauma
An explanation of the WMS, its content, and its psychometric
structure is in order previous to a review of its sensitivity to
brain damage effects.
Wechsler (1945) desired to develop a "rapid, simple memory
examination.11 Since he desired that the memory quotient scoresbe
made comparable and equivalent to the then popular Wechsler-Bellevue
Intelligence Scale, he calculated correction scores for an age group
that could be added to the total score of the WMS so that the
patient's memory quotient would be equal to his or her full scale
intelligence quotient score. In this way he had hoped that memory
deficits relative to the patient's overall cognitive functioning
could be estimated.
He included seven subtests. Subtests 1 and 2 are concerned
with personal and current Information as well as orientation. These
subtests essentially determine whether the patient is communicative
or oriented to surroundings sufficiently to take the test. A third
section, mental control, requires the patient to rapidly execute
overlearned material, such as reciting the alphabet, counting by 3s,
and doing other simple tasks under the pressure of time. Logical
memory is the fourth section and requires the person to recall a
number of items presented in two short-story passages which are read
to the patient. The fifth section which is memory span is identical
to digit span on WA1S. The sixth section which is visual reproduc
tion requires the person to draw from memory geometric designs after
he or she has been shown them for 10 seconds. The seventh section
is associate learning which requires a patient to listen to a number
of paired words and then to recall the accompanying word when the
word list is read. The patient is allowed three trials, and the
words are divided into easy words which are logically associated and
hard words which have no logical association.
The factor structure of the memory scale was investigated by
Davis and Swenson (1970). The factor analytic study produced two
factors which they labeled long- and short-term storage and the
third factor called "mental control" which they claimed as having
much in common with the "freedom-from-distractability factor" or the
ability to maintain concentration under more complex stimulus demand
conditions. Kear-Colwell (1977) did a factor analysis of WMS re
sults in 112 patients referred for investigation of possible organic
pathology of the brain. While the pathology type was not specified,
a similar factor structure emerged. This consisted of (a) the re
call of immediate novel information to both visual and auditory
modalities, (b) attention concentration and rapid processing of
verbal information, and (c) orientation to place and time and the
15
recall of simple long established verbal information. Their idea of
emphasizing a novel learning or "short-term memory" function along
with attention and concentration capability is very similar to the
earlier study of Davis and Swenson (1970). Still another factor
analysis of the test carried out by Dujovne and Levy (1971) compared
the factor analytic results of normals to a patient group of mixed
psychiatric diagnosis. They extracted three factors for normals
which they called (1) general attentiveness, (2) simple learning,
and (3) associational flexibility. For patients the relevant dimen
sions were named (1) mental control, (2) associational flexibility,
and (3) cognitive dysfunction. In the patient group, associative
flexibility and cognitive dysfunction accounted for the majority of
the common variance. The factor of associative flexibility appeared
to load on associative learning with the cognitive dysfunction factor
loading on visual reproduction, logical memory, and the verbal asso
ciates portions of the test. The overall results were interpreted
to indicate that, as a whole, the WMS may be sensitive to dysfunc
tions in "associative flexibility" such that the capacity for analy
sis and synthesis of complex stimuli and a rather immediate reten
tion of this stimuli was impaired.
This would suggest that the WMS is also sensitive to variables
which serve as a platform for short-term memory processes. This
would seem to suggest that in a normal sample the WMS was sensitive
to a factor of "general retentiveness" that is closely related to
verbal intelligence. Therefore, the WMS may be detecting cognitive
abnormalities related to concreteness and an inability to rapidly
16
synthesize information.
Dye (1982) did a factor analysis of the WMS in results obtained
from patients over 50 years of age. She reported that the factor
structure in this age group was more similar to the patient than the
nonpatient population of previous studies. That is, a short-term
memory factor of retention and a second factor of attention and con
centration that governs rapid organization and input appear to
determine the performance in this group as well as in the patient
population. These factor analytic studies appear to portray the WMS
as an instrument that may not be sensitive to pure memory function,
but rather cognitive processes which determine the rapid and flex
ible acquisition of stimuli.
The more recent validation study by Kear-Colwell and Heller
(1978) replicated the presence of the factors of learning and imme
diate recall of information and attention and concentration in a
population of normal subjects divided into a young and an old age
group. The overall results obtained were construed to suggest that
these two factors were indeed a measure of short-term memory. In
the most recent large scale review of most of the literature con
cerning the WMS, Prigatano (1978) summed up the positive features of
the test as including a relatively constant factor structure in the
areas of attention and concentration to novel information and gen
eral retentiveness in the verbal mode. He noted that it has limita
tions, however, in that it appears to be a poor measure of nonverbal
memory and fails to adequately address a more full range of memory
functions, such as recent or long-term memory.
17
An emergent Impression of the WMS based upon the review cited
thus far suggests that it is too "narrow" to serve as an actual
battery of memory testing. However, its sensitivity to operations
requiring flexibility without distractability and its involvement
with a general factor of retentiveness would suggest that it has
merit in measuring dysfunction in brain trauma syndromes where these
types of learning deficits are considered to be key events.
These learning deficits will be shown to be important to the
diffuse head injury processes, which are theoretically related to
closed head injury, at another point in this study. However, a re
view of the relationship of the WMS to more specific processes of
brain trauma is warranted.
The literature relating the WMS to the effects of brain trauma
and closed head injury has shown a great deal of growth within the
last 10 years. The WMS (Wechsler, 1945) has been in existence for
over 30 years and in its most recent history has been probably the
most frequently used instrument for investigating memory deficits in
head trauma (Prigatano, 1978). Since it continues to be widely used
as a neuropsychological technique, a review of the relationships
between this test and head trauma will be carried out here.
In one of the earliest comparison studies, Cohen (1950) com
pared patients with psychoneurotic difficulties, "organic pathology,"
and schizophrenics. The organic group was divided into cases of
encephalitis, brain tumor, and posttraumatic syndromes. His study
lacks completeness in definition of cerebral pathology. The groups
were compared in terms of their performance on the WMS, but in
18
addition, It Included the Wechsler-Bellevue full scale Intelligence
quotient and memory quotient discrepancy score, which is the topic
of this study. He found that the memory functioning of these three
groups could not be discriminated in terms of WMS subtest scores or
deviation from Wechsler-Bellevue intelligence quotient scores. It
should be noted that there was no description of the "traumatic"
cases. There was no attempt to discriminate closed head injury from
other types. Another similar early study by Howard (1950) compared
epileptics, various encephalopathies, and 35 paretics. He was not
able to distinguish any of these groups from the control group with
the exception of some of the paretics. Again, this study lacked a
clear definition of cerebral pathology and appeared to exclude
closed head injury as a syndrome.
Bachrach and Mintz (1974) compared two groups of neuro
psychiatric patients. Group 1 consisted of 42 patients with a clear
diagnosis of cerebral dysfunction. These were compared to psychiat
ric patients with no such diagnosis. They stated that four subtests
from the WMS (information, logical memory, designs, and associative
learning) significantly discriminated between those impaired and
unimparied patients. These authors did not specify the type of dys
function that they thought the WMS might be sensitive to. Rather,
they hoped it to be a more general tool for the prediction of "mild
cerebral dysfunction." This claim is contested by Prigatano (1977).
He compared 31 adult patients with confirmed brain dysfunction to 26
adult patients without similar dysfunction. He noted that these
patients generally performed poorly on subtests of the WMS, but
19
these differences drastically decreased when intelligence quotient
levels were matched. This suggested that the WMS was a poor screen
ing device for brain dysfunction and that consistent with its pre
viously established norms (Wechsler, 1945) it was not discrepant or
additive to cognitive functions as measured by the Wechsler Adult
Intelligence Scale (WAIS).
Although the WMS is described in the above studies as being
insensitive to undefined cerebral dysfunction, its failure may not
necessarily reflect uselessness due to the criticisms inherent in
patient selection processes. There is no reason to believe that all
organic disorders of the central nervous system will necessarily
cause memory dysfunction in the same manner. A unitary concept of
brain damage may not be adequate to understand the relationships
between memory function and organic brain disorders.
In a factor analytic study in which factor scores were calcu
lated for patients with different types of organic pathologies,
Kear-Colwell (1973) investigated the performance of 250 patients on
the WAIS and WMS. They located three factors. The first loaded
highly on complex short-term memory. The second appeared to consist
of mental control and seemed to involve processes of immediate atten
tion, concentration, and the ability to process verbal nonsemantic
information. Factor 3 appeared to load primarily on very long-term
memory processes. These factors were continuous across groups with
the first factor correlating highly with intelligence. The group
with mixed head injuries scored significantly different on Factors
1 and 2 with major deficits appearing on Factor 1. These figures
20
suggest that compared with normal subjects of similar age and
intellectual ability, the head injury subjects evidence memory dys
function on two to three factors described in this study. Their
research also appeared to support the conclusion that while some
localized and focal lesions do not produce measurable memory defect,
diffuse head injury manifestations create greater memory disturbance
as measured by the WMS.
In one of thefirst studies to closely control the type of
cerebral pathology, Brooks (1976) examined 82 patients with "severe"
head injury (posttraumatic amnesia lasting at least 2 days). He
found rather conclusive results in this study that these head in
jured patients had severe memory difficulties, particularly on
logical memory and associative learning. Posttraumatic amnesia and
length of coma were related to poor performance, as was increasing
age. He also noted, however, that more localizing neurological
signs, such as skull fracture or dysphasia, were not related to dif
ficulties in performance.
These results are interesting in the context of the present
study because they appear to relate diffuse processes to deficits in
WMS performance and suggest that focal signs, when present, are less
clearly predictors of poor memory function as defined by WMS per
formance. Secondly, this study marks very clear-cut results of
memory deficit in closed head injuries which are apparently remain
ing many months after injury. Also, Brooks (1976) suggested that on
a long-term presentation ( 1 hour), retention proved very difficult
for head injured patients. On logical memory, head injured patients
21
were significantly poorer at immediate and delayed recall although
their rate of forgetting was not significantly greater than controls.
This suggests that the acquisition of the stimuli affects both imme
diate retention as well as the longer term retention which follows.
In general, Brooks observed that patients with a length of coma of
1 week or less were much less affected on memory tasks than patients
with a comatose period lasting greater than 1 week. He also sug
gested that recovery of memory to a stable but low level may take
place possibly within the first A to 6 months after injury.
In another study of head injured patients, Kear-Colwell and
Heller (1980) performed a factor analysis study comparing head in
jury subjects with a control group from the general population.
This study isolated three factors which were persistently identified
in other factor analytic studies of the WMS and cited earlier. This
involved the learning and immediate recall (short-term memory) of
new semantic and complex information. It also included attention
and concentration or freedom from distractability with the third
factor being the overall orientation to time, place, and the recall
of simple long-established information. The differences between the
head injured group and the general population were very striking
with differences obtained at all three factors, particularly Factors
1 and 2. Although the head injury group produced decrements on all
aspects of performance on the WMS, it had a particularly marked
effect on tasks that required the immediate learning of new verbal
information, i.e., logical memory and associative learning. It was
concluded that the WMS has useful validity for describing these
22
types of cognitive deficits and in identifying these particular
types of cognitive dysfunctions in head injured patients.
A rather unique application of the WMS to the measurement of
brain damage was carried out by Kljajic (1975). He attempted to
compensate for the lack of scale scores on the WMS by comparing
"don't hold subtest" on the WMS (the associate learning subtests)
with the "hold subtests" (orientation, mental control, and digit
span). He was able to discriminate head injured from noninjured
groups based upon their performance on these two subtests, with the
brain injured group performing much more poorly on the "don't hold
subtest" in relation to the "hold subtest." This may point out the
susceptibility of the head injured groups to performance deficits on
the associative learning subtest. It also demonstrates the apparent
difficulty the head injured populations have in dealing with the
more complex subtests in terms of the types of material learned.
The head injured group correspondingly had a much easier time in the
performance of overlearned simple material which did not require the
rapid efficient organization of multiple verbal stimuli such as that
found in the associate learning subtest.
In another study which did control for the type of pathology,
Black (1973) compared WMS performance on groups who have suffered
both penetrating missile wounds and closed head injuries. As pre
dicted, the degree of impairment on three tests of memory was sig
nificantly greater in closed head injury, as compared to brain
damage secondary to penetrating missile wounds. Although there was
considerable variation in individual test performance, suggesting
23
that a "pure" type of head trauma is improbable, the group tendency
was very clear. These results suggest that discrete and localized
brain trauma does not result in as great a memory impairment as the
more diffuse types of head injury processes.
The effects of unilateral lesions on WMS performance was
studied by Bornstein (1982). He demonstrated that the WMS was
sensitive to the effects of lateralized lesions. As expected,
right lesion patients scored poorly on visual reproduction, while
left lesion patients scored poorly on logical memory and verbal
associate learning. The former finding corroborates the observation
of E. W. Russell (1975), who developed a scoring system to test
recent memory and found visual reproduction to function well as a
diagnostic indicator of right hemisphere lesion. Thus, although
diffuse head injury processes result in greater overall decrements
in memory processes, it would appear that some modality specific
memory deficits can be measured by the WMS if the lesions are spe
cific in regard to laterality. E. W. Russell (1982) used the same
revised scoring system of the WMS in a factor analytic study of the
Halstead-Reitan Neuropsychological Battery. In the context of that
battery of tests sensitive to a variety of dysfunctions, he found
five types of memory functions loaded with many nonmemory functions
on the same factors. This led him to conclude that the processing
of immediate or short-term memory in any given modality (i.e.,
verbal or visual) is carried out by discrete brain areas, but that
these discrete and modality specific storage functions act in a unit
for purposes of integrated activity for more consolidated long-term
24
memory activity. This would explain why highly lateralized and
specific lesion processes result in modality specific performance
on the WMS. It would also explain why the diffuse head injury pro
cess resulting from closed head injuries has an even greater adverse
effect upon WMS performance. That is, a disruption of connections
between lower brain centers and the cortex as well as intercortical
disruption occurs causing impairment of memory processing across
modalities with a special impact upon memory processes requiring
simultaneous integrating activity.
The studies cited above appear to emphasize the thesis of this
study. That is, particular types of memory and acquisitional
deficits result from the diffuse process of closed head injury. The
bulk of the literature reviewed on the results of head trauma upon
WMS performance also suggests that the more complex portions of the
WMS, or those requiring the simultaneous integration of complex
verbal stimuli, are also areas of particular susceptibility in dif
fuse closed head injury trauma. A discrepancy between these more
complex types of short-term memory processes with other cognitive
functions which require less efficiency or adaptation to complex
novel information would seem likely. In a following section, a re
view of the WMS and WAIS relationships may attempt to clarify this
possibility by examining the relationship between the WMS and the
WAIS for both normal and brain damaged populations. It would seem
logical that, if the WAIS contained many cognitive tasks which do
not require rapid acquisition of new memory, the pathological
25
affects of closed head injuries would be less noticeable on portions
or all of the WAIS than it is on the WMS.
The WAIS and Its Relationship to Brain Trauma
The WAIS will be reviewed in terms of the most recent and im
portant literature in regard to the performance of brain damaged
individuals on the WAIS. Previous to that review, however, a func
tional description of the test is provided such that the construc
tion of the subtests will be understood along with a shared under
standing of the domains of behavior measured by the subtests. A
description of those subtests will be provided by paraphrasing the
originator of the test (Wechsler, 1958).
Since Wechsler (1958) believed that intelligence was the net
result of interaction between various factors of intelligence, he
constructed his intelligence scale to be a battery of tests in order
that those factors might be measured separately. In choosing sub
tests he used criteria which consisted of an analysis of standard
ized tests already in use. An attempt was then made to evaluate the
validity of each subtest on the basis of correlations with other
recognized tests and empirical ratings of intelligence. An attempt
was then made to relate these tests to the clinical experience of
practitioners. Following the selections of subtests, some 2 years
were devoted to experimental work on individuals with known intelli
gence levels. He originally selected 12 tests and later dropped one
of the subtests called cube analysis.
The following is a listing and description of the subtests
following the most important elements in Wechsler's (1958) descrip
tions. A first subtest called the Information Subtest examines an
individual's range of information by accessing the amount of common
information an individual has the opportunity to acquire. The Com
prehension Subtest requires an individual to furnish his or her
answer to questions requiring problematic but practical thinking.
These subtests, however, are presented at different levels of com
plexity, such that the most complex requires some abstract thinking.
The Arithmetic Subtest was added to access computational skills in
volving practical calculations which must be done without the aid of
written reference. The Digit Span Subtest, which is used both on
the WMS and the WAIS, is one of the most specific subtests in terms
of the type of ability it measures and was included to estimate
attention and retentiveness. The Similarities Subtest requires
individuals to associate elements or concepts in terms of their com
mon elements and thereby stresses concept formation and to a certain
degree is controlled by the "logical character of the subject's
thinking processes." The Picture Arrangement Subtest, which begins
the performance section of the WAIS, requires individuals to sequen
tially understand disarranged picture stories and to correct the
order. It is supposed to measure the subject's ability to access a
total situation as well as its parts in sequence. The Picture Com
pletion Subtest requires analysis of a visual scene with the intent
of identifying a specific missing part thereby measuring speed in
visual analysis and attention detail. The Block Design Subtest is
basically a task of imitation requiring a person to assemble colored
blocks to match geometric patterns. It requires spatial perceptual
skills and ability to visually analyze form. The Digit Symbol Sub
test requires a subject to associate certain symbols with numbers in
a manner which requires visual speed and accuracy along with dexter
ity. The Object Assembly Subtest consists of four figure form-
boards consisting of an elephant, a maniken, a hand, and a face and
profile. These are presented to the subject in disarray with the
instruction to assemble them into a finished object. It thus has a
puzzle-like quality but visual discovery is important on the more
difficult items. The Vocabulary Subtest is basically a measure of a
person's ability to define words presented both verbally and visu
ally. It simply calls for a defined response to presented words.
The above subtests comprised the verbal portion of the battery
(Information, Comprehension, Vocabulary, Similarities, Arithmetic,
and Digit Span). The "performance" section of the battery includes
Block Design, Picture Arrangement, Picture Completion, Digit Symbol,
and Object Assembly. A group intelligence quotient score is ob
tained for the verbal subtest, as well as the performance subtest,
and then these scores are combined to obtain an age corrected full-
scale intelligence quotient.
A discussion of the subtests on different factors obtained in
factor analysis is beyond the scope of the present review. However,
it should be noted that Wechsler (1958) described three factors
which are currently identified. These are a verbal comprehension
factor, a nonverbal organizational factor, and a general intellectual
28
factor. Most of the verbal subtests load heavily on the verbal com
prehension factor with performance subtests loading highly on the
nonverbal organizational factor. Besides the general factor of
intelligence which shows relatively good correlations with most of
the subtests, a fourth factor has also emerged which is termed by
Wechsler to be "memory" or freedom from distractability. The meth
ods section contains a review of the basic normative data that per
tains to the WAIS (and its more recent form the WAIS-Revlsed Form).
Studies reflecting the effect of head trauma or organic brain
syndrome upon the WAIS have frequently been preoccupied with path
ology affecting either the right or left hemispheres. More recently,
researchers have attempted to contrast actual pathology types, such
as comparing closed head injury to other specific types of wound
processes. Representative studies which relate recovery of intel
lectual functions, as measured by the WAIS, to closed head injury
have been relatively recent. A range of this research will be
covered presently to help elucidate the meaning of the WAIS in the
context of measurement of brain dysfunction and to survey the most
recent relevant research of the psychometric properties of the WAIS
in respect to brain injury.
It has long been typical for studies to compare groups of brain
damaged persons with known lesions in either the right or left hemi
sphere on the WAIS. In this manner, several reports appear indi
cating that lesions of the left cerebral hemisphere were related to
impaired ability to deal with language and right hemispheric lesions
have been associated with impaired ability in the area of
29
visuo-spatial tasks (Fitzhugh, Fitzhugh, & Reitan, 1962). Other
relationships of WAIS performance patterns to type of cerebral dys
function became apparent, however. Fitzhugh and Fitzhugh (1964)
compared patients with lateralized cerebral dysfunctions to those
with diffuse cerebral dysfunction. In addition to the more simple
relationship of right and left hemispheric effects, they discovered
that the performance subtests were more impaired in the diffuse
lesion groups than were verbal subtests. Heilbrun (1956) also
called attention to the possibility that the performance subtests
are a more "heterogeneous" group of sub tests and speculated that
they may not be truly "nonverbal."
Parsons, Vega, and Burn (1969) compared bilateral lesions (both
hemispheres involved) with unilateral lesions (only right or left
hemisphere damage). They did support the hypothesis that left hemi
sphere damage results in impairment of language abilities and right
hemisphere damage results in decrement of visuo-spatial ability.
They investigated this hypothesis using only Vocabulary as a lan
guage score and Block Design as a performance score, however. Using
this technique, they could not discriminate a group with bilateral
head injury from a unilateral lesion group. They did not specify
the nature of the bilateral head injury subjects used in the study.
Zimmerman, Whitmyre, and Fields (1970) investigated differences
in the factor structure of the intelligence scores in right, left,
and "diffuse" cerebral dysfunction patients. In this group only a
minor portion of the diffuse dysfunction group were closed head in
juries. However, they did find that the greatest similarity in
30
factor structure appears between the right hemisphere injury group
and the diffuse injury groups. They interpreted these findings to
mean that the lowering of performance scores may be more predictive
of cerebral involvement than the lowering of verbal scores, though
performance scores, overall, are less predictive of the question of
unilateral versus diffuse cerebral dysfunction. Their study did,
however, support the presence of a "verbal factor, performance fac
tor and memory efficiency or freedom from distractability factor."
Simpson and Vega (1971) also studied the effects of unilateral
brain damage on the WAIS and attempted to control the confounding
effects of advancing age by using age corrected scaled scores to
study patterns of intellectual deficits associated with damage to
one or the other cerebral hemispheres. They achieved the expected
results. That is, brain damaged groups show more impairment than
controls in overall levels of function. However, they noted that
there was no significant difference between right hemisphere brain
damage, left hemisphere brain damage, and bilateral hemisphere brain
damage groups on the performance subtests. Woo-Sam (1971) offered
partial experimental verification that age compensated scores
achieve spurious results due to the affect of aging factors on the
performance subtest. He compared 21 subjects identified as right
hemisphere damaged and 15 subjects considered as left hemisphere
damaged within a single age group (ages 30 to 36) and failed to
report clear differences between right brain dysfunction and left
brain dysfunction in respect to verbal or performance intelligence
factors. His study, however, again does not consider overall
31
severity of injury. It does not include a description of localiza
tion of dysfunction or type of trauma.
In a group of mostly chronic brain damaged subjects, E. W.
Russell (1972) demonstrated that while WAIS performance is affected
by brain damage, its basic factor structure is not. He confirmed a
factor of general intelligence, verbal intelligence, and visual-
organizational intelligence along with memory and freedom from dis
tractability factors. He substantiated the reported history that
Digit Symbol was particularly affected by a variety of brain damage
conditions, but suggested strongly that laterality of brain damage
was not related to verbal and performance factors in these more
chronic brain damaged cases. The study suffers again from the
heterogeneity of the brain damaged types in the population study.
Lansdell and Smith (1975) studied 268 head injured servicemen
divided between cases of penetrating head wounds, closed head in
juries, and mixed cerebral damage. They applied factor analytic
methods and found a verbal factor loading on the Vocabulary Subtest
with a second factor loading on the Object Assembly Subtest which
was apparently linked to visual organizational skills. The patterns
of ability uncovered are similar to other studies. They interpreted
their results to indicate a resiliency for a verbal factor and a
tendency toward remaining impairment for nonverbal factors after
brain damage. They hypothesized that the WAIS performance intelli
gence quotient may contain a verbal component, making it more vulner
able to impairment in all types of brain damage.
In a similar vein, Todd, Coolidge, and Satz (1977) investigated
WAIS functioning in 335 patients. These were split between psychi
atric controls, 69 "diffuse" brain damaged cases, 46 right lateral
ized brain damaged cases, and 68 left lateralized brain damaged
cases. Like other studies, their data for localization was based
upon standard neurological procedures; however, their patient group
was again very mixed. The categories of brain damage included head
trauma, cerebral vascular accidents, tumors, degenerative diseases,
epilepsy, and a large unclassified group. They applied the WAIS
verbal intelligence quotient/performance intelligence quotient dis
crepancy index. This is the degree of difference between the per
formance intelligence quotient and the verbal intelligence quotient.
The results suggested that the verbal intelligence quotient/perform
ance intelligence quotient discrepancy not be used as a diagnostic
screening tool due to the fact that these discrepancy scores were
not consistently related to status or type of lesion. In all groups
they noted that mean performance intelligence quotients were sig
nificantly lower than mean verbal intelligence quotients.
The above studies concerning the lateralized indicators of
brain damage have much in common. They often use clearly heteroge
neous populations. For the purposes of this study it is important
that closed head injury as a distinct traumatic phenomenon is
largely ignored. While the same factors of intelligence are present
and substantiated for most control groups consisting of normal popu
lations, these studies also substantiate that the clear and simple
use of verbal intelligence quotient/performance intelligence
33
quotient discrepancy scores or differences are not easily relatable
to localization of trauma or type of brain damage. Also, on the
whole, they yield findings indicating a greater susceptibility of
the performance intelligence quotient factor to most types of brain
trauma with the exception of some cases showing highly localized
unilateral dysfunction.
These findings also generally lend support to Wechsler's (1958)
claim that brain damage to either or both hemispheres significantly
lowers the performance intelligence quotient but does not affect the
verbal intelligence quotient as dramatically. Moreover, these re
sults lend further credence to Smith's (1966) results. He found
that over 65% of his left hemisphere damaged group had a higher
verbal intelligence quotient than performance intelligence quotient.
He suggested that his findings indicated that impairment secondary
to a wide number of brain trauma types causes impairment in perform
ance intelligence quotient. Smith's subjects also varied dramati
cally in the type and recency of their brain damage. So while the
general tendency of liability of performance intelligence quotient
appears in Smith's study, the same criticism of mixed patient groups
apply.
A WAIS performance comparison between brain damaged populations
and neurotic adjustments was done by Ladd (1964). His brain damaged
group consisted of 50 male patients with undefined cerebral disease.
Also he purposely excluded grossly impaired brain damaged individ
uals. Interestingly, like many previous studies his results showed
that the brain damaged group was significantly lower in full scale
intelligence quotient and performance intelligence quotient but they
were not lower than the neurotic group in verbal intelligence quo
tient. Within group comparisons in that study suggested that verbal
intelligence quotient and performance intelligence quotient were
roughly equivalent; however, the verbal intelligence quotient and
performance intelligence quotient discrepancy in the brain damaged
group was significantly larger. There was no particular pattern of
individual subtest deficits in this heterogeneous group. This par
ticular study raised the question of whether any particular pattern
of subscore variation could be used in individual diagnosis of brain
damage. Similar general results were obtained by Vogt and Heaton
(1977) . They studied patients with extreme values on the impairment
index of the Halstead-Reitan Neuropsychological Battery. These are
a large group of tests which are sensitive to a variety of cerebral
dysfunction. When comparing impaired groups with unimpaired groups
on that index, they found that the impaired group did worse on all
11 WAIS subtests. Compared to the nonimpaired group, however, im
paired subjects appeared to do relatively poorer on Block Design and
Object Assembly and did relatively better on Information, Comprehen
sion, Vocabulary, and Picture Completion. These results are similar
to other results and suggest a greater sensitivity of the perform
ance subtest to impairment. In addition to using the Halstead
Impairment Index as a definition of impairment, however, this study
is important from the standpoint that it appears to support
Wechsler's (1944) idea of a "deterioration quotient."
35
In developing that deterioration quotient, Wechsler hypothe
sized "don't hold" and "hold" subtests. He suggested that, under
conditions of deterioration of brain function, "hold" subtests would
be less subject to performance decrements; and the "don't hold" sub
tests would be more subject to performance decrements following any
condition of deterioration. He hypothesized "don't hold" subtests
to be Digit Span, Digit Symbol, Arithmetic, and Block Design.
"Hold" subtests are Information, Object Assembly, Picture Completion,
and Comprehension. The above results generally support Wechsler's
contention with the exception of Object Assembly. It is seen in the
present review that Object Assembly appears to be part of a percep
tual organizational factor which is managed very poorly by brain
damaged individuals.
In another study which compared WAIS performance with the
Halstead-Reitan Neuropsychological Battery, Logue and Allen (1971)
confirmed a correlation between Wechsler full scale intelligence and
predicted category errors. The Categories Test on the Halstead-
Reitan Neuropsychological Battery is its single most valuable indi
cator of cerebral dysfunction. It is basically a concept formation
task involving higher level problem-solving skills. It requires
flexibility in judgment, problem solving, and analysis. Its rela
tionship to the Wechsler Adult Intelligence Scale full scale intel
ligence quotient is documented and the above authors developed a
table that would enable direct comparison scores to evaluate Cate
gory Test performance while correcting for intelligence quotient.
The nature and description of the Halstead-Reitan Neuropsychological
36
Battery and the specific indicators on that battery will be reviewed.
Mandleberg (1975) and Mandleberg and Brooks (1975) studied WAIS
performance as indicators of cognitive recovery after severe head
injury. Unlike other studies, they selected rather carefully for
the diffuse damage which results from the pathophysiological effects
of closed head injuries. Their findings are instructional from the
standpoint that they documented remarkable recovery in overall WAIS
intelligence quotient for their group of patients with severe head
injuries (posttraumatic amnesia greater than 1 week). During post
traumatic amnesia they discovered that verbal ability appeared to be
relatively intact, while nonverbal skills were extremely poorly exe
cuted. It should be realized, however, that they tested individuals
during a period where they were oriented but not necessarily capable
of responding to complex stimulation, including test instructions.
Over the period of serial testing, verbal subtest scaled scores
showed less initial impairment and were faster to recover than were
all nonverbal subtest scores. Verbal intelligence quotient scores
for a head injured group approached that of a comparison group
within about 1 year following the injury, while the performance
intelligence quotient continued to be deficient for a period last
ing over at least 3 years. They also noted that the pattern of
impairment on the WAIS for this closed head injury group corres
ponded to that typically displayed by the right hemisphere or bi
laterally damaged groups. That is, the discrepancy between verbal
and performance subtests continued to be present, with the perform
ance subtest being more poorly performed. Mandleberg (1976) also
37
related WAIS performance to length of posttraumatic amnesia. While
verbal intelligence deficits were related to posttraumatic amnesia
duration at 3 months, performance intelligence quotient deficits
continued to be correlated to posttraumatic amnesia after 6 months.
These relationships had disappeared by 30 months after injury. In a
study of a similar nature, Becker (1975) had administered the WAIS
to 10 patients who had sustained closed head injuries in auto acci
dents. This head injured group displayed initially severe deficits
with subsequent improvement on the performance subtests, especially
Digit Symbol, Block Design, and Object Assembly. The susceptibility
of perceptual organizational skills to diffuse head injury was re
inforced by the fact that Digit Span and Block Design appeared to be
the most sensitive indicators of improvement.
Other systematic studies of diffuse and closed head injuries
syndromes were carried out by E. W. Russell (1979, 1980). In both
studies, E. W. Russell was concerned with patterns of brain damage
as defined by the WAIS sub test profile. As has been the case with
much of the review of the literature, he suggested that diffuse
degenerative brain damage has about the same pattern of WAIS per
formance as does right hemisphere damage. He concluded that the
WAIS has defects from the standpoint of neuropsychological testing
in that the verbal portion does not have subtests that are highly
sensitive to brain damage. On the other hand, Digit Symbol, Block
Design, and Object Assembly, which constitute a majority of the per
formance subtests, are all sensitive to brain damage.
WAIS and WMS Relationships
The effects of brain dysfunction on the performance of the WMS
and the WAIS have been reviewed in separate sections previously.
While they are different in regard to sensitivity to brain dysfunc
tion, they have rather high intercorrelations as, indeed, they were
intended to (Wechsler, 1945). Libb and Coleman (1971) used out
patients with disability groups including mental retardation, psychi
atric disorders, and physical problems to study WAIS and WMS rela
tionships. They found significant correlations and a very close
relationship between the WAIS full scale intelligence quotient and
Wechsler memory quotient. This was true for all three groups, none
of which showed cr documented memory disturbance or brain damage,
however.
In a study which did correlate Wechsler memory quotients with
the Wechsler Adult Intelligence full scale intelligence quotient,
Fields (1971) found that in a brain damaged population, a similar
high and significant correlation between the WMS and the WAIS. How
ever, this study also failed to control for the effects of qualita
tively different types of pathology. This research population was
mixed to include focalized and nonfocalized lesions in an extremely
heterogeneous sample. Field (1971) argued that the WMS and WAIS
measured the same factor of "general intelligence." Likewise, Hall
and Toal (1957) suggested that the overlap between the WMS and the
Wechsler-Bellevue Intelligence Scale was so large as to not justify
giving both tests. Hall and Toal (1957) also noted occasional
subjects who achieved very low intercorrelations between the WMS
subtests. This was' interpreted as low reliability due to the brev
ity of the test. An alternative explanation, however, may be found
in the fact that all the subtests are combined to achieve a memory
quotient. Owing to the fact that the previously cited factor
analytic studies of the WMS have noted a complex audio-verbal factor
and an attention and concentration factor, it is likely that the
Wechsler memory quotient often is masking true variability in dif
ferent facets of memory performance and, hence, eliminating con
sistent differences when compared to the WAIS. A more pervasive and
sweeping difficulty in attention or concentration among brain in
jured might more favorably highlight WAIS and WMS differences. That
is, the freedom from distractability factor may not be as important
in normals or mixed .patient groups as it is in amnestic neurological
disabilities, particularly closed head injury.
Thus, while memory quotient and full scale intelligence quo
tient are highly correlated in individuals of average intelligence
and in mixed pathology (Hall & Toal, 1957; Libb & Coleman, 1971),
it would be important to investigate whether there are memory quo
tient and intelligence quotient differences in individuals with
syndromes which are clearly expected to evidence these differences
based on history or neuroanatomical evidence. In a group of 15
patients with Wernike-Korsakoff Syndrome (a syndrome known for short
term memory difficulties), Victor, Herman, & White (1959) found that
memory deficits as measured by the WMS did exist when compared with
normal verbal intelligence. The Korsakoff patients had mental
quotients that were generally in the 70s and well below their intel
ligence test scores. Interestingly, alcoholic patients without the
Korsakoff Syndrome did not show the Wechsler full scale intelligence
quotient and memory quotient discrepancy effect (Parsons & Prigatano,
1977) .
In one of the most specific studies of intelligence quotient
and memory quotient difference scores, Zaidel and Sperry (1974) com
pared those scores in eight patients who had undergone complete or
partial commissurotomy. These patients, who underwent separation of
the hippocampal commissure, showed memory quotient scores which were
always 12 points or more below their obtained intelligence quotient
scores. Wide ranging memory deficits were reported by individuals
who lived with those patients. Case reports of individuals with
known lesions of the hippocampus (Victor, Angevine, Mancall, &
Fisher, 1961) also show that those patients, who have expected
memory deficits by virtue of anatomical evidence, show rather clear-
cut intelligence quotient and memory quotient difference scores with
the latter being inferior.
Quadfasel and Pruyser (1955) compared scores from the Wechsler-
Bellevue Intelligence Scale and the WMS in 38 epileptic patients.
More than half of these manifested abnormal anterior temporal lobe
electroencephalogram (EEG) recordings. They were compared to
another 19 subjects with no focal abnormalities but with generalized
EEG abnormalities. The temporal lobe abnormal group showed a sig
nificantly greater memory quotient discrepancy than the group with
the generalized abnormal EEG. At that time Quadfasel and Pruyser
41
(1955) suggested that memory deficit would be said to exist when it
is 11 or more points below the intelligence quotient. The research
also suggested the WMS to be a test which was sensitive to dysfunc
tion in the temporal lobe, as well as being an instrument sensitive
to deficits in audio-verbal memory. A similar conclusion was
reached by Milner (1975), who suggested that a WMS quotient about 12
points below the full scale intelligence quotient is indicative of a
verbal memory impairment.
The above studies link aspects of pathophysiology that are ex
pected to produce brain impairment to actual deficits on the WMS in
relation to intelligence quotient. Other studies, as noted before,
have, however, shown less clear-cut discrepancy between full scale
intelligence quotient and memory quotient. Note that Howard (1950)
failed to find a memory quotient that could discriminate encephali-
tics from epileptics and paretics. Also, Cohen (1950) sought to use
the memory quotient discrepancy index to define memory impairment in
tumor, encephalics, and posttraumatic cases. He found no clear dis
crepancy between full scale intelligence quotient and memory quo
tient in these groups. He did not suggest, however, why he may have
expected to find memory difficulties as a specific syndrome in this
mixed group.
In 1974, Prigatano (cited in Prigatano, 1978, p. 823) compared
WAIS full scale intelligence quotient scores minus memory quotient
scores in 15 head injured patients who sustained coma after head
trauma. He compared these with a psychiatric control group without
coma. Average discrepancy scores for the head trauma group were
42
significant in showing a nearly 10-point discrepancy, on the average,
of memory quotient below intelligence quotient. The psychiatric
patients manifested no similar relative lowering of the memory quo
tient score and, in fact, showed somewhat higher memory quotient
scores as opposed to intelligence quotient scores. Interestingly,
he also found that there was a very high correlation between esti
mated time of unconsciousness and discrepancy scores. At that time
Prigatano (1978) suggested that was a preliminary sign which needed
to be evaluated further.
A summary of WMS quotient and WAIS quotient comparisons in
brain damaged individuals would suggest that memory quotient is
sensitive to highly specific lesions which are expected to produce
memory deficits, as reviewed. Also, we have seen that in groups of
mixed brain damage regular discrepancy scores are less frequently
obtained. Thirdly, the material reviewed suggests a probable intel
ligence quotient/memory quotient discrepancy in diffuse head injury
cases caused by closed head injury. Prigatano (1978) suggested that
low memory quotient scores should be expected in any patient with
diffuse brain dysfunction who shows overall cognitive impairment.
To this end, it would seem logical to compare cognitively impaired
closed head injury patients with individuals manifesting impairment
from localized lesions and to study the degree to which the full scale
intelligence quotient and memory quotient discrepancy correlates
with severity. These hypotheses will be more specifically stated
along with an integration of the literature review in the final
43
section following a review of the sensitivity of the Halstead-Reitan
neuropsychological indicators.
The Halstead-Reitan Neuropsychological Battery
The Halstead-Reitan Battery serves as a measure of brain impair
ment severity in this study. This section is devoted to a descrip
tion of its historical development and the research relevant to its
sensitivity as a measure of brain dysfunction.
Neuropsychological testing and the Halstead-Reitan Battery grew
out of attempts to understand the nature of intelligence. Halstead
(1947) sought to set forth a new conception of intelligence. At
that time he hypothesized that there were three categories of intel
ligence which included psychometric intelligence, clinical intelli
gence, and neurological intelligence. He considered the concept of
"psychometric intelligence" to stem largely from the investigative
efforts of psychologists. This was presumed to include such things
as "judgment," "abstract thinking," and other measures of adaptabil
ity as manifested by quantitative testing and scoring methods. He
felt that "clinical intelligence" was a product largely of clinical
investigations and involved those clinical descriptions oriented
toward describing the effectiveness of an individual's behavior.
Neurological intelligence was associated primarily with the effects
of brain lesions on human beings and lower animals. Halstead (1947)
believed that deficiencies existed in each of these explanations.
Halstead proposed the concept of "biological intelligence" which he
achieved through a factoral analytic study that utilized several
newly developed tests. He found four factors to be the central
integrative field factor, the abstraction factor, the power factor,
and the directional factor.
Besides Halstead’s concept of "biological" intelligence, the
importance of his early work was the development of a battery of
tests which he applied to brain damaged and control subjects in an
attempt to define the basic attributes of adaptive brain functioning.
A very large portion of Halstead’s Battery was adopted by
Reitan (1955a) for his early research on brain dysfunction. The
majority of the early Halstead tests of biological intelligence be
came the foundation of the Halstead-Reitan Neuropsychological
Battery. In his earliest attempt to validate Halstead's tests of
biological intelligence, Reitan (1955a) compared persons with symp
toms of cerebral damage with a normal control group. He found that
composite scores based upon a number of Halstead's tests were ex
tremely sensitive to organic brain damage. In particular, he found
that Halstead's category test differentiated clinical groups nearly
as well as the total impairment index from all of the other nine
subtests. This subtest will be described shortly. Later Reitan
(1956) administered the Wechsler-Bellevue Scale of Adult Intelligence
as well as Halstead's 10 tests again to 50 brain damaged patients
and 50 control subjects. He found that the Wechsler-Bellevue vari
ables were much more highly intercorrelated than were Halstead's
measures. He also found that Halstead's tests were much more sensi
tive to brain damage than the Wechsler-Bellevue subtests although
there was considerable overlap. Reitan concluded that while there
was a relationship between "psychometric" and "biological" intelli
gence, the latter did indeed appear to be more related to the integ
rity of the central nervous system and he continued on a span of
research into tests sensitive to brain dysfunction which has lasted
to the present.
Before revealing the sensitivity of the Halstead-Reitan Battery
that has been shown in more recent research, a description of that
battery will be provided. The following descriptions are paraphrased
from Reitan (1967).
The Category Test utilizes a projection apparatus (or more cur
rently a booklet form). The subject is told to select one of four
pictures which appears either on the page or the screen. Only one
response is allowed for each selection and he or she is given feed
back regarding whether his or her selection is right or wrong after
each trial. He or she is simply told that he or she has to learn a
"principle" that will allow him or her to select correctly one of
the four designs. The first group requires the matching of numerals.
In the second group the subject must learn to select one of four
numbers corresponding to the number of items appearing on the screen.
In a third group of items the subject must learn to select one
figure which is unique from the other three. In this section the
task becomes quite a bit more complex as it progresses through dif
ferent shapes, sizes, and colors of figures such that the stimulus
field constantly changes while the subject is required to maintain
a response consistent with an underlying principle. Additional
principles must be learned as the subject proceeds through the seven
subtests. The Category Test is therefore a nonverbal measure of
complex concept formation which requires an abstract ability to note
similarities and differences in constantly changing stimulus mate
rial while postulating hypotheses to serve as selection principles.
The Tactual Performance Test is a three-part test in which a
subject sits in front of a large board blindfolded. The "form
board" has a number of recessed shapes into which a specific shaped
block will fit. The subject's task is to fit the blocks into their
proper spaces using first his or her preferred hand, then his or her
nonpreferred hand, and finally both hands together. It requires the
development of a visualization of the spatial configuration of the
board without actually viewing it. Later, the patient is required
to draw the entire board from memory and place the shapes in their
respective correct locations. It therefore has a time, memory, and
localization component.
The Rhythm Test is a subtest of the Seashore Test of Musical
Talent requiring the subject to differentiate 30 pairs of rhythmic
beats. This requires sustained perceptive attention in the auditory
modality and auditory differentiation.
The Speech Sound Perception Test consists of 60 spoken nonsense
words with four nonsense words spoken on each trial period. The
subject's task is to discriminate the spoken word, and in doing so,
select that word from four closely sounding alternatives. It is a
test of audio-verbal receptivity and audio-verbal attention.
The Finger Oscillation or Tapping Test is a test of pure visual
motor speed using the index finger on the preferred and nonpreferred
47
hand in five consecutive 10-second trials.
The Trail Making Test consists oftwo parts, (a) and (b). Part
(a) consists of 25 circles distributed over a white sheet of paper
and numbered from 1 to 25.Part (b) consists of 25 circles numbered
from 1 to 13 and lettered from A to L. The subject is required to
connect these circles in consecutive order according to the numbers
or letters in rapid sequence. It requires a maintained sense of
sequence, attention and memory to the most recent point of depar
ture, and rapid visual scanning.
In the Fingertip Number Writing Perception Test, the subject is
required to report numbers which are written upon the fingertips of
each hand. It requires the tactile recognition of symbolic form.
On the subtest called Tactile Finger Recognition, the subject
is required to call out a number which is assigned to each finger.
In Tactile Form Recognition, the test requires the subject to
identify by touch alone: pennies, nickles, and dimes. Additionally,
the nonvisual and tactile recognition of shapes is also required.
In addition to the above tests, the neuropsychological battery
consists of a brief aphasia screening battery, a spatial relation
score which is derived from the Block Design of the WAIS in conjunc
tion with a Creek cross drawing, and sensory perceptual examination
which tests the sensory intactness on the face and upper extremities.
An indication of grip strength is also obtained.
Finally, the Digit Symbol subtest of the WAIS is included as a
separate subtest that contributes to the impairment index.
48
Since the previously cited studies (Reitan, 1955a, 1956), a
large amount of research has occurred in regard to the Halstead-
Reitan Battery aimed both at evaluating its sensitivity in the de
tection of various types of cerebral pathology and in attempts to
understand patterns of test scores indicative of types of lesions or
pathological processes. Reitan (1958a) administered the Halstead
Battery and the Wechsler-Bellevue Scale to 50 brain damaged patients
and 50 controls to assess whether that battery would show qualita
tive differences between the brain damaged and control groups or
whether simple quantitative differences would occur. His results
suggested that while brain damaged patients tend to show impaired
abilities, their manner of failure, as defined by types of errors on
a variety of subtests, are of the same type as normals. In particu
lar, Reitan discovered that the Category Subtest was extremely sensi
tive to brain damage despite the fact that the distribution of error
scores for brain damage was similar to the distribution for errors
for normal groups.
Reitan (1959a) more directly compared the effects of brain
damage on the Halstead Impairment Index with the Wechsler-Bellevue
Scale. The brain damaged group consisted of mixed cerebral path
ology. The Halstead impairment index was found to be significantly
more sensitive to brain damage than was the Wechsler-Bellvue intel
ligence quotient. All of the Halstead-Reitan subtests proved to be
more sensitive with the exception of two subtests called Critical
Flicker Frequency and Time Estimation. Later Reitan dropped both of
these subtests from the battery. Although, overall, the intelligence
49
quotient tends to be lowered by brain damage, it showed a smaller
magnitude of effect regardless of the Wechsler-Bellevue Scale
selected for comparison. In this study Reitan also found that the
Category Subtest was almost as sensitive as the entire neuro
psychological battery in discriminating cerebral dysfunction regard
less of the type of dysfunction present.
Other validating research on the Halstead-Reitan Neuro
psychological Battery was carried out with the aim of attempting to
make correct predictions regarding the presence or absence of brain
damage through statistical methods. Three studies (Wheeler, 1964;
Wheeler, Burke, & Reitan, 1963: Wheeler & Reitan, 1963) indicated
that correct classification of individual subjects in accordance
with neurological criteria ranged from 98.8% to 81%. The brain
damaged subjects included bilateral, unilateral, and diffuse brain
damage. In spite of these impressive levels for correct prediction
of the presence or absence of brain damage, they are scarcely better
than those obtained by impairment criterion based on simple cutting
scores on the Halstead-Reitan Battery (Reitan, 1967).
Other cross-validation studies have been done. Vega and Parsons
(1967) attempted to revise the nature of the impairment index. The
impairment index that was originally used was based on cutting
scores which defined either impairment or nonimpairment on any par
ticular subtest. Halstead's impairment index required that the
7 out of 10 subtests be within the impaired range resulting in the
necessity of a score of .7 to indicate impairment. Vega and Parsons
(1967) substituted T scores for each subtest, then calculated average
50
T scores. Using such a modified index, they were able to produce
nearly 80% correct classifications with a variety of established
brain damage types ascertained through neurological methods.
The Halstead-Reitan Neuropsychological Battery as a test which
is potentially usable to specify lateration, localization, and pro
cess aspects of brain lesions was studied by Filskov and Goldstein
(1974). Their validational study attempted to compare diagnostic
statements obtained by a clinical interpretation of the Halstead-
Reitan Neuropsychological Battery to diagnostic statements based on
commonly used physical diagnostic procedures. They claimed a per
fect success rate for the Halstead-Reitan Neuropsychological Battery,
demonstrating the contention that a neuropsychological diagnostic
approach, when combined with a clinical/actuarial approach, was
preferable to either clinical or actuarial approaches used alone.
Other research suggests that a subtest interpretation approach is
usable in identifying static versus rapidly growing brain lesions
resulting from tumors (Fitzhugh, Fitzhugh, & Reitan, 1961).
As successful as the Halstead-Reitan Neuropsychological Battery
has been for the detection and the specification of brain damage,
another subtest, in addition to the Categories Test, has shown to be
very highly sensitive to the variety of brain damaged conditions.
The Trail Making Test, as described previously, was very successful
in differentiating a mixed group of brain damaged patients from con
trols (Reitan, 1955b). In later studies, Reitan (1958b, 1959b)
administered the Trail Making Test to patients with brain damage and
achieved highly significant differences in performance of normals
and brain damaged groups. It was noted that brain damaged individ
uals usually do very well on Part (a) but have a great deal of dif
ficulty on Part (b) of Trail Making.
Reitan (1959b) computed correlation coefficients between the
Trail Making Test and the WAIS variables. These showed generally
significant correlations and at that time Reitan suggested that the
WAIS be used to establish ejcpected levels of performance so that the
performance requirements for the Trail Making Test could be estab
lished that would more specifically ascertain the presence or
absence of brain damage. In a study of Trail Making Test perform
ance in normal and brain damaged children, Reitan (1971) noted that
not only were brain damaged children more impaired than normals on
Trail Making, but also the factor of brain damage was a far more
potent predictor of test results than was chronological age among
children. This is a rather important indication of its selective
sensitivity to the effects of brain damage.
The Trail Making Test, like the Category Test, requires a
degree of alertness and visual comprehension of stimulus material,
interpretation of symbols in sequence, and a large degree of purpose
ful visual scanning. It also requires the patient to maintain a
memory of sequence simultaneously. In this respect, Part (b) of
Trails seems somewhat similar to the requirements of the Halstead
Category Test in which the patient must assimilate information from
one stimulus figure after another in an effort to grasp an organiz
ing principle which applies to the series. It Is little wonder, in
light of the literature reviewed, that these two indicators are of
52
special significance in terms of being powerful single indicators of
brain lesions.
Summary of Literature Review
In this section an attempt was made to highlight the major
findings of the literature reviewed and to draw tentative conclu
sions which lend themselves to the researchable hypotheses regarding
the relationship of the WAIS, the WMS, and closed head injury.
It was initially stated that memory was documented as a rapidly
plateauing variety of cognitive dysfunction and this is probably
related to severity in closed head injury. It was seen as a pos
sibly more vulnerable and long standing residual deficit following
closed head injury, especially in respect to short-term memory.
The WMS was reviewed and it was suggested that while it is not
a complete measure of memory functioning, it has repeatedly been
factored into short-term complex audio-verbal memory, attention
concentration, and freedom from distractability factors; and to a
certain extent, a third factor occurs which includes orientation and
access to simple overlearned information. A review then of the
impact of brain lesions and trauma upon the WMS, with some excep
tions, indicates that the above factors are also likely significant
in the poor performance obtained by closed head injury patients.
These patients performed more poorly than normals or patients with
some forms of localized brain lesions. This appears to be especially
true for complex audio-verbal tasks, i.e., Logical Memory and Paired
Associate Learning subtests. An inability of closed head injury
53
patients to adapt to novel and complex stimuli was suggested by
these findings.
Unlike the WMS, the WAIS was introduced and shown to be more
refractory to the effects of diffuse head trauma or closed head
injury. While the WAIS was seen as being sensitive to some forms
of highly localized lesions, at least the verbal portions are less
impaired following diffuse or closed head injury. Also, WAIS per
formance was seen to improve to higher levels with time and recovery.
This was more true for the verbal subtests, however, and it is sug
gested that the performance subtests, by virtue of their active pro
cessing requirements, are more vulnerable to a variety of trauma
effects. Therefore, neuropsychological indicators which also demand
novel or active integrative capacity are more likely to parallel
performance intelligence quotient scales.
A review of the direct WMS quotient and WAIS quotient compari
sons was presented. While the WMS quotient is discrepant and low in
comparison to the WAIS on diffuse and closed head injury groups,
less clear discrepancy scores were obtained for mixed pathological
populations. The WMS may be most sensitive to temporal lobe lesions
and verbal memory deficiencies.
Finally, the Halstead-Reitan Neuropsychological Battery was
introduced in terms of its sensitivity to a wide variety of brain
damage aspects. It was reviewed in terms of its sensitivity and
capacity for discrimination on a wide variety of organic brain syn
drome features. This was particularly true of the Categories and
Trails (b) subtests. Its universal sensitivity appears to be a
54
factor that apparently discriminates it from the WAIS as a measure
of residuals of brain trauma.
The above stated differences in sensitivities of the Wechsler
Memory Quotient and Wechsler Adult Intelligence Quotient may present
the unique opportunity to use them in concert in cases of closed
head injury or other cases of brain trauma where impairment of
short-term memory or impairment of active audio-verbal processing
is a suspected deficit area.
Much of the value in comparing the WMS to the WAIS may lie in
the long recognized differences between fluid and crystallized
intellectual abilities.
Halstead (1947) postulated the existence of two general forms
of intellectual abilities, calling them psychometric and biological
intelligence. Cattell (1943) used a similar concept even earlier
when he named two forms of intelligence to be fluid and crystallized
intelligence. Both of these theories can be explained by emphasiz
ing that fluid (biological) intelligence involves active mental pro
cessing of new material while crystallized (psychometric) intelli
gence is composed of well learned historically developed abilities.
It is possible that sensitivities of the WMS and WAIS are dif
ferent by virtue of the fact that the verbal portions of the WAIS
and some portions of the performance aspect of the WAIS measure
crystallized abilities. Hence, they recover more quickly following
specific types of brain damage and are more refractory to impairment
as a long-term residual of closed head injury. Conversely, neuro
psychological indicators, as discussed previously, are more sensitive
55
to the process of rapid perceptual integration or novel problem
solving which could be considered to be more of a fluid ability.
This concept would not only explain the differences between the WMS
and WAIS in response to brain damage but also would explain the
historical difference between the verbal subtests and the perform
ance subtests with the latter being considered as fluid abilities
and hence more chronically impaired under conditions of diffuse
cerebral trauma.
From a previous review, Horn (1976) advanced this hypothesis
directly by suggesting that most of the performance subtests of the
WAIS indeed measured right hemispheric, but also fluid abilities.
This made them more susceptible to impairment. E. W. Russell (1980)
tested this hypothesis specifically and found that in diffusely
organically damaged subjects the performance subtests were indeed
more affected than the verbal subtests. At that time E. W. Russell
commented that a test of fluid verbal abilities was not in common
use.
In addition to being a measure of short-term memory function,
the WMS may indeed be a measure of more fluid verbal abilities. Its
demands for new learning, concentration, and active mental process
ing of verbal material would possibly make it separate from the
verbal subtests of the WAIS in terms of its overall requirements.
This interpretation would be consistent with the literature review
carried out in this study which showed generally good correlations
between the WAIS and WMS except in some brain damage or special
populations. This observation has been supported in other research.
56
The known feature of closed head injury as previously discussed
emphasizes a disruption of the integrative aspects of brain function
ing. With such a hypothetical disruption of integrative functions
and memory difficulties as outlined previously, it is possible that
the WMS's discrepancy (that is, the WMS quotient below the WAIS full
scale score) would be an important measure both in estimating the
nature of neuropsychological disability and as screening information
regarding processing defects of specific types, such as fluid verbal
abilities and related memory functions.
Smith (1983) has recently included the WMS in the Michigan
Neuropsychological Battery and he has suggested that it be used
along with the Digit Symbol Modalities Test as an indicator of
deficits in rapid and fluid processing capabilities.
General Research Hypothesis
The general research hypothesis will be stated here in three
parts reflecting, therefore, the three major hypotheses to be
tested.
Hypothesis 1 : The WMS memory quotient score will be lower than
the WAIS full scale score in closed head injury patients. Such a
WMS performance discrepancy will not be present in a group of
localized brain damaged patients and a psychiatric control group.
Hypothesis 2 : The short coma group will 6how a smaller WMS
discrepancy score than will the long coma group. That is, the less
severely head injured will have more equal WAIS and WMS performances.
Hypothesis 3: Large WMS discrepancy scores will be associated
with large impairment ratings on the Halstead-Reitan Neuro
psychological Battery in the closed head injury group. Conversely,
low WMS discrepancy scores will be associated with lower impairment
ratings on the Halstead-Reitan Neuropsychological Battery.
Limitations of the Study
A limitation of this study derives from the fact that a number
of separate events in nature are studied that are not subject to
scientific control. The study relies upon pathological groups. In
this case, a closed head injury group, a localized trauma group, and
an unimpaired psychiatric control group were formed from existing
clinical populations. This does not provide confidence that each
group has unity in terms of the exact dysfunctions assumed or
measured. Every effort has been made to analyze presenting path
ology so that the truly localized brain syndrome patients and closed
head injury patients qualified for their respective groups. However,
dependence upon posttrauma information gathering is a liability as
is the absence of experimental controls over a number of unseen
variables which may contribute to these naturally occurring phenome
non.
A second limitation rests in the weaknesses inherent in static-
group comparison designs and correlational studies. An unknown num
ber of factors could contribute to the traits in each group in a
predetermined way or these effects might be independent of the named
pathology groupings.
CHAPTER II
DESIGN AND METHODOLOGY
Subjects
The total subject population in the present study consisted of
115 inpatients and outpatients from a large inpatient psychiatric
and rehabilitation facility. All subjects were referred specifi
cally for neuropsychological evaluation. The closed head injury
group, localized injury group, and unimpaired psychiatric group did,
however, differ slightly in terms of the source and reason for re
ferral.
The Unimpaired Control Group
The unimpaired psychiatric control group consisted of 25 males
and 20 females. The age range of the control group was 18 to 56
years old with a mean age of 28.4 years and a median age of 25 years.
An initial pool of 71 inpatients was identified who had been re
ferred for neuropsychological testing. From the original pool of 71
cases, 18 were excluded due to incomplete test files or data. Four
of the original pool were excluded due to a stated history of mild
closed head injuries which had occurred in their distant past but
were noted to have resulted in at least some notable residuals,
hospitalization, or a diagnosis of head injury. Two others were
excluded from the original pool due to actual neuropsychological
ratings which placed them within the defined range of impairment.
58
One additional case was dropped due to an inability to read and a
history of learning disability. An additional case was excluded due
to the presence of a cerebral vascular accident in recent history.
The total number of excluded subjects was 26 leaving 45 for a total
number in the group.
"Pure" clinical histories were difficult to obtain for the psy
chiatric unimpaired group as well as the closed head injury group
and localized trauma group. It should be noted that the psychiatric
control group evidenced a number of individuals (nearly half of
those sampled in all groups) to have suffered minor blows to the
head which they claimed resulted in very transient effects with no
stated residuals, posttraumatic amnesia, or confusion of any type.
It is assumed that these very transient head injuries are ubiquitous
in the general population and no attempt was exhaustively made to
screen out every individual who reported some blow to the head at
sometime in their personal history. Three persons in the psychi
atric control group showed abnormal EEGs at sometime in their per
sonal history. This group also included two cases with possible
encephalitis.
All individuals in the unimpaired control group were referred
largely from an inpatient psychiatric facility. At the time of re
ferral, the treating clinician usually attempted to screen out
subtle processes of thought disorder, memory difficulties, or other
difficulties, such as concentration. A large variety of the refer
rals were motivated by histories of alcohol or street drug abuse.
60
Referrals with this type of history constituted almost half of the
referred group.
The Closed Head Injury Group
The closed head injury group consisted of 45 cases. These were
drawn from an original data pool of 59 cases who were examined neuro-
psychologically. Nine of these cases were dropped due to incomplete
testing or missing scores. Two were excluded due to equivocal his
tories of head injury. That is, it could not be ascertained either
in their history or from their report whether they had a diagnosed
history of head injury with any resulting symptomology. One was
excluded due to a coexisting learning disability and reported birth
trauma which influenced intellectual development. An additional
case was dropped due to a long standing history of hearing impair
ment which lowered the verbal intelligence quotient. One additional
case was dropped due to penetrating head wound which coexisted with
closed head injury.
Of the remaining 45 cases in the closed head injury group, 16
showed some additional localizing effects. That is, their histories
were positive for hemiparesis, focal signs, hematomas, or some signs
of abnormal EEG or seizure activity. Four of the group were neuro-
logically suspected of suffering an anoxic effect. Fifteen to
twenty percent of the cases had head injuries severe enough to show
decerebrate posturing at the time of emergency room admission.
There were 35 males and 10 females with a mean age of 30 years,
range of 16 to 54 years, and a median age of 27.5 years. It was
61
predominately a young sample. Twenty-two of the 45 cases fell below
age 26. There were 13 cases over 40 years of age and 10 cases were
in their 30s.
In terms of recency of injury, the closed head injury group
showed a mean of 45 months post-injury. There were four cases where
the head injury was very remote, that is, 12 years or more. When
these were excluded from the 45 cases, the average recency of injury
dropped to 30.6 months post-injury. Recency of injury ranged from a
high of 20 years to a low of approximately 9 months.
In terms of length of coma, the head injury sample showed a
median length of coma of 14 days. This ranged from zero days to 90
days. Seventeen of the cases had comas of 5 days or less. Seven
teen cases had comas of 28 days or more. An additional 13 cases had
comas ranging from 14 to 28 days. In those cases with zero days of
coma, there were documented confusional stages, memory impairment,
or posttraumatic amnesia, such that even though coma was not present,
rather clear stated or documented residuals from the head injury
occurred.
The Localized Trauma Group
The localized trauma group consisted of an original pool of 39
non-closed head injury patients who suffered some localized lesion.
From this data pool four were dropped due to incomplete measures or
data. An additional two cases were dropped due to the confounding
presence of closed head injury in their history.
The localized lesion group, since it contained many stroke patients,
was a more aged group. They had a median age of 60 with a mean age
of 54.9 years. The ages ranged from 29 to 78 years. The group con
sisted of 15 males and 10 females. Of the total group 19 suffered
either cerebral vascular accidents or aneurysms. Six suffered
tumors. In terms of lateralization, seven of the sample has lesions
clearly lateralized to the left hemisphere. Nine members of the
sample had lesions clearly lateralized to the right hemisphere. An
additional nine individuals showed equivocal lateralization based on
their neurological history, and some were unable to report specific
localizing data.
Table 1 summarizes the sex and age comparisons between the
groups. Also, Table 1 provides a comparison of the mean neuro
psychological impairment ratings. This variable was included here
to afford a direct comparison of the level of severity of each group
in terms of measured deficits. As might be expected, the unimpaired
psychiatric group was very nearly average in terms of their group
performance (an average, or unimpaired, score is 1.0).
The localized lesion group is clearly oldest and most impaired.
Group comparisons in terms of educational level are not tabulated.
Although the educational level for all groups is estimated to be
near high school completion, the closed head injury group and the
psychiatric group evidenced lower levels of occupational and educa
tional attainment. This is likely associated with the high number
of serious drug abusers and troubled adolescents comprising the
psychiatric group and the large number of very young adults and
63
teenagers in the closed head injury group who had educational and
occupational progress interrupted by their injuries.
Table 1
Comparison of Closed Head Injury Group, Localized

Lesion Group, and Psychiatric Unimpaired Group
by Sex, Mean Age, and Mean Neuropsychology
Impairment Rating
Sex Mean
Mean
neuropsyc.
Variable N age
impairment
M F (yrs.)
rating
Closed head
injury group 45 35 10 30.0 1.98
Localized
lesion group 25 15 10 54.9 2.33
Psychiatric
unimpaired
group 45 25 20 28.4 1.02
Subject Selection and Group Assignment
The subjects utilized in the present study were patients who
had undergone either clinical treatment or formal assessment to
understand, rehabilitate, or treat clinically presenting conditions.
Therefore, the three groups, which consisted of the psychiatric
unimpaired group, the localized lesion group, and the closed head
injury group, in many ways constitute naturally occurring phenomenon.
Therefore, the clinical entities, which comprise the three groups
are not considered to be "pure." Also, the lack of purity in clini
cal syndromes is a recognized liability in such group comparisons.
As was noted In the description of the subject composition for each
group, it was difficult to find patients with absolutely no other
confounding factors in their history which might have contributed
some measurable type of brain dysfunction. It is likely that these
events are commonplace in nature such that minor blows to the head,
high fevers, history of fainting spells, abnormal EEGs, or other
features would be impossible to rule out.
Subject selection criteria relied upon patient history and any
accompanying neurological data. An effort was made to be very con
servative and not to include subjects in any group for which the
self-report or accompanying neurological history or data was un
certain.
Individuals in the closed head injury group were required to
have suffered a head injury for which they were treated and which
resulted in some residuals which were brought to the attention of
treatment personnel. Self-report was used as a criterion in a minor
portion of the closed head injury cases. In this instance patients
had to exhibit a clear recognition of whether coma exceeded 5 days.
Cases where self-report was equivocal of this regard, or where other
data were not present, were omitted. Those cases with no clear evi
dence of coma were referred due to posttraumatic confusion and to
ascertain cognitive dysfunction which was residual and persisting
from the time of trauma.
The localized lesion group had a neurological history of either
cerebral vascular accident, aneurysm, or tumor. They were included
in this group regardless of severity or localization of the injury.
65
In this group all individuals were referred to ascertain the pos
sible residuals of these events. It was important with this group
to demonstrate the absence of diffuse brain damage history or closed
head injury. Therefore, patients with any such history were dropped
from consideration as noted in the description of that population.
A number of those individuals who were reporting extremely minor
head injuries were retained if it was clear that there were no re
ported Immediate or retained sequelae to these injuries. Since
these events were considered to be frequent in the population,
exclusion criteria to rule out the most minor of head injuries would
seriously have limited candidates for inclusion in a non-closed head
injury sample.
Subjects included in the psychiatric control group were in
cluded if they had no history of stroke or localized lesion. Also,
it was required that they have no known closed head injury. The
criterion for the inclusion of this group was an unimpaired rating
of the Halstead-Reitan Neuropsychological Battery. Therefore, it
serves as a contrast which requires subjects not to have lesions of
the type involved in the other two groups. A cut-off of an impair
ment rating of 1.55 was utilized. That is, individuals obtaining
scores of 1.56 or higher were excluded and defined as impaired fol
lowing the subtest averaging method described by E. W. Russell,
Neuringer, and Goldstein (1970). These authors considered an impair
ment rating of 1.55 to be the dividing line between the average
range of functioning and the mildly impaired range of functioning.
In the present study three cases with an impairment rating of 1.58
66
were included due to the fact that the examiner in the individual
case offered the clinical judgment of nonimpairment. Thus, it was
not felt that these three cases would harm the integrity of the
assumption of nonimpairment for the control group. There was no
attempt made to control for education, age, or other demographic
factors. The possible confounding effects of group selection vari
ables and their limitations for interpretation will be discussed at
a later time.
Criteria Instruments
The three measures used in this research were the WAIS, the
WMS, and the Halstead-Reitan Neuropsychological Battery. An addi
tional variable was length of coma. The origin, description, and
validity data relative to these instruments were provided in appro
priate sections of Chapter I. The major reporting data regarding
their structure and validity, however, will be summarized.
The WMS has repeatedly been shown to be sensitive to factors of
very short-term registration memory, attention and concentration,
and associational flexibility despite its weakness as a measure of
the wider variety of memory functions (Dye, 1982; Kear-Cowell, 1977;
Kear-Colwell & Heller, 1978; Prigatano, 1978). It was designed to
be comparable to the Wechsler-Bellevue Intelligence Scale (Wechsler,
1945), and in the absence of severe cerebral dysfunction, it is not
usually discrepant from the WAIS (Prigatano, 1977). This instrument
is included here to serve as a screening instrument for the types of
short-term memory and attentional deficits noted in the diffuse
67
closed head injury.
The WAIS has been shown to be sensitive to four major factors
of intelligence. These are a verbal comprehension factor, a non
verbal or visual organizational factor, a general intellectual fac
tor, and a freedom from distractability factor (E. W. Russell, 1972;
Wechsler, 1958). Furthermore, rather consistent findings in the
literature suggest that while the verbal and performance intelli
gence quotient scores are sensitive to left hemisphere and right
hemisphere lesions, respectively, many types of brain damage, and
more particularly diffuse brain injury, result in greater impairment
of the performance intelligence quotient score (Lansdell & Smith,
1975; Smith, 1966; Todd, Coolldge, & Satz, 1977). The Wechsler
Adult Intelligence Scale and its replacement version, the Wechsler
Adult Intelligence Scale-Revised Form (Wechsler, 1981), is the most
widely standardized individual intelligence test currently used. In
the standardization of the Wechsler Adult Intelligence Scale, large
stratified samples were used based on the census data so that scores
would be representative of the race, geographic region, age, sex,
and occupational variety reflected in the United States census cal
culations.
Wechsler (1981) reported reliability and validity data in his
manual. Split half reliability and test-retest reliability coeffi
cients were obtained. The test-retest reliability coefficient for
the verbal intelligence quotient is reported to be .95. The per
formance intelligence reliability quotient is reported to be .88,
and the full scale intelligence reliability quotient score is
68
reported as .96.
Validity studies have been undertaken to compare WAIS results
with global indicators of intelligence, such as level of education,
as well as job and school performance. Such correlations are typi
cally on the order of .50 (Woo-Sam, 1971; Zimmerman et al., 1970).
A comparison of average intelligence quotients reveals the WAIS
intelligence quotient to be about eight points higher than the cor
responding intelligence quotient of the WAIS-Revised Form (Wechsler,
1981). To make WAIS-Revised Form scores comparable to WAIS scores,
eight points were subtracted from the full scale WAIS-Revised Form
score. In this manner, all subjects who were assessed with the WAIS-
Revised Form were given an adjusted full scale intelligence quotient
score.
The Halstead-Reitan Neuropsychological Battery has been previ
ously reviewed, as a part of this study, detailing its sensitivity
to various types of brain damage (see Chapter I). It has been
normed on clinical populations with known types and degrees of cere
bral trauma. It was shown to be more sensitive to brain damage than
the Wechsler-Bellevue intelligence quotient (Reitan, 1953). It was
also shown to discriminate brain damage from non-brain-damaged indi
viduals with a correct classification of 81% to 98.8% (Wheeler,
1964; Wheeler et al., 1963). That the Halstead-Reitan Neuro
psychological Battery is potentially usable to specify lateraliza
tion, localization, and process aspects of brain damage has been
affirmed by Filskov and Goldstein (1974).
Research Design
Data Collection Procedure
Data were collected in the present study through a review of
neuropsychological data which were collected in a private psychiat
ric and rehabilitation facility beginning in 1976 and ending in
November of 1983. The subjects tested before 1976 could not be uti
lized as they were not administered the WMS in conjunction with
neuropsychological testing. All subjects were tested by psycho
metric technicians who had formal training in the administration of
the Halstead-Reitan Neuropsychological Battery. Each case was also
subjected to a clinical interview which was part of the testing re
view and report writing procedure. With each neuropsychological
case which was eligible for the data pool, a formal intake question
naire was utilized. This is a self-report format which was individ
ually administered and accessed the client's history in terms of
important signs of cerebral trauma or pathology. Additional data
were often gathered by referral sources. In many cases of closed
head injury, additional historical data were accomplished through
interview or questions directed at family members. Therefore, data
collection involved the ongoing collection of suitable candidates
and a record of relevant data as they became available. It also
involved a review of previously stored cases by category of injury.
In most cases, the type of cerebral pathology was very clear-cut.
In some cases, as was described in the section on subject selection,
equivocal results were obtained and this resulted in exclusion.
Design of the Study
The current study is an ex post facto analysis utilizing static
group comparisons. In naturalistic observations, such as the brain
trauma subjects involved in the present study, it is necessary to
test variations and alternative aspects of the hypothesis to add
strength to the tentative conclusions gained from lack of control
over independent variables (Kerlinger, 1973). Therefore, in the
present study the WAIS and the WMS discrepancy score was related to
closed head injury both by comparing it to groups chosen to hypo
thetically contrast in terms of their discrepancy score outcome and
by testing for correlation with other measures which relate severity
of closed head injury to the WMS discrepancy score.
These alternative hypotheses are not claimed to compensate for
threats to internal validity or generalizability of results inherent
in this design (Campbell & Stanley, 1966). Rather the analysis was
used to determine what relationship may exist between severity of
closed head injury as defined by performance on the Halstead-Reitan
Neuropsychological Battery and the WMS discrepancy score.
The statistical analysis used to test Hypothesis 1 was a one
way analysis of variance (ANOVA) to determine if there was a sig
nificant difference between the localized trauma group, the closed
head injury group, or the unimpaired psychiatric group. The pro
tected least squares difference procedure was applied to this three-
group ANOVA. This was done to determine which of the three group
means were significantly different from each other. The expected
71
result for Hypothesis 1 would be a significant difference between
the mean WMS discrepancy score of the localized lesion group, the
closed head injury group, and the unimpaired psychiatric control
group. More specifically, it should also be found that the mean WMS
discrepancy scores for the closed head injury group are signifi
cantly different from the mean WMS discrepancy scores of the local
ized lesion group and the psychiatric control group. If large WMS
discrepancy scores are truly unique to closed head injury, it would
also suggest that there will be no difference between the mean WMS
discrepancy scores of the localized lesion group and the unimpaired
control group.
The statistical analysis used to test Hypothesis 2 was a one
way ANOVA to establish F probability. This was used to determine if
the short coma group was significantly different from the long coma
group in regard to the WMS discrepancy score. The expected quanti
tative outcome for Hypothesis 2 is a significant difference between
the mean WMS discrepancy scores for the long and short coma groups.
This difference between groups should account for the greatest pro
portion of the discrepancy score differences. Specifically, the
short coma group should have smaller discrepancy scores than the
long coma group.
The expected result of Hypothesis 3 is a linear relationship
between the magnitude of the Halstead-Reitan Neuropsychological
Battery severity score and the magnitude of the WMS discrepancy
score. That linear relationship should be positive. Increases in
neuropsychological severity rating should occur with increases in
discrepancy score.
An alpha level of .05 was used to test all of the null hypothe
ses against their respective alternate hypotheses.
Additional statistical procedures used to clarify outcomes are
presented along with results of those procedures in Chapter III.
Research Hypotheses (Operational Hypotheses)
The three central hypotheses in the current study are presented
here in terms of specific parameters. The specific hypothesis to be
tested (the null hypothesis) will accompany each hypothesis.
Hypothesis 1 : There will be a significant difference between
the mean WMS discrepancy score of a closed head injury group from
the mean WMS discrepancy scores of an unimpaired psychiatric control
group and a localized lesion group.
Null hypothesis: There will be no difference between the mean
WMS discrepancy scores of the closed head injury group, the local
ized lesion group, and the psychiatric unimpaired control group fol
lowing analysis of variance.
Hypothesis 2 : The mean WMS discrepancy score of a short coma
group will be significantly less than the mean WMS discrepancy score
of a long coma group.
Null hypothesis: There will be no difference between the means
of a short coma group and a long coma group following analysis of
variance.
Hypothesis 3: There will be a significant positive correlation
between severity of head injury as measured by the Halstead-Reitan
73
Neuropsychological Battery impairment rating and the size of the WMS
discrepancy score.
Null hypothesis: The Pearson product-moment correlation co
efficient between the Halstead-Reitan Neuropsychological Battery
impairment rating and the WMS discrepancy score will be zero.
CHAPTER III
DATA ANALYSIS
This research attempted to investigate the relationship between
the WMS discrepancy score and closed head injury. Conceptually, it
was hypothesized that closed head injury results in deficient WMS
performance relative to WAIS performance. This hypothesized deficit
in short-term memory processing is explored by comparing a less
severe head injury group with a more severe head injury group as
defined by length of coma, comparing closed head injury WMS discrep
ancy scores with other impaired and unimpaired groups, and finally,
by correlating WAIS discrepancy scores with a neuropsychological
test battery severity score. Data are presented in this chapter
that will portray these comparisons. Initially presented are data
from the entire research population that are descriptive. Research
data pertaining to group difference and hypothesis testing will
follow. Additional statistical analysis will then be presented.
Analysis of Population Data
The descriptive data for the entire research population are
presented in Table 2, where the mean, range, and standard deviation
scores are presented for the WAIS, WMS, Halstead-Reitan impairment
rating, and the WMS discrepancy scores.
As can be noted, the mean scores for both the WAIS and the WMS
are within the average range. However, the range and standard
74
75
deviation for the WMS are comparatively large with a range of 91
points and a standard deviation of 19 points. Thus, relative vari
ability in WMS performance is suggested with the implication that
it is more subject to individual differences within the whole popu
lation. The widely varying discrepancy score (standard deviation,
11.61) is also likely associated with the wider variance in WMS
scores.
Table 2
Means, Ranges, and Standard Deviations for the WAIS, WMS,

Halstead-Reitan Neuropsychological Impairment
Rating, and WMS Discrepancy for the
Entire Research Population
Mean Range SD
WAIS 99.30 66.00 12.34
WMS 94.53 91.00 19.00
Halstead-Reitan
impairment rating 1.67 2.91 .79
WMS discrepancy score 4.77 56.00 11.61
The average neuropsychological severity rating is close to the
nonimpaired cut-off score of 1.55 with an average severity rating of
1.67 and a standard deviation of .79 points. While the range of
severity scores is from .42 (superior) to 3.33 (moderately impaired),
they suggest a less variable distribution across the entire sample
than do scores associated with the WMS. These descriptive observa
tions may be due to the fact that both the Halstead-Reitan test and
the WAIS assess a wide variety of individualized cognitive functions
and are therefore less likely to evidence extremes in range and
variance in their composite scores. It would appear important to
note, however, the rather extreme range and variance in the WMS
scores as compared to other performance measures obtained.
Hypothesis 1: In order to test Hypothesis 1, an ANOVA was com
puted to test the null hypothesis that there would be no difference
between the means of the unimpaired psychiatric control group, the
localized lesion group, and the closed head injury group. Addition
ally, the protected least squares difference test was applied to
ascertain which group differences were significant. The ANOVA did
reveal significant differences between the groups, F(2, 112) = 6.241,
£ = ,0027 (see Table 3). Therefore, the null hypothesis of no dif
ference between group means could be rejected at the established .05
level of confidence.
The least squares procedure confirms the expected group differ
ences. That is, only the null hypothesis of no difference between
the means of the closed head injury group and the other two groups
could be rejected at the .05 level. At the .05 level of confidence
a null hypothesis of no difference between the psychiatric unimpaired
group and the localized lesion group could not be rejected.
Hypothesis 2 : The null form of Hypothesis 2 states that there
will be no difference in the mean WMS discrepancy scores between a
short coma group and a long coma group. Hypothesis 2 was tested by
computing a one-way analysis of variance and values are shown in
Table 4.
77
Table 3
ANOVA for Three Ranges of WMS Discrepancy Scores

(Closed Head Injury Group, Localized Lesion
Group, and Psychiatric Unimpaired Group)
Source SS df MS F £
Between 1540.90 2 770.45

6.241 .0027
Within 13827.22 112 123.45
Total 15368.12 114
Group N X SD
Psychiatric unimpaired group 45 2.00 11.82
Localized lesion group 25 1.50 12.77
Closed head injury group 45 9.33 9.23
Table 4
ANOVA for Two Ranges of WMS Discrepancy

Scores by Length of Coma
Between 336.56 1 336.56

4.237 .0456
Within 3415.43 43 79.42
Total 3752.00 44
Group N X SD
Short coma group 17 5.82 8.00
Long coma group 28 11.46 9.41
78
The obtained F value is (1, 43) = 4.237, £ = .0456. These
values are sufficient to reject the null hypothesis at the .05 level
of confidence. It does support the conceptual hypothesis of differ
ences in WMS discrepancy scores between these long and short coma
groups.
Hypothesis 3 : The null form of Hypothesis 3 states a zero
Pearson product-moment correlation coefficient between the neuro
psychological impairment rating and the WMS discrepancy score in a
group of closed head injured.
The obtained correlation was _r = .3236, £ < -05 (j> = .030).
The null hypothesis could, therefore, be rejected at the .05 level
of confidence. The conceptual hypothesis of a linear positive rela
tionship between the size of WMS discrepancy score and the neuro
psychological impairment rating is supported.
Additional Statistical Analysis
In order to further examine the relationships between type of
cerebral trauma, intelligence, and WMS performance, additional sta
tistical procedures were run in an attempt to clarify the sensitiv
ity of specific instruments to type of cerebral trauma.
In order to examine the specific sensitivity of the WAIS to the
three groups, a one-way analysis of variance was computed, along
with the protected least squares procedure to determine any differ
ences between groups on the WAIS full scale IQ score. The values
for the analysis of variance are presented in Table 5. The result
would suggest a rejection of a hypothesis of no difference between
79
those groups, F(2, 112) = 7.511, £ = .0009. Computed results of
least square tests further indicate, however, that significant dif
ferences are present between only the psychiatric unimpaired group
and the closed head injury group.
Table 5
ANOVA of WAIS Full Scale Scores for Closed Head Injury

Group, Localized Lesion Group, and a
Psychiatric Unimpaired Group
Between 2054.36 2 1027.18

7.511 .0009
Within 15315.98 112 136.74
Total 17370.34 114
Group N X SD
An ANOVA and protected least squares test was also computed for
the WMS to determine differences in performance between the closed
head injury, localized lesion, and psychiatric unimpaired groups.
ANOVA results suggest a rejection of the null hypothesis of no dif
ference between groups based on WMS performance also, F(2, 112) =
10.73, 2 = .0001.
. The accompanying protected least squares procedure
suggested a significant difference between the psychiatric unimpaired
group and the closed head injury group in addition to the significant
difference between the localized trauma group and the closed head
injury group (.05 level). Those results are presented in Table 6.
These results are suggestive that performance on WMS in the closed
head injury population is impaired in comparison to the localized
trauma group and the psychiatric control group.
Table 6
ANOVA of WMS Quotients for Closed Head Injury Group,

Localized Lesion Group, and a Psychiatric
Unimpaired Group
Source SIS cif MS F £
Between 6618.57 2 3309.28

10.73 .0001
Within 34538.07 112 308.37
Total 41156.64 114
Group N X SD
An additional analysis of variance was computed for the WAIS
by the length of coma. These results suggest no significant differ-
ence in WAIS performance between short and long coma groups,
F(l, 43) = 2.113, £ = .1533. Computed values are provided in
Table 7.
81
Table 7
ANOVA of WAIS Full Scale Scores for

Long and Short Coma Groups
Between 310.27 1 310.27

2.113 .1533
Within 6313.36 43 146.82
Total 6623.64 44
Group N X SD
The suggestion of no significant difference between the long
and short coma groups on the WAIS variable was unlike the results of
the analysis of variance for the WMS quotient between short and long
coma groups. Significant difference was obtained between the short
and long groups beyond the .05 level, F^(l, 43) = 5.063, £ = .0296
(see Table 8). These results would confirm that while the WAIS may
not be sensitive to the length of coma variable, the WMS quotient
is more sensitive to that dimension of severity.
Pearson product-moment correlation coefficients were computed
to obtain additional information about the degree of relationship
between the Halstead-Reitan Neuropsychological Battery impairment
rating and other variables, including the WMS memory quotient, WAIS,
and WMS discrepancy score. These correlations were computed for
both the localized trauma group and the closed head injury group.
Table 8
ANOVA of WMS Quotients for Long and

Short Coma Groups
Source SS df MS
-- F p
—
1
Between 1293.14 1 1293.14

5.06 .0296
Within 10983.16 43 225.42
Total 12276.31 44
Group N X SD
The WAIS full scale IQ score correlates with the neuro
psychological impairment rating at -.75 (jj = .00) in both the closed
head injury group and the localized lesion group. The WMS memory
quotient shows a correlation with the neuropsychological impairment
rating of -.73 (j> = .00) in the closed head injury group, and a
correlation of -.66 (ja = .00) in the localized lesion group. The
correlation between the WMS discrepancy score and the neuro
psychological impairment rating in the localized lesion group is
.33 (j> = .105). Recall that this is nearly the same as the correla
tion coefficient between the WMS discrepancy score and the neuro
psychological impairment rating in the closed head injury group
(.32, £ = .03).
The above correlations do permit the rejection of the null
hypothesis of a zero correlation between the WMS discrepancy score
83
and the neuropsychological impairment rating. These correlations do
not lend strength, however, to the corollary hypothesis that such a
relationship is unique to closed head injury.
Discussion
Hypothesis 1 predicts a difference in the WMS discrepancy score
between the closed head injury group, the localized lesion group,
and the unimpaired control group. This hypothesis was strongly
supported by the results. ANOVA did reveal differences between the
groups on the dimension of WMS discrepancy score. The null hypothe
sis of no group difference was rejected at a confidence level of
.05. Furthermore, the nature of those group differences were as
hypothesized. The closed head injury group was significantly dif
ferent from both the unimpaired psychiatric control group and the
localized lesion group. The results of the protected least squares
difference procedure could not support an assumption of difference
between the psychiatric unimpaired control group and the localized
lesion group on the WMS discrepancy dimension.
A clear difference is seen on the examination of the mean WMS
discrepancy scores for each group. The mean WMS discrepancy score
in the closed head injury group is nearly 9.5 points. The psychiat
ric unimpaired group and the localized lesion group show mean WMS
discrepancy scores of 2 and 1.5 points, respectively. Not only are
these WMS discrepancy scores low in comparison to those obtained by
the closed head injury group, but the small differences between the
psychiatric unimpaired group and the localized lesion group are
suggestive of similarity in terms of WMS performance.
Two other ANOVAs were performed to ascertain what differences
there might be in WAIS and WMS performance between the three groups.
The ANOVA and protected least squares difference test revealed only
a significant difference (at the .05 level) between the closed head
injury group and the unimpaired psychiatric group on the measure of
WAIS full scale IQ. However, an ANOVA and protected least squares
procedure reveal significant difference (.05 level) between the
closed head injury group and both other groups on the dimension of
WMS memory quotient performance.
The comparisons of WAIS and WMS performances by group are sup
portive of the conceptual hypothesis of greater WMS deficits in
closed head injury and, hence, the occurrence of the WMS discrepancy
score. Some other descriptive data have a bearing upon this hypothe
sis also, and it is revealing of the differential sensitivity of the
WAIS and the WMS to brain damage.
The closed head injury group, the localized lesion group, and
the psychiatric control group show mean WAIS scores of 94.6, 98.7,
and 104.2, respectively. All groups are within the average range
(not considering the correction factor of adding eight points to the
WAIS-Revised Form). This is true even of the two groups with ascer
tained brain damage. To have an average WAIS full scale IQ even
within the low average range in a diffuse closed head injury group
might testify to the relative resistance to impairment of overall
WAIS performance in the closed head injured.
85
In contrast, the data indicate that the WMS is affected by the
diffuse closed head injury process. The mean WMS memory quotient is
nearly equal to the WAIS full scale IQ in the localized lesion group
as in the psychiatric unimpaired group. They are about two points
below the WAIS intelligence quotient in both instances. However,
the WMS performance in the closed head injury group is inferior.
The mean WMS quotient in the closed head injury group is nearly 12
points below the mean WMS quotient of the localized trauma group and
nearly 17 points below the mean WMS quotient of the psychiatric un
impaired group.
An interpretation that localized lesion groups do not evidence
low WMS scores is not warranted, however. This group shows the
greatest overall variability in WMS performance with a memory quo
tient range extending from a low of 73 points to a high of 143
points.
Hypothesis 2 conceptually stated that there would be a larger
WMS discrepancy score for the long coma group (more than 5 days)
than in the short coma group (5 days or less). The null hypothesis
of no difference between the group means of the long and short coma
groups was rejected at the .05 level of confidence following an
ANOVA.
The above results are supportive of the hypothesized increase
in the size of WMS discrepancy scores with a longer period of coma.
Additional data portray the effect of length of coma on both the WMS
and the WAIS. The WMS scores and the WAIS scores were both sub
jected to an ANOVA. The results suggested only a tendency for the
86
WAIS mean IQ scores to differ between the short and long coma groups
with the probability level of mean differences being £ = .153. In
contrast, the results of ANOVA of WMS memory quotient scores between
the short and long coma groups yielded a significant probability of
difference, £ = .029.
These results would tend to suggest that while both measures
are affected by length of coma, the WMS is more sensitive to that
indicator of severity.
An inspection of the mean WAIS IQ score and the mean WMS memory
quotient for both the long and short coma groups reveals other des
criptive information. In the short coma group there is about a six-
point discrepancy when the WAIS mean score is compared to the WMS
mean score. When these same scores are compared in the long coma
group, the difference between the WAIS score and the WMS score in
creases to more than 11 points. Also, while the mean score for the
WAIS in the long coma group remains in the low average range (92.6),
the mean WMS score for the same group is a low score (81.2).
The present data regarding the WhS discrepancy score, and its
relationship to closed head injury severity as measured by length of
coma, is convergent. It is supportive of greater WMS discrepancy
with larger periods of coma. It would appear that not only is the
WMS deficient in relation to the WAIS following closed head injury,
but also this WMS discrepancy increases after longer periods of coma.
Data pertaining to Hypothesis 3 suggest that a linear relation
ship exists between the WMS discrepancy score and the neuro
psychological impairment rating in the closed head injury group.
87
However, a number of results indicate that this relationship is not
strong. Moreover, the relationship between the WMS discrepancy
score and the neuropsychological impairment rating is not unique to
the closed head injury group. The correlation between the neuro
psychological impairment rating and the WMS discrepancy score in the
closed head injury group was .32. The same correlation in the local
ized trauma group was .33. The smaller sample size in the localized
trauma group likely prevented a smaller probability than the value
obtained, = .105.
A number of interactions between the psychometric variables
could act to prevent a higher correlation between the neuro
psychological severity rating and the size of the WMS discrepancy
score. There is a higher negative correlation between the WAIS
full scale IQ and the neuropsychological severity rating (-.75 for
both the closed head injury group and the localized lesion group).
Despite the high negative correlation of the WMS memory quotient
(-.73 for the closed head injury group), such a correlation between
the WAIS IQ and the neuropsychological impairment rating obscures
the sensitivity of the WMS to neuropsychological impairment. That
is, as WMS scores decline with severity, so do WAIS scores. In
light of the high negative correlations obtained between the neuro
psychological impairment rating and the WAIS, it is remarkable that
a significant correlation was obtained between the neuropsychological
impairment rating and the WMS discrepancy score.
The above data suggest, therefore, that the WMS discrepancy
score and neuropsychological impairment ratings are somewhat related
but probably measuring quite different processes. This should not
be surprising. A neuropsychological impairment rating is based on
an extremely wide range of deficit types and it would logically show
a higher correlation with the WAIS battery than to a more specific
measure such as the WMS. These equivocal results also likely stem
from the widely varying pattern of deficits and dysfunction found
even within a single patient population. The assumption of a linear
relationship and a high significant correlation between WMS discrep
ancy score and neuropsychological impairment rating presupposes that
these two measures would be consistently sensitive to the same
attributes in every brain damaged subject. While the WMS may be
more uniform in terms of the abilities or attributes tested, cer
tainly the neuropsychological battery will be influenced by a much
wider variety of specific dysfunctions which are not related to WMS
performance.
Hypothesis 3 is only equivocally supported, therefore. While
the size of the WMS discrepancy score is related to neuro
psychological impairment ratings in a moderate and positive way,
this relationship is not unique to closed head injury and is prob
ably overshadowed by stronger linear relationships between the
neuropsychological impairment rating and other more diverse cogni
tive measures.
CHAPTER IV
FINDINGS, CONCLUSIONS, AND RECOMMENDATIONS
Summary
The goal of the present research was to understand the relation
ship between the WMS and the WAIS in the context of brain damage.
It is hypothesized that short term auditory memory, which is empha
sized on the WMS, is more impacted by diffuse brain injuries that
are secondary to closed head injury. Thus, the WMS is presumed to
show a relative deficiency when compared to the WAIS in a group of
closed head injury patients. This research investigated the conten
tion that the WMS memory quotient, when significantly below the WAIS
full scale IQ (WMS discrepancy score), signals the diffuse injury
characteristic of closed head injuries.
Prigatano (1977) has stated that the WMS is a poor screening
device for a wide range of head injuries. At the same time, its
factor structure, consisting of the recall of immediate complex
information and attention and concentration (Kear-Colwell, 1973),
suggests that it might be used in concert with the WAIS to assess
the severity of closed head injury or to screen for the presence of
diffuse closed head injury. Also, it would be an important indi
cator of verbal short-term memory difficulties experienced clini
cally by people suffering closed head injuries. Theoretically,
therefore, WMS discrepancy scores should be related to the head
89
90
injury process as a distinctive feature. That is, it should be re
lated to severity of the head injury as defined by the length of
coma (Brooks, 1976). The WMS discrepancy score would also logically
be related to the severity of closed head injury as defined by neuro
psychological severity indicators. Its increasing use as a screen
ing technique, both singly and in neuropsychological batteries, war
rants further knowledge about these relationships. Thirdly, to
assume that deficient WMS performance in comparison to WAIS perform
ance is a distinctive factor of closed head injury, WMS deficits
should be greater for closed head injury groups when measured
against comparison populations.
The use of WMS in the context of neuropsychological evaluations
is increasing (Smith, 1983). Therefore, the WMS discrepancy score
was studied as a variable which is relevant to a number of psycho
metric dimensions and patient types. The fact that the WMS was de
veloped to be a memory test which could be directly comparable to the
WAIS full scale IQ score (Wechsler, 1945) adds to the value of under
standing the conditions under which WMS performance will vary. Al
though the WMS is known to be deficient among closed head injury
patients (Prigatano, cited in Prigatano, 1978), the exclusivity of
this pattern of discrepancy is less well established. In summary,
the hypothetical problem of deficient WMS performance in relation to
WAIS performance is important from the standpoint of understanding
the types of head injury in which these scores are comparable and
what parameters of brain injury or dysfunction underlie inferior WMS
performance. If the WMS alone is ineffective as a screen for brain
91
damage, then possibly the WMS discrepancy score might be used as a
brain damage screen, at least in closed head injury populations.
The sample population for this research consisted of 115
patients who had undergone complete neuropsychological evaluations
but who had also been given the WMS. Their neurological histories
were used to divide them into three groups according to category and
type of trauma. They were divided into a rather small localized
group of patients who had undergone an ascertained brain trauma
secondary to a stroke, aneurysm, or tumor; compared to a group who
had suffered a closed head injury; compared to a group of psychiat
ric controls who were referred for the suspicion for brain dysfunc
tion but who developed normal neuropsychological battery test re
sults.
The statistical analysis used to test Hypothesis 1 was an ANOVA
to assess any differences between group means of the psychiatric
unimpaired group, the localized lesion group, and the closed head
injury group. Analysis of Hypothesis 2 was also an ANOVA to test
for differences between group means of the short coma group and the
long coma group. The statistical analysis of Hypothesis 3 was
carried out utilizing the Pearson product-moment correlation
coefficient.
Findings
All three null hypotheses were rejected and the theoretical
hypotheses for the WMS discrepancy score were supported. The re
sults suggest that it is at least a common feature of closed head
92
injury and it may be a marker of closed head injury so far as it is
related to severity. Hypothesis 1, which theoretically predicted
larger WMS discrepancy scores in the closed head injury group than
in the unimpaired psychiatric group or the localized lesion group,
was supported. Hypothesis 2, which conceptually predicted larger
WMS discrepancy scores for the long coma group, was also supported.
Hypothesis 3 predicted a significant and positive correlation be
tween the size of the WMS discrepancy score and the magnitude of a
neuropsychological impairment rating. While this hypothesis was
technically supported, the strength of the finding was substantially
weakened by the low order of the correlation and the fact that the
localized lesion group showed a similar correlation. Neuro
psychological impairment ratings were interpreted as being too
diverse in range of measurement to correlate highly with a singular
condition verbal short-term memory deficiency.
Additional statistical analyses were carried out which sub
jected scores on the WAIS and the WMS individually to an ANOVA pro
cedure. It was found that there were significant differences in the
dimension of WAIS full scale intelligence quotient only between the
closed head Injury group and the unimpaired psychiatric group. How
ever, on the WMS memory quotient measure, the closed head injury
group was significantly different from both the localized trauma
group and the psychiatric unimpaired group.
It was discovered that the WMS discrepancy score showed a sig
nificant correlation with the neuropsychological impairment rating
in the closed head injury group. These findings are inconclusive,
93
however, due to the much higher negative correlations obtained be
tween the neuropsychological impairment rating and the WAIS and WMS
in both the localized lesion and closed head injury groups. Overall,
the findings are summarized to suggest that the WMS discrepancy
score is found to persist to a greater degree in closed head injury
populations. This was true despite the fact that the WAIS also
shows the greatest level of impairment in the population of closed
head injury subjects used in this study.
Conclusions
In concept, this study has been directed toward affirming the
conceptual hypothesis that the WMS, as a measure of both short-term
audio-verbal memory and "fluid" verbal abilities, would be particu
larly sensitive to the diffuse effects of closed head injury. More
specifically, it was hypothesized that the WMS might be used in con
cert with the WAIS to identify those individuals who suffer the
types of injuries which put them at high risk for deficits which
involve fluid verbal abilities and the other performance factors of
the WMS which include short-term verbal memory, attention, concentra
tion, and the freedom from distractability factor. It was also
hypothesized that closed head injury as a pathological process would
involve diffuse cerebral disruption and cortex/midbrain disruptions
which are important in causing short-term memory impairment. The
WMS discrepancy score is also investigated due to the supposed com
parability of the scores in normals (Wechsler, 1945). In diffuse
head injuries, as the literature has supported, there tends to be a
94
predominance of performance Intelligence quotient score deficits in
closed head injuries (Mandleberg, 1975; Mandleberg & Brooks, 1975).
The WMS was hypothesized to be a more sensitive indication of fluid
as opposed to crystallized verbal abilities. The use of the WMS
could therefore be evaluated by understanding the manner in which
the WMS might be deficient to WAIS performances. Additionally, it
might solve the problem of measuring fluid verbal capabilities and
associated short-term verbal memory problems. It therefore could
serve as a valuable addition to the WAIS which measures mostly
crystallized verbal abilities which are refractory to diffuse head
injuries (E. W. Russell, 1980).
In the present study the relationship of head injury to the WMS
discrepancy score was established. The relationship of the WMS dis
crepancy score to length of coma was clearly supported. The present
study only differentiates short from long coma groups, but it does
nicely corroborate the findings of Prigatano (cited in Prigatano,
1978, p. 823). The discrepancy of the WMS from the WAIS in increas
ing severity as defined by length of coma is not only supported by
Prigatano's study but also the average WMS discrepancy score of 10
points established in his study is highly consistent with the pres
ent study. These findings do support the general conceptual hypothe
sis that increasing severity of diffuse closed head injury results
in a disproportionally greater decrease in the more fluid and memory
based aspects of verbal processing as measured by the WMS.
The relationship between neuropsychological severity scores and
WMS discrepancy scores was investigated to further support the
notion that, with increasing severity, WMS discrepancy scores would
be more characteristic of closed head injuries as compared to local
ized head injuries and a psychiatric control group. This hypothesis
was supported but seems considerably weaker. The closed head injury
group performance showed a significant correlation between the WMS
discrepancy score and neuropsychological severity ratings. While
these findings certainly are in the expected direction, they are not
strong in supporting conclusively that WMS discrepancy is unique to
closed head injury with increasing severity. In fact, the localized
lesion group also showed correlation between WMS discrepancy score
and neuropsychological severity score approaching significance.
Difficulties in sample size in the present study may have prevented
more conclusive results regarding this hypothesis.
Possibly the most conclusive evidence in the present study is
the obtained significant differences which discriminate the closed
head injury from both the localized trauma group and the unimpaired
control group on the measure of WMS discrepancy score. While the
closed head injury group was different from both of the comparison
groups, the localized trauma group and the unimpaired control group
were not different from each other. Therefore, except for possibly
the most severe of localized lesions, group differences in the dis
crepancy scores are most notable. It is possible that as localized
pathological process such as tumors or strokes become more general
ized they begin to overlap on the dimension of WMS discrepancy with
the diffuse head injuries. This may be due to multiple lesion sites,
the more widespread involvement of the cerebral cortex, or by the
radiating effects of more widespread brain damage. This observation
certainly would be in line with the observation of Smith (1966), who
found that 65% of left hemisphere lesion patients showed the contra
dictory findings of having higher verbal intelligence quotient
scores.
However, the present study has much less to say conceptually
about memory difficulties or WMS discrepancy in regard to localized
lesions. The fact that WMS discrepancy was less typical of the
group does not obviate the fact that some localized lesion subjects
showed markedly inferior WMS scores. The use of the WMS discrepancy
scores with localized lesion groups will obviously remain a matter
of individual clinical neuropsychological assessment. Given the
extremely large range of WMS scores in that group, it would demand
that the neuropsychology clinician be sensitive to the potential
variance among localized lesion groups on the dimension of memory
function. The fact that localized lesion patients as a group do not
show a regularly occurring pattern of short-term memory deficiency
only requires greater vigilance to those numbers of cases who may
show such a condition. These observations tend to reinforce the
importance of the present findings in terms of using a technique ir
addition to the WAIS to screen for deficits in the area of memory
processes regardless of the site of lesion.
Another observation which can be derived from results of this
study pertains to the differences in deficit pattern between the
closed head injury group and the localized lesion group. The local
ized lesion group showed the greatest overall neuropsychology
97
impairment rating. Despite this fact, they showed higher WAIS
scores and much smaller WMS discrepancy scores. It is tempting to
speculate that the Halstead-Reitan Neuropsychological Battery, by
virtue of its specific sensitivity to cortical areas, may under
estimate or overestimate total brain dysfunction depending upon the
type of lesion process. If that were true, then it could be pos
sible that the neuropsychological impairment rating may under
estimate diffuse lesions and overestimate focal lesions in terms of
their total impact upon a patient's adjustment and cognitive capaci
ties. While this possibility is a matter for further research, it
suggests clinical caution when using neuropsychology impairment
ratings as estimates of difficulty in living.
A number of important features emerge from the present findings.
Despite the fact that WAIS performance was most devastated in the
closed head injury group, the discrepancy effect remained. That is,
although closed head injury patients showed the greatest disruption
of intellectual processes as measured by the WAIS, they still show
the largest relative WMS discrepancy effects.
This researcher believes this effect further underscores the
importance of assessing interaction effects among subtests when
carrying out neuropsychological evaluations. This observation cer
tainly is not new to neuropsychology. The field has long depended
upon the differential sensitivity of subtests to estimate brain dys
function. This is most true of those subtests constructed to be
sensitive to specific sites of dysfunction within the cortex. The
addition of memory testing to neuropsychological batteries, however,
98
provides a new dimension. That is, it supplies the examiner with
information about the rate and nature of information processing and
retention. Memory processing and other rapid integrative activity
are most frequently not the product of activity within a well local
ized field of the cortex. The interactions between neuro
psychological subtests which test more localized cortical functions
and those which are more globally sensitive to diffuse brain dys
function need to be compared. WMS and WAIS performance comparisons
may therefore be most important not because of any specific ability
to discriminate memory deficiency; rather, they may provide a set of
interactions or comparisons which can be used to discriminate local
ized from diffuse injury processes or even to estimate subcortical
brain damage effects which have a bearing upon memory, arousal,
attention, and other fluid adaptive capacities.
The present study discloses closed head injury to have an aver
age WMS discrepancy score of almost 9.5 points. Other authors
(Milner, 1975; Prigatano, 1978; Quadfasel & Pruyser, 1955) have
reported, from studies of groups with lesions known to produce mem
ory deficits, discrepancy scores of 12 points, 10 points, anti 11
points, respectively. Zaidel and Sperry (1974) reported a WMS dis
crepancy score of nearly 12 points in a group of patients with known
short-term memory impairments secondary to hippocampal commissurot
omy. This places the average discrepancy score for the closed head
injury group in the present study (9.33) to be close to those groups
with known memory producing lesions. The discrepancy of the local
ized trauma group and the unimpaired psychiatric control group were
99
1.5 and 2.0 points, respectively. They do not approach the level of
WMS discrepancy required for defined memory impairment as suggested
by the previous authors and further confirm the overall tenant of
the present study.
The findings of the present study have some important implica
tions for clinical assessment of brain dysfunction following closed
head injury. Although the population of closed head injured sub
jects used in the present study were not selected on the basis of
known memory dysfunction, they manifested a distinct tendency to
resemble memory impaired groups on the WMS discrepancy score measure.
This was true of the present closed head injury population despite
the fact that it contained a substantial percentage of "mild" head
injuries as defined by length of coma. Clinical neuropsychologists
cannot afford to ignore the dimension of relative short-term memory
impairment, especially when evaluating closed head injury. This
implies that an absence of a memory measure, which could be used to
contrast with intelligence and neuropsychological findings, would
jeopardize the applicability of neuropsychological battery findings.
The present study does not imply that only the WMS be used as a
measure of short-term memory. Rather, it would suggest that an
instrument be developed or used which requires rapid assimilation of
new information. Also, the information to be assimilated should
vary according to modality and complexity.
One interpretive danger to the present study consists of the
fact that it has failed to control for the amount of time following
injury. Although most of the subjects in the present study are past
100
the 6 months critical period of restitution (Mandleberg, 1975), it
should be recognized that a very few individuals with less severe
closed head injuries had relatively recent injuries which would tend
to emphasize abnormally low WMS in those groups who were still re
covering function.
Recommendations
The following recommendations include suggestions for both
further research and clinical practice. Initially some points will
be discussed which may have some importance for further research and
then a general discussion of the relationship of the present find
ings to clinical practice will be offered.
Future research should be aimed at avoiding methodological
weaknesses in the present study as well as being directed toward
extending the understanding of the WMS discrepancy score. A criti
cism of the present study might be a failure to control in an exact
ing way for the site, location, and severity of lesion in the local
ized lesion group. The availability of more definitive data which
describes the exact site and location of lesions in this group would
have provided valuable additional data regarding the impact of those
specific lesion sites on the discrepancy score. It is possible that
a different magnitude of WMS discrepancy score will occur in right
sided as opposed to left sided lesions. Also, there may be further
relationships between the exact localization of these lesions within
a functional territory of the cortex and the discrepancy score. For
example, temporal lobe lesions could be compared with frontal or
101
posterior lobe lesions in a way that furthers the understanding of
ways in which some localized lesions might be more similar to closed
head injury or diffuse lesions in terms of discrepancy scores and
memory deficits. Such research obviously requires the availability
of large numbers of patients on whom data can be gathered in a
highly systematic and ongoing fashion. This approach, however,
appears to be indispensable for highly specific research involving
the clinical correlates of brain lesions.
It would be important to investigate the discrepancy score on
subjects with known levels of premorbid functioning. This could
provide a description of the relative deficits on the WMS perform
ance or an increase in the WMS discrepancy score following any spe
cific lesion type. Again, it requires data collection which is
specific and concurrent rather than a post hoc analysis.
The present study found rather conclusive evidence that severity
of closed head injury as defined by length of coma correlates highly
with WMS discrepancy score. That is, as length of coma increases,
so does relative WMS discrepancy score. It would be a substantial
clinical benefit to record specific length of coma or posttraumatic
amnesia such as number of days and to relate increasing length of
coma to WMS discrepancy. While the effects of generally worsening
WMS performance with increasing severity of head injuries has been
documented by Mandleberg and Brooks (1975), it will be important to
know if there are specific bench marks in terms of length of coma or
posttraumatic amnesia which cause greater corresponding increases in
WMS discrepancy. The current suggestion of coma state lasting 5 days
102
or more as an indicator of severity producing memory results cer
tainly seems to be established.
It would be important for further research to aim at the under
standing of the relationship between closed head injuries and the
most severe types of localized lesion. Overall, the localized
lesion group and the closed head injury group were dissimilar for
purposes of the WMS score and the WMS discrepancy score. At the
same time, many localized lesion individuals showed very poor WMS
performance. This may suggest that, as localized lesions become
more involved, they more closely resemble closed head injury in
terms of memory performance. A similarity in performance on the WMS
discrepancy score between closed head injury and more severe local
ized lesions would not only be an important clinical finding but
also would be theoretically important for understanding the suffi
cient pathophysiological causes of memory impairment.
This study continues to be heuristic in supporting the notion
that a fluid verbal factor is present and may be in fact measurable
by such tests as the WMS. While the concept of fluid as opposed to
crystallized abilities in both verbal and visually based skills has
long been theorized, the concept has been most recently applied in
test construction by Kaufman and Kaufman (1983). The effects of
closed head injuries on fluid and crystallized intelligence deserve
further study.
In addition to these more general research suggestions, spe
cific research ideas are generated by the findings in the study.
The WMS discrepancy score has been shown to be psychometrically
associated with closed head injury. It would enhance the WMS dis
crepancy score concept and allow more specific clinical description
if modality specific comparisons between the WMS and WAIS could also
be made. For example, the WAIS verbal IQ score could be contrasted
with the paired associates or logical memory subtests. Similarly,
figural memory could be contrasted with the WAIS performance IQ
score. Such a study might provide both a visual and verbal memory
discrepancy score. This could provide an index of "fluid" and
"crystallized" abilities in both visual and verbal spheres and pro
vide important additional information in the context of a neuro
psychological assessment. To best carry out such a research project,
the exact locus of lesion should be specified in control groups. In
this way, more specific comparisons can be made regarding the effect
of lesions on discrepancy scores.
An immediate problem which needs to be solved if WMS discrep
ancy scores are to be useful is the problem of establishing WMS and
WAIS-Revised Form (WAIS-R) comparisons. As discussed, the WAIS-R
understates IQ level in comparison to the older WAIS. A more broad
study of WAIS-R/WMS differences would provide descriptive informa
tion about comparisons between the WAIS-R and WMS in different popu
lations, but more importantly it would help establish a new WMS
discrepancy score "benchmark" so that the WMS could more predictably
be used with the newer version, the WAIS-R.
The current study focuses almost entirely on short-term memory
processes. In light of the current findings it would be beneficial
to investigate the discrepancy effect utilizing recent memory, or
104
even longer periods of retention. In such a study, WAIS-R perform
ance could be contrasted with short-term memory and then, in turn,
contrasted with memory requiring retention of 30 minutes or more.
This would carry the benefit of measuring more pure memory functions
in addition to performance skills requiring attention and arousal.
E. W. Russell's (1979) multiple scoring system utilizing the WMS
would lend itself nicely to such an investigation.
Many computer software programs are currently marketed to offer
cognitive retraining that is claimed to improve memory capabilities
in the head injured patient. The WMS discrepancy could be used in a
test-retest fashion to investigate whether true short-term memory
gains are occurring or whether improved memory performance is really
an aspect of more general cognitive improvement. This study empha
sizes the problem of short-term memory in the closed head injured,
and more specific evaluations of the conditions of "true" memory
improvement in this group are required.
The present study was unconvincing in regard to WMS discrepancy
scores and their correlation with neuropsychological outcomes. It
is likely that the extreme diversity of neuropsychological subtests
which contribute to global impairment ratings will not have a pre
dictable relationship to the WMS discrepancy score. It would remain
desirable to be able to use the WMS discrepancy score in conjunction
with neuropsychological indicators to estimate diffuse or severe
dysfunction. Such comparisons could be made possible by research
which investigates the relationship between the WMS discrepancy
score and those neuropsychological subtests which require integrative
105
ability. Specifically, on the Halstead-Reitan Battery, the Cate
gories Test and the Trails (b) Test might be used in concert with the
WMS discrepancy score as a marker of diffuse head injury. Again,
this research would be most valuable if such inter-test comparisons
are made utilizing the strictest control possible over the type of
lesions which comprise comparison groups. Such specific comparison
of lesion types will be most important in future research aimed at
understanding the complex interactions of neuropsychological assess
ment instruments.
In terms of clinically relevant indications from the present
studies, it clearly shows that clinicians should be very aware of
verbal processing deficits in cases of closed head injury. Moreover,
this is even true in cases where a normal verbal intelligence score
is obtained. That is, the rather regular deficiencies on the WMS
(even with its strong component of audio-verbal processing) suggests
that deficits in verbal or auditory learning may exist as hidden
features which significantly compound the treatment and rehabilita
tion of the diffusely involved closed head injury patient. Specifi
cally, the WAIS alone should not be used as an indicator of restored
verbal capacity.
The study suggests that the actual treatment of the closed head
injured should be oriented toward a reduction in the complexity and
rate of verbal processing demands. The repetition of written lan
guage and the use of visual as well as spoken verbal messages may be
used simultaneously to enhance memory traces for instructional or
therapeuatic purposes. The hidden deficits relative to verbal
106
efficiency and fluency may be responsible for a great deal of the
concreteness and inflexibility and adjustment observed in closed
head injury patients who have difficulty adapting even after normal
verbal intelligence quotient scores are obtained.
The results of the study do not suggest that the WMS discrep
ancy score be used as a screening tool for organic brain dysfunction.
While the score is certainly more descriptive than the WMS scale
alone, it would appear that it is much too variable to be used as a
reliable indicator of the presence of a lesion. This would appear
to be especially true in those categories of brain dysfunction that
do not involve closed head injury or diffuse dysfunction.
If a shortened screening battery for closed head injury patients
were required, the WMS discrepancy score would be best used along
with a global test of impairment, a test of spatial relations, and
an aphasia screening tool. In this manner, intelligence, short-term
memory, fluid processing capabilities, visual spatial organization,
and language functions could contribute to a functional description
of deficits. Such a combination of instruments still could not be
used to infer the site of a lesion, however.
The WMS should probably be included in most batteries attempt
ing to generate information relevant to rehabilitation or treatment.
This is obvious due to the immediately preceding comments about
hidden deficiencies in verbal fluency and acquisition. Prigatano's
(1978) general criticism of the WMS as a general test of memory may
still be true. However, with E. W. Russell's (1979) complex scoring
system, it may still be possible to assess immediate and fluid
verbal processes as well as recent memory of up to 30 minutes. Addi
tionally, it would be possible to gauge the type of memory diffi
culties which are specific to an individual or to achieve informa
tion regarding modality specific memory deficits. This approach may
solve the problem of addressing fluid as opposed to crystallized
verbal functions while also retaining the capability of measuring
qualitative disruptions in the memory process proper and enabling
the clinician to describe more specifically the type of breakdown
in memory processes in the individual patient.
It is not likely that the results of the present study are
strong enough to indicate that the WMS discrepancy score can be used
in individual patients to describe the severity of closed head in
juries. Too many factors are involved in the idiosyncratic patho
physiological makeup of each injured patient. However, the system
atic inclusion of the WMS in cases which use neuropsychological
batteries or the WAIS for screening purposes might provide an addi
tional valuable point of analysis which can be combined with other
psychometric features and clinical judgment in managing the diagno
sis and treatment following head injury.
APPENDIX
108
109
Raw Scores for WAIS, WMS, Neuropsychological

Impairment Rating and Length of
Coma by Group
Neuropsych. Neuropsych
ID WAIS WMS impairment ID WAIS WMS impairment
rating rating
Psychiatric unimpaired group 37 104 105 .50

38 99 105 1.09
1 105 76 1.16 39 98 90 1.08
2 114 112 .92 40 108 106 .75
3 123 118 .42 41 118 120 .58
4 113 126 .67 42 99 80 1.25
5 106 96 1.58 43 87 114 1.50
6 111 99 .83 44 121 135 1.33
7 109 105 1.08 45 95 108 1.08
8 124 140 .50
9 92 92 1.25 Localized lesion group
10 110 99 .67
11 99 106 1.17 46 81 80 2.50
12 89 92 1.13 47 126 132 .92
13 93 70 1.17 48 104 77 1.60
14 118 112 .67 49 129 143 1.00
15 109 87 .83 50 85 77 3.27
16 100 118 .92 51 96 81 2.58
17 113 129 .58 52 121 143 1.25
18 92 105 1.42 53 93 103 1.67
19 108 108 .67 54 109 112 1.25
20 93 86 1.33 55 96 100 2.92
21 121 124 .64 56 90 73 3.08
22 103 105 1.00 57 96 87 2.42
23 111 103 .75 58 95 79 2.75
24 102 87 .67 59 91 94 3.17
25 108 97 1.00 60 100 97 2.33
26 116 114 1.08 61 92 84 2.25
27 102 101 .50 62 108 90 2.88
28 101 96 1.00 63 91 94 1.83
29 87 92 1.25 64 106 93 2.08
30 110 106 1.58 65 105 108 2.50
31 101 99 1.50 66 105 129 2.63
32 84 66 1.33 67 81 77 2.75
33 106 106 .83 68 89 89 3.11
34 99 100 1.17 69 100 116 2.50
35 98 80 1.33 70 80 73 3.11
36 91 84 1.16
110
Neuropsych. Length
ID WAIS WMS Impairment of
rating coma
Closed head Injury group
71 120 116 1.33 Short

72 107 92 2.50 Long
73 95 90 1.33 Short
74 124 105 1.42 Long
75 111 110 1.08 Long
76 111 106 .58 Short
77 86 97 2.83 Short
78 107 112 1.58 Long
79 81 64 2.16 Short
80 86 92 2.17 Long
81 92 92 1.67 Long
82 93 69 2.50 Long
83 101 94 1.75 Short
84 88 73 1.50 Long
85 95 99 1.80 Short
86 79 62 2.75 Long
87 63 52 3.08 Long
88 98 94 1.58 Short
89 94 81 1.50 Long
90 90 83 2.67 Long
91 107 103 1.50 Long
92 92 94 1.50 Short
93 86 67 2.64 Long
94 103 106 1.50 Short
95 108 89 1.00 Short
96 99 99 1.50 Long
97 102 93 1.42 Long
98 106 93 2.16 Short
99 89 72 2.17 Long
100 81 62 3.18 Short
101 90 81 2.25 Short
102 79 57 2.67 Long
103 88 74 1.50 Long
104 114 110 .92 Short
105 92 80 2.17 Long
106 97 70 1.92 Long
107 83 73 2.56 Long
108 88 64 3.17 Long
109 97 89 1.67 Long
110 92 70 2.33 Long
111 75 61 3.33 Long
112 100 96 1.25 Short
113 84 84 2.75 Long
114 102 99 2.00 Long
115 83 77 2.33 Short
BIBLIOGRAPHY
Bachrach, H., & Mintz, J. The Wechsler Memory Scale as a tool for
the detection of mild cerebral dysfunction. Journal of Clinical
Psychology, 1974, _30, 58-60.
Becker, B. Intellectual changes after closed head injury. Journal

of Clinical Psychology, 1975, _31, 307-309.
Benton, A. Behavioral consequences of closed head injury. In

National Institute of Neurological and Communicative Disorders
and Stroke, Central nervous system trauma research status report,
1979. Pp. 220-231. (Mimeo)
Black, F. W. Cognitive and memory performance in subjects with

brain damage secondary to penetrating missile wounds and closed
head injury. Journal of Clinical Psychology, 1973, 29, 441-442.
Bornstein, R. A. Effects of unilateral lesions on the Wechsler

Memory Scale. Journal of Clinical Psychology, 1982, _38, 389-392.
Brooks, D. N. Memory and head injury. The Journal of Nervous and

Mental Disease, 1972, 155, 350-355.
Brooks, D. N. Long and short term memory in head injured patients.

Cortex, 1975, 11, 329-340.
Brooks, D. N. Wechsler Memory Scale performance and its relation

ship to brain damage after severe closed head injury. Journal
of Neurology, Neurosurgery, and Psychiatry, 1976, _39, 593-601.
Campbell, D. T., & Stanley, J. C. Experimental and quasi-

experimental designs for research. Chicago: Rand McNally,
1966.
Cattell, R. B. The measurement of adult intelligence. Psychological

Bulletin, 1943, 3, 153-193.
Cohen, J. Wechsler Memory Scale performance of psychoneurotic,

organic, and schizophrenic groups. Journal of Consulting
Psychology, 1950, 14, 371-375.
Conkey, R. C. Psychological changes associated with head injuries.

Archives of Psychology, 1938, 232, 1-62.
Davis, L. J., & Swenson, W. M. Factor analysis of the Wechsler

Memory Scale. Journal of Clinical and Consulting Psychology,
1970, 35, 430.
Ill
Dujovne, B. E., & Levy, B. I. The psychometric study of the
Wechsler Memory Scale. Journal of Clinical Psychology, 1971,
27, 351-354.
Dye, C. L. Factor structure of the Wechsler Memory Scale in an

older adult population. Journal of Clinical Psychology, 1982,
38, 163-166.
Fields, F. R. Relative effects of brain damage on the Wechsler

memory and intelligence quotients. Diseases of the Nervous
System, 1971, 32, 673-675.
Filskov, S. B., & Goldstein, S. G. Diagnostic validity of the

Halstead-Reitan battery. Journal of Consulting and Clinical
Psychology, 1974, 42^, 382-388.
Fitzhugh, K. B., & Fitzhugh, L. C. WAIS results for S's with long
standing, chronic, lateralized and diffuse cerebral dysfunction.
Perceptual and Motor Skills, 1964, 19^, 735-739.
Fitzhugh, K. B., Fitzhugh, L. C., & Reitan, R. M. Psychological

deficits in relation to acuteness of brain dysfunction. Journal
of Consulting Psychology, 1961, 25, 61-66.
Fitzhigh, K. B., Fitzhugh, L. C., & Reitan, R. M. Wechsler-Bellevue

comparisons in groups with "chronic" and "current" lateralized and
diffuse brain lesions. Journal of Consulting Psychology, 1962,
26, 306-310.
Hall, J. C., & Toal, R. Reliability (internal consistency) of the

Wechsler Memory Scale and correlation with the Wechsler-Bellevue
Intelligence Scale. Journal of Consulting Psychology, 1957, 21,
131-135.
Halstead, W. C. Brain and intelligence: A quantitative study of

the frontal lobes. Chicago: University of Chicago Press, 1947.
Heilbrun, A. B. Psychological test performance as a function of

lateral localization of cerebral lesion. Journal of Comparative
and Physiological Psychology, 1956, 49^, 10-14.
Horn, J. L. Human abilities: A review of research and theory in

the early 1970's. Annual Review of Psychology, 1976, 27, 437-
485.
Howard, A. R. Diagnostic value of the Wechsler Memory Scale with

selected groups of institutionalized patients. Journal of
Consulting Psychology, 1950, 14, 376-380.
Kaufman, A. S., & Kaufman, N. L. Kaufman assessment battery for
children: Interpretive manual. Circle Pines, MN: American
Guidance Service, 1983.
Kear-Colwell, J. J. The structure of the Wechsler Memory Scale and

its relationsuip to brain damage. British Journal of Social and
Clinical Psychology, 1973, 12, 384-392.
Kear-Colwell, J. J. The structure of the Wechsler Memory Scale.

Journal of Clinical Psychology, 1977, J33, 483-485.
Kear-Colwell, J. J., & Heller, M. A normative study of the Wechsler

Memory Scale. Journal of Clinical Psychology, 1978, _34, 437-442.
Kear-Colwell, J. J., & Heller, M. The Wechsler Memory Scale and

closed head injury. Journal of Clinical Psychology, 1980, 36,
782-787.
Kerlinger, F. N. Foundations of behavioral research (2nd ed.).

Chicago: Holt, Rinehart and Winston, 1973.
Kljajic, I. Wechsler Memory Scale indices of brain pathology.

Journal of Clinical Psychology, 1950, 14, 376-380.
Ladd, C. E. WAIS performances of brain damaged and neurotic adults.

Journal of Clinical Psychology, 1964, 20, 114-117.
Lansdell, H., & Smith, F. J. Asymmetrical cerebral function for two

WAIS factors and their recovery after brain injury. Journal of
Consulting and Clinical Psychology, 1975, 4_3, 923.
Levin, H. S., Benton, A. L., & Grossman, R. G. Neurobehavioral

consequences of closed head injury. New York: Oxford University
Press, 1982.
Levin, H. S., & Grossman, R. G. Behavioral sequelae of closed head

injury: A quantitative study. Archives of Neurology, 1978, 35,
720-727.
Levin, H. S., Grossman, R. G., & Kelly, P. J. Short-term recogni

tion memory in relation to severity of head injury. Cortex,
1976, 12, 175-182.
Libb, J. W., & Coleman, . Correlations between the WAIS and

revised BETA, Wechsler Memory Scale and Quick Test in a voca
tional rehabilitation center. Psychological Reports, 1971, 29,
863-865.
Logue, P. E., & Allen, K. WAIS: Predicted category test scores

with the Halstead Neuropsychological Battery. Perceptual and
Motor Skills, 1971, 33, 1095-1096.
114
Mandleberg, I. A. Cognitive recovery after severe head injury:

2. Wechsler Adult Intelligence Scale during post-traumatic
amnesia. Journal of Neurology, Neurosurgery, and Psychiatry.
1975, 38, 1127-1132.
Mandleberg, I. A. Cognitive recovery after severe head injury:

3. WAIS Verbal and Performance IQs as a function of post-
traumatic amnesia duration and time from injury. Journal of
Neurology, Neurosurgery, and Psychiatry, 1976, _39, 1001-1007.
Mandleberg, I. A., & Brooks, D. N. Cognitive recovery after severe

head injury: 1. Serial testing on the Wechsler Adult Intelli
gence Scale. Journal of Neurology, Neurosurgery, and Psychiatry,
1975, 38, 1121-1126.
Milner, B. Psychological aspects of focal epilepsy and its neuro

surgical management. Advances in Neurology, 1975, 299-321.
Ommaya, A. K., & Gennarelli, T. A. Cerebral concussion and trau

matic unconsciousness: Correlation of experimental and clinical
observations on blunt head injuries. Brain, 1974, JF7, 633-654.
Parsons, 0. A., & Prigatano, G. P. Memory function in alcoholics.

In I. M. Birnbaum & E. S. Parker (Eds.), Alcohol and human
memory. Hillsdale, NJ: Elbraum Associates, 1977.
Parsons, 0. A., Vega, A., & Burn, J. Different psychological

effects of lateralized brain damage. Journal of Clinical and
Consulting Psychology, 1969, _33, 551-557.
Prigatano, G. P. The Wechsler Memory Scale is a poor screening test

for brain dysfunction. Journal of Clinical Psychology, 1977, 33,
772-777.
Prigatano, G. P. Wechsler Memory Scale: A selective review of the

literature. Journal of Clinical Psychology, 1978, 34, 816-832.
Prigatano, G. P. Personal communication, July 19, 1983.
Quadfasel, A. F., & Pruyser, P. W. Cognitive deficit in patients

with psychomotor epilepsy. Epilepsia, 1955, 4^(3), 80-90.
Reitan, R. M. An investigation of the validity of Halstead's

measures of biological intelligence. AMA Archives of Neurology
and Psychiatry, 1955, 7 3 , 28-35. (a)
Reitan, R. M. The relation of the Trail Making Test to organic

brain damage. Journal of Consulting Psychology, 1955, lj?, 393-
395. (b)
115
Reitan, R. M. Investigation of relationships between "psychometric"

and "biological" intelligence. Journal of Nervous and Mental
Diseases, 1956, 123, 536-541.
Reitan, R. M. Qualitative versus quantitative mental changes

following brain damage. Journal of Psychology, 1958, 46, 339-
346. (a)
Reitan, R. M. Validity of the Trail Making Test as an indicator of

organic brain damage. Perceptual and Motor Skills, 1958, 8, 271-
276. (b)
Reitan, R. M. The comparative effects of brain damage on the

Halstead Impairment Index and the Wechsler-Bellevue Scale.
Journal of Clinical Psychology, 1959, 15, 281-285. (a)
Reitan, R. M. Correlations between the Trail Making Test and the

Wechsler-Bellevue Scale. Perceptual and Motor Skills, 1959, 9,
127-130. (b)
Reitan, R. M. Psychological assessment of deficits associated with

brain lesions in subjects with and without subnormal intelligence.
In J. L. Khanna (Ed.), Brain damage and mental retardation: A
psychological evaluation. Springfield, IL: Charles C. Thomas,
1967.
Reitan, R. M. Trail Making Test results for normal and brain

damaged children. Perceptual and Motor Skills, 1971, 33, 575-
581.
Russell, E. W. WAIS factor analysis with brain-damaged subjects

using criterion measures. Journal of Consulting and Clinical
Psychology, 1972, 39, 133-139.
Russell, E. W. A multiple scoring method for the assessment of

complex memory functions. Journal of Clinical and Consulting
Psychology, 1975, 4^3, 800-809.
Russell, E. W. Three patterns of brain damage on the WAIS. Journal

of Clinical Psychology, 1979, 35, 611-620.
Russell, E. W. Fluid and crystallized intelligence: Effects of

diffuse brain damage on the WAIS. Perceptual and Motor Skills,
1980, 51, 121-122.
Russell, E. W. The pathology and clinical examination of memory.

In S. B. Filskov & T. J. Boll (Eds.), Handbook of Clinical
Neuropsychology. New York: John Wiley and Sons, 1981. Pp. 287-
319.
Russell, E. W. Factor analysis of the revised Wechsler Memory Scale
tests in a neuropsychological battery. Perceptual and Motor
Skills, 1982, 54, 971-974.
Russell, E. W., Neuringer, C., & Goldstein, G. Assessment of brain

damage: A neuropsychological key approach. New York: John
Wiley and Sons, Wiley-Interscience, 1970.
Russell, W. R. Cerebral involvement in head injury. Brain, 1932,

35, 549-603.
Schacter, D. L., & Crovitz, H. F. Memory function after closed head

injury: A review of the quantitative research. Cortex, 1977,
13, 150-176.
Silverstein, M. L., Rosenbaum, G., & Rennick, P. M. Decay and

interference processes in short-term retention of normal and
brain damaged patients. Journal of Clinical Psychology, 1979,
40, 86-92.
Simpson, C. D., & Vega, A. Unilateral brain damage and patterns of

age-corrected WAIS sub test scores. Journal of Clinical
Psychology, 1971, 2J_, 204-208.
Smith, A. Certain hypothesized hemispheric differences in language

and visual functions in human adults. Cortex, 1966, 2^, 109-126.
Smith, A. Personal communication, July 1983.
Todd, J., Coolidge, F., & Satz, P. The Wechsler Adult Intelligence
Scale discrepancy index: A neuropsychological evaluation.
Journal of Consulting and Clinical Psychology, 1977, 45, 450-454.
Vega, A., & Parsons, 0. A. Cross-validation of the Halstead-Reitan

tests for brain damage. Journal of Consulting Psychology, 1967,
31, 619-625.
Victor, M. Angevine, J., Mancall, E. L., & Fisher, C. M. Memory

loss with lesions of hippocampal formation. Archives of
Neurology, 1961, _5, 26-45.
Victor, M., Herman, K., & White, E. E. A psychological study of

Wernicke-ICorsakoff syndrome. Quarterly Journal of Studies on
Alcohol, 1959, 20, 467-479.
Vogt, A. T., & Heaton, R. K. Comparison of Wechsler Adult Intelli

gence Scale indices of cerebral dysfunction. Perceptual and
Motor Skills, 1977, 45, 607-615.
Wechsler, D. The measurement of adult Intelligence. Baltimore:
Williams and Wilkins, 1944.
Wechsler, D. A standardized memory scale for clinical use. Journal

of Psychology, 1945, _19, 87-95.
Wechsler, D. The measurement and appraisal of adult intelligence.

Baltimore: Williams and Wilkins, 1958.
Wechsler, D. Wechsler Adult Intelligence Scale-Revised (Manual).

New York: The Psychological Corporation, 1981.
Wheeler, L. Complex behavioral indices weighted by linear dis

criminant functions for the prediction of cerebral damage.
Perceptual and Motor Skills, 1964, 19, 907-923. (Monograph
Supplement)
Wheeler, L., Burke, C. J., & Reitan, R. M. An application of dis

criminant functions to the problem of predicting brain damage
using behavioral variables. Perceptual and Motor Skills, 1963,
16, 417-440. (Monograph Supplement)
Wheeler, L., & Reitan, R. M. Discriminant functions applied to the

problem of predicting cerebral damage from behavioral tests: A
cross validation study. Perceptual and Motor SkilJs, 1963, 16,
681-701.
Woo-Sam, J. Lateralized brain damage and differential psychological

effects: Parsons et al., re-examined. Perceptual and Motor
Skills, 1971, 33, 259-262.
Zaidel, D., & Sperry, R. W. Memory impairment after commissurotomy

in man. Brain, 1974, 9_7, 263-272.
Zimmerman, S. F., Whitmyre, J. W., & Fields, F. R. Factor analytic

structure of the Wechsler Adult Intelligence Scale in patients
with diffuse and lateralized cerebral dysfunction. Journal of
Clinical Psychology, 1970, 2 6 , 462-465.

Wechsler Adult Intelligence Scale - Wechsler Memory Scale Differenc

Uploaded by

Copyright:

Available Formats

Wechsler Adult Intelligence Scale - Wechsler Memory Scale Differenc

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Wechsler Adult Intelligence Scale - Wechsler Memory Scale Differenc

Uploaded by

Copyright:

Available Formats

Western Michigan University

Dissertations Graduate College

Wechsler Adult Intelligence Scale/Wechsler Memory Scale

Follow this and additional works at: https://scholarworks.wmich.edu/dissertations

Part of the Psychoanalysis and Psychotherapy Commons

This Dissertation-Open Access is brought to you for free

Western Michigan University

Don A. Boyd, Ed.D.

Western Michigan University, 1984

Performance differences between the Wechlser Adult Intelligence

WMS Memory Quotient (WMS discrepancy score) were compared across

three groups consisting of a closed head injury group (N = 45), a

localized lesion group (N^ = 25), and a psychiatric group (N = 45)

who were unimpaired on the Halstead-Reitan Neuropsychological Bat­

tery. Also, WMS discrepancy scores were compared across a long-coma

group and a short-coma group and the correlation coefficient between

the WMS discrepancy score and neuropsychological impairment rating

The WMS discrepancy score was hypothesized to be a marker of

the long-coma patients would show greater WMS discrepancy scores

than a short-coma group, and that WMS discrepancy scores would be

related to severity as defined by impairment on the Halstead-Reitan

Neuropsychological Battery. Results confirmed the expected differ­

long-coma group had significantly larger WMS discrepancy scores.

The correlation between WMS discrepancy score and neuropsychological

they showed the lowest WAIS scores. Results were interpreted as

marker of diffuse closed head injury.

Implications for the use of the WMS discrepancy score and

neuropsychological ratings were discussed. The WMS discrepancy

score may be more descriptive of severity in closed head injury than

neuropsychological impairment ratings. Despite weaknesses in the

use of the WMS as a comprehensive test of memory, it may be of clini­

culties in the area of fluid verbal skills or information processing.

The systematic inclusion of the WMS discrepancy score in the evalua­

tion of closed head injuries seems warranted. Recommendations for

further research were offered.

T his re p ro d u c tio n was made fro m a copy o f a d o c u m e n t sent to us fo r m ic ro film in g .

T h e fo llo w in g e xp lan atio n o f techniques is provided to help c la rify m arkings o r

2. W hen an image on the film is o b lite ra te d w ith a ro u nd black m a rk , it is an

4. For illustrations th a t c an no t be satisfactorily reproduced by xerographic

WECHSLER ADULT INTELLIGENCE SC ALE/W ECHSLER MEMORY SCALE

Western Michigan University Ed.D. 1984

I would like to acknowledge the support and effort of my

chairman, Dr. Robert Hopkins, in helping complete this important

I wish to express appreciation to Dr. Richard Gay, friend and

mentor, who was most helpful in providing direction and creating a

I want to thank my parents, Francis and Alice Boyd, for their

lifelong ability to understand the importance of things that are

fun and interesting.

to understand how much more meaning they give to the completion of

Thanks to my extraordinary wife, Deborah Jo. Her talent,

effort, and love were essential to the completion of the disserta­

tion. Her character has contributed much to the happiness of

others around her.

LIST OF T A BLES .............................................. v

I. THE PROBLEM AND THE BACKGROUND..................... 1

Review of Literature ........................... 4

Definition of Closed Head Injury .............. 5

Definition of Memory Processes ............... 7

Closed Head Injury and Memory Impairment . . . . 9

who were unimpaired on the Halstead-Reitan Neuropsychological Bat

Neuropsychological Battery. Results confirmed the expected differ

use of the WMS as a comprehensive test of memory, it may be of clini

The systematic inclusion of the WMS discrepancy score in the evalua

effort, and love were essential to the completion of the disserta

Wechsler Adult Intelligence Scale. More specifically, it is con

below the WAIS intelligence quotient should logically predict diffi

The validity of the WMS as a screening device for memory dys

capability to measure functions other than short-term memory pro

Despite its published shortcomings as a memory test, this re

closed head injury patients. In an outpatient rehabilitation set

inability to show the capabilities for short-term memory and reten

the type and severity of head injury. However, these neuro