"The contribution of neuropsychology to psychiatry"

Abstract (Summary)
OBJECTIVE: Neuropsychological test data are applied with increasing frequency in
research studies and clinical practice in psychiatry. This article addresses three popular
assumptions about neuropsychological test data and describes the limitations and
contributions of neuropsychological assessment of patients with psychiatric disorders.
METHOD: All research articles from major journals in psychiatry and clinical
psychology since 1991 that focused on neuropsychological assessment of psychiatric
patients were reviewed. Other journals and earlier studies were reviewed selectively.
RESULTS: Neuropsychological test data have made significant contributions to the
development of hypotheses about abnormal brain structure and function in patients with
psychiatric disorders, yet many findings from neuropsychological assessments of
psychiatric patients are misinterpreted. The extent to which neuropsychological test data
in psychiatric populations can be interpreted to reflect abnormalities in brain structure
and function is frequently exaggerated, as is the ability of neuropsychological measures
to serve as specific cognitive probes in imaging studies of physiological activation. On
the other hand, the utility of neuropsychological test batteries as measures of the
patterns of cognitive strength and deficit in individuals with specific psychiatric
disorders is frequently underestimated. CONCLUSIONS: In addition to testing models
of regional brain dysfunction in psychiatric disorders, neuropsychological tests can
provide researchers in psychiatry with an improved understanding of the relation
between central cognitive impairments and symptoms and serve to identify cognitive
predictors of course of illness, and they may provide a method for discriminating among
heterogeneous forms of some psychiatric disorders. Clinically, neuropsychological test
data can be used to develop treatment strategies tailored for an individual's specific
cognitive strengths and deficits.

» Jump to indexing (document details)

Full Text
(8468 words)
Copyright American Psychiatric Association Jan 1995

Neuropsychology is generally understood to be the study of the relation between brain

function and behavior (1). Neuropsychological assessment has traditionally focused on
determining specific changes in mental processes in patients with discrete lesions
following normal development. In this manner, neuropsychology in clinical practice has
enabled practitioners to determine the locus of insult or disease as well as the functional
capacities of patients in treatment. Neuroscientists have benefited from human
neuropsychological data, since studies of the relation between brain lesions and specific
patterns of functional deficit have made substantial contributions to understanding the
role of specific neuroanatomic regions in normal mental processes.

There are many applications of neuropsychology in psychiatry, including the

identification of brain lesions in psychiatric patients, the evaluation of cognitive
deterioration over time, and the advancement of theories regarding the neuroanatomic
localization of the symptoms of various psychiatric disorders. The purpose of this article
is to discuss the contributions that neuropsychology and neuropsychological assessment
can make to psychiatry. It is based on a review of all research articles from major
journals in psychiatry and clinical psychology since 1991 that focused on
neuropsychological assessment of psychiatric patients. Other journals and earlier studies
were reviewed selectively. The article considers the validity of three common
assumptions in which neuropsychological data are regarded as supportive of hypotheses
about the specific regions of impairment in specific major psychiatric disorders. It also
emphasizes underutilized practical applications of neuropsychological assessment in
psychiatry. One of the most important contributions of neuropsychological assessment
is that it makes possible an objective evaluation of behavior in the context of the ability
to perform basic tasks. When applied properly, a battery of neuropsychological tests
yields an understanding of the cognitive and behavioral abilities and weaknesses of an
individual or group of individuals. It provides the clinician or investigator with an
objective description of what areas of behavior and cognition are likely to be a problem
for the psychiatric patient and what areas are not. In this manner, neuropsychological
data serve as a window into the everyday mental processes of the psychiatric patient.

ASSUMPTIONS ABOUT NEUROPSYCHOLOGICAL TEST PERFORMANCE

DATA

Neuropsychological test performance data have made significant contributions to the

development of hypotheses about abnormalities of brain structure and function in
patients with several major psychiatric disorders, including schizophrenia, major
depressive disorder, and Alzheimer's disease. While these contributions have helped to
reshape the focus of investigations into the etiology and pathophysiology of these
disorders, findings in this area have often been either exaggerated or underemphasized,
and frequent erroneous conclusions have been made about neuropsychological
assessments. Three of the most common assumptions are addressed here.

ASSUMPTION 1: Neuropsychological Tests Measure Specific Functions, and Poor

Performance on a Single Test Indicates a Specific Neuropsychological Deficit

A frequent conclusion of research reports in psychiatry and psychology is that

performance on a particular neuropsychological test (or very small number of tests) in a
particular group of patients is worse than normal, indicating impairment in the function
that the test measures. This conclusion is drawn even in some of the best research
studies (2-13). Neuropsychological test batteries, when constructed properly, assess a
broad range of functions, including, among others, perception in all sensory modalities,
attention, learning and memory, motor skills, verbal and nonverbal skills,
comprehension and expression of language, spatial abilities, laterality, abstraction, and
"executive" functions. There is great overlap among these categories, and many tasks
include redundant features. For example, both abstraction, as measured by the
interpretation of proverbs, and verbal memory require intact verbal comprehension.
Almost all individual neuropsychological tests of higher mental processes, because of
their inherent complexity, involve a number of different functions, and poor
performance can be attributed to any variety of functional deficits. Formal
neuropsychological test batteries, such as the Halstead-Reitan (14) and the Luria-
Nebraska (15), are designed to measure a broad range of functions with a fixed set of
tests. The Halstead-Reitan is based on a series of tests devised in the late 1940s to
discriminate between patients with frontal lobe lesions and control subjects. The Luria-
Nebraska is an attempt to standardize the assessment procedures used by Luria. These
formal batteries are useful in making general conclusions about the presence of
impairment in individuals without psychiatric disorder, but they are not flexible enough
to allow the investigator to assess the variety of functions that may underlie a
performance deficit on a complex cognitive task.

The level of complexity of some tests of higher mental functions can be seen clearly in a
consideration of the assessment of executive functions with the Wisconsin Card Sorting
Test (16), which was originally developed by Grant and Berg in 1948 and is currently
the most extensively used neuropsychological test in psychiatry research. The standard
version of this task in clinical practice (17) requires patients to sort cards by a series of
principles that they must deduce. The cards vary in the shape, color, and number of
stimuli printed on them. There is an arbitrary correct sorting principle that patients must
learn through feedback from the tester ("right" or "wrong" after each sorting trial). The
subject demonstrates knowledge of the principle by placing each card under a set of
"key" cards which differ in form, color, and number. The principle changes once the
subject has responded correctly for 10 consecutive trials, indicating that she or he has
learned the principle. To perform normally on the Wisconsin Card Sorting Test, an
individual must have normal capacities for at least the following functions: memory of
the goal of the task, auditory attention to the description of the task demand and to the
feedback from the tester, visual attention to the characteristics of the sorting cards and
the key cards, adequate motor skills to allow placement of the cards where the subject
intends, learning of the principles to be acquired, proper abstraction to allow learning of
the principles, categorization of the cards, maintenance of the categorization in
"working memory," and executive control of the many cognitive functions at work at
the same time. This list is only partially complete, but it serves to emphasize the
complexity of functions involved in the completion of many neuropsychological tests.
Thus, conclusions about any single poor test performance must include a consideration
of the possibility that any of a number of functions may be causing the deficit.

While this reserve should be applied to any neuropsychological evaluation, it is

particularly important that it be applied to the evaluation of patients with psychiatric
disorders, since there are many nonspecific reasons that their performance may suffer.
To address the many possibilities that can contribute to poor performance on a
neuropsychological test, consider further the example of using the Wisconsin Card
Sorting Test as a measure of executive functions in patients with schizophrenia.
Schizophrenic patients make various types of errors on the Wisconsin Card Sorting Test
(18). One of the key measures of impairment on this test is the number of perseverative
errors committed. These are scored when a subject continues to sort according to a
principle despite feedback indicating that this strategy is incorrect. This measure has
been associated with frontal lesions in studies by Milner (19) and others (20-22),
although not in all studies (23). A similarity in the number of perseverative errors made
by patients with frontal lobe damage and patients with schizophrenia has contributed to
the development of the notion that the frontal cortex is dysfunctional in schizophrenia
(24, 25). Schizophrenic patients make more perseverative errors on the Wisconsin Card
Sorting Test than comparison subjects, including patients with affective disorders, and
these differences exist even when factors such as premorbid intelligence and symptom
severity are controlled (6). However, there are many mechanisms by which
schizophrenic patients can make perseverative errors on this test, even if they are
cooperative and motivated to perform well. As a result of verbal information-processing
or verbal learning deficits, they may never gain a proper understanding of the purpose
of the test. Because of impaired short-term or working memory functions, they may lose
their awareness of the purpose of the test. Impaired auditory attention may lead them to
misunderstand the nature of the feedback they receive. Finally, the symptoms of
psychosis may interfere with their ability to process the feedback. While any of these
possible, and not unlikely, situations would lead to a great number of perseverative
errors, none of the errors could be specifically attributable to impairment in executive
functions. Thus, the conclusion that there is a specific deficit can rarely be made on the
basis of poor performance on a single test.

A neuropsychological deficit in a psychiatric patient should be determined through a

broad assessment of the functions described above. Because of the increased likelihood
of spurious poor performance on any test or set of tests by a single individual, this is
especially important in a clinical evaluation. A clinical finding based on a single test
performance can be useful in generating hypotheses about areas of impairment, but such
a finding requires considerable substantiation from additional measures in the same area
of functioning. Hence, one of the important considerations in the use of standard formal
test batteries such as the Halstead-Reitan is that the clinical investigator should pursue
important findings with tests outside of the limited formal battery.

In research studies of a group of patients, the time required to complete a full battery for
each subject is generally prohibitive. Any individual test performance of the patient
group, however, should be compared to an estimate of how the general population
performs, as determined in the process of test standardization. (This comparison can be
problematic, however, if the standardization sample is not reflective of the general
population. For example, users of the Heaton manual for the Wisconsin Card Sorting
Test [17] may note that the average number of perseverative errors among "normal"
individuals is 12.6, but since this standardization sample had an IQ of 114--about 1 SD
above the mean--and Wisconsin Card Sorting Test performance is correlated with IQ
[17], this standardization sample is not reflective of the general population.) Ideally,
patient and control groups will be shown to differ on a task that is identical to a control
task with the exception of one feature, which is matched between tasks for difficulty,
ceiling effects, and floor effects. In an optimal comparison, the scores in each group
should be normally distributed, with perfect performance a rarity. This prevents results
such as spurious between-group differences due to near-perfect performance by patients
and control subjects on an extremely easy control task. A specific deficit identified in
comparisons of groups' performance in this manner has been referred to as a
"differential deficit" by Chapman and Chapman (26), since the difference between the
two tests discriminates the performance of the two groups, enabling one to conclude
that a group demonstrates a performance deficit which can be accounted for by a
specific feature and not simply by general dysfunction. For example, a research finding
that the performance of patients with major depression is impaired on a
neuropsychological test with a timed motor component is not of interest, because these
patients tend to perform worse than normal subjects on almost all relatively difficult
tasks that are timed (27). This general deficit may be attributable to any number of
problems, including, among others, lack of motivation, lethargy, distraction secondary
to intrusive thoughts, and general motor retardation. Furthermore, it is often not correct
to conclude that the "worst" deficit in a battery of tests reflects the most basic or central
cognitive disturbance in a disorder, even if the scores on a series of tests are made
statistically comparable to one another with a "standardized residualized scores"
approach (28). It is more likely that the largest differences between a group with a
disorder and a control group will occur on the most difficult tests. With the use of
appropriate statistical procedures, attempts to replicate findings of specific deficits in
psychiatric disorders have often resulted in finding that the disorders are attributable to
generalized deficits (29). Control subjects may be able to summon the energy or
motivation to perform well on particularly difficult tests, whereas psychiatric patients
may not have this capacity. One of the more impressive findings in schizophrenia
research is that while schizophrenic patients may demonstrate deficits on verbal
recognition tests in comparison with normal control subjects, their verbal recall deficit
is more severe (30, 31). This finding is impressive because the differential deficit
persists even when the tasks are matched for difficulty in such a way that
nonschizophrenic control subjects perform similarly on both tests (32).

Finally, a strategy to match patients and control subjects on a specific measure must
consider the impact of the psychiatric disorder on the variable chosen. For instance,
since level of education and socioeconomic status can be reduced by disorders such as
schizophrenia, a strategy to compare the cognitive performance of schizophrenic
patients and control subjects matched on these variables would result in a group of
schizophrenic patients who are likely to have exceeded the level of functioning expected
of the average individual with schizophrenia, and they may thus perform better than the
average schizophrenic patient on the test of interest (26). A better strategy is to match
groups on variables not affected by the illness, such as the pronunciation of irregular
words as measured by performance on the National Adult Reading Test (33), which has
been found not to be compromised by schizophrenia (34) or Alzheimer's disease (35).
Another alternative is to match patient and control groups on the level of education or
socioeconomic status of the parents of the patients and control subjects, as long as the
parents are unaffected by the disorder under study (26).

The strategy of administering to patients and control subjects a battery of tests of similar
difficulty, some of which do not differ between groups, makes possible not only a more
accurate determination of changes in mental processes associated with psychiatric
disorder but also the identification of preserved functions in the patient group.
Determining normal abilities in individuals with severe psychiatric disorders can be as
important as identifying deficits. While the finding that a patient or group of patients
has impairment in a specific area of neuropsychological function but has other areas of
normality may or may not lead to confirmation of a particular model of deficient mental
processes, it certainly provides data about the experience of these patients and about
which mental functions may be failing them and which are preserved.

ASSUMPTION 2: Abnormal Neuropsychological Test Performance Indicates Specific

Regional Brain Dysfunction

Possibly the most prevalent assumption made about neuropsychological test data is that
if psychiatric patients' performance on a specific test is equivalent to the performance of
patients with discrete lesions in a specific region, this suggests the presence of similar
neuroanatomic abnormalities in the psychiatric patients. This assumption is found even
in many of the best published papers involving the assessment of neuropsychological
functions in psychiatric patients (2-5, 9-13, 36-40). It is very likely not to be true. Poor
performance may be the result of a broad range of possibilities, including damage to one
of several areas, the accumulative effect of mild deficits in multiple areas, or factors
unrelated to specific brain dysfunction.

As I have mentioned, any individual or group of individuals may perform poorly on a

neuropsychological test or series of tests for many different reasons. The more complex
the measure, the more likely it is that poor performance will result from nonspecific
factors. The best example of this phenomenon is that psychiatric patients with reduced
motivation will often demonstrate neuropsychological profiles consistent with frontal
lobe disease, since in a battery of standard neuropsychological tests, the most highly
complex tasks are generally chosen as measures of frontal lobe function. Thus, the
complexity of functions involved in any single neuropsychological test almost always
prohibits the conclusion, based solely on test performance, that a specific brain region is
impaired.

Even if patients are able to comprehend and maintain the concepts required for task
completion and are motivated to perform well, poor performance on a test is not
necessarily related to dysfunction of specific brain regions. Throughout the history of
the study of regional brain function, attempts have been made to isolate specific
functions of specific brain regions, but the success of these mapping efforts has been
mixed with overestimations of the capacities of the map at any given time. In 1861
Broca proposed that the third frontal convolution of the left hemisphere, area 44, is
associated with expressive speech, and damage to this region results in expressive
language deficit. This disturbance, now termed "Broca's aphasia," continues to serve as
a leading indicator of disease or insult in this region. Other findings, such as the
laterality of motor perception and movement and the role of the occipital cortex in
vision, have had a tremendous impact on our current understanding of regional brain
function. In addition, data from animal studies suggest that some brain regions mediate
very specific functions. In monkeys, for example, memory of the precise spatial location
of visual stimuli has been found to be subserved by neural pathways leading to specific
regions of the prefrontal cortex (41). However, most recent data support the notion that
higher brain functions cannot be isolated to a specific region. Many important
behavioral and cognitive functions are mediated by complex networks of neurons
throughout the brain, and damage to an area does not always result in impairment of the
function associated with that region, possibly due to the plasticity of regional brain
function.

Luria (42) described higher mental functions in terms of complex functional systems
that "cannot be localized in narrow zones of the cortex or in isolated cell groups, but
must be organized in systems of concertedly working zones, each of which performs its
role in the complex functional system, and which may be located in completely different
and often far distant areas of the brain" (p. 31). Studies of damaged cortical connectivity
(43, 44), including commissurotomy (45), and physiological studies on "neural
networks" in human and nonhuman primates (41, 46-49) have supported Luria's
conceptualizations of higher cortical functions. They suggest that higher cortical
functions may be impaired by damage to any aspect of the complex system that
regulates them. These concepts have become particularly relevant for understanding
brain dysfunction in psychiatric patients (50-53), since poor neuropsychological test
performance is found in many psychiatric disorders without established regional brain
dysfunction.
Studies of the organization of relations among brain regions in psychiatric patients
suggest that identifying the regional pattern of electrical activity may improve
psychiatric diagnosis and treatment (46). An example of the complexity of the relation
between brain structure and neuropsychological function in psychiatric patients can be
seen in the strong positive correlation between hippocampal size and physiological
activation of the dorsolateral prefrontal cortex (but not other regions) during the
Wisconsin Card Sorting Test in schizophrenic twins compared with their normal
cotwins (54). Since activation of the dorsolateral prefrontal cortex during this test is
related to task performance (55), it suggests that structural deficits may lead to
performance deficits farther "downstream" in the neural circuitry associated with a
complex task. Thus, the complexity of neural networks associated with a
neuropsychological task suggests that damage to any part of the network has the
potential to impair performance on the task.

In considering the relation between neuropsychological deficit and brain impairment in

psychiatric patients, it is also important to consider the ability of the brain to adapt to
damage. According to Luria (56), higher mental functions result from a unified system
of components with a complex and plastic, not concrete, structure in which the initial
and final links of the system (the task and the effect) remain constant and unchanged,
and the intermediate links (the means and the performance of the task) may be broadly
modified. The functional capacities of different brain regions are quite flexible in
humans and other animals. If one area is damaged, the functioning of different brain
areas is altered to accommodate for the damage, which may lead to greater adaptive
capacities in the individual or greater impairment (57-64). Thus, it seems unwarranted
to draw the conclusion that there is specific regional damage in psychiatric patients on
the basis of isolated neuropsychological deficit.

The effect of insult to the brain may depend heavily on the stage of development at
which the insult occurs (65), which may have important implications for developmental
models of psychiatric illness (58, 66). In the early stages of development, basic
functions such as visual perception depend heavily on "lower" areas of the cortex, but in
the later stages of development, the activity starts to depend on a different neural
network (67). For example, a lesion of the lower regions of the visual cortex in early
childhood leads to systemic underdevelopment of the higher regions responsible for the
integration of visual concepts. In contrast, a similar lesion in an adult causes only partial
deficiencies of visual analysis and synthesis, leaving the more complex cognitive
functions, formed at an earlier stage, unaffected (42). This developmental model of
brain insult and adaptation may eventually be useful in accounting for psychiatric
illnesses with developmental etiologies, since neural network models of psychiatric
disorders propose that a variety of intrinsic and extrinsic factors, even early emotional
experiences, may affect the course of brain development (68).

In sum, the assumption that abnormal neuropsychological test performance indicates

specific regional brain dysfunction is often not true in brain-injured patients with
otherwise normal histories; it is even less likely to be true in the assessment of patients
with psychiatric disorders. Most mental processes, especially the higher cognitive
functions relevant to psychiatric disorders, are regulated by complex neural networks
that connect various brain regions. Therefore, deficits on neuropsychological tests may
derive not only from nonspecific factors, such as lack of motivation in psychiatric
patients, but from dysfunction in any of the connected brain regions that form the neural
networks that guide higher mental processes, and even in the connections themselves.
Increasingly, research on the neuropsychology of psychiatric disorders needs to focus
on the function and dysfunction of these neural networks.

ASSUMPTION 3: "Hypoactivity" During Functional Imaging Procedures With

Cognitive Activation Tasks Suggests Regional Brain Dysfunction

Another frequent assumption about neuropsychological test performance and

psychiatric disorders is that less than normal regional activation during a test, as
measured by a variety of functional imaging technologies including regional cerebral
blood flow, single photon emission computed tomography, and positron emission
tomography, is an indication of pathology in that region in the disorder under
examination (55, 69-73). This conclusion is viewed as bolstered by evidence that
reduced metabolism is positively correlated with poor performance on the test used for
activation. This assumption, although more subtle than assumption 2 discussed above,
can also be erroneous. It is likely that patients who perform poorly on a
neuropsychological test are not processing the task in the same manner as normal
subjects who perform well on the task. There are enormous differences between patient
groups and normal subjects in the manner in which they approach tasks, and these
differences are almost certain to result in different patterns of activation that are
correlated with performance, especially since level of difficulty within a single task can
greatly affect regional activation (74). Thus, the correlation between poor performance
and reduced activation reported in numerous functional imaging studies may actually
reflect the different approaches of patients and control subjects, possibly due to
differences in task difficulty, rather than reflecting regional brain dysfunction in the
patients. Particularly striking are reports of an absence in patient groups of the lateral
asymmetry found in normal subjects. In general, among groups of normal control
subjects with left-hemisphere dominance, verbal tasks are known to activate left-
hemisphere regions in functional imaging studies, while nonverbal visual tasks activate
right-hemisphere regions (75). Schizophrenic patients, who perform worse than normal
subjects on both verbal and nonverbal visual tasks (28, 76, 77), have been reported to
have less extreme laterality of physiological activation during execution of these tasks
(73, 78, 79), including the Continuous Performance Test, a test of visual attention (78,
79). It seems most parsimonious to understand these findings in terms of the inability of
the patients to approach the task in the same way as normal subjects who perform well.
Specifically, if schizophrenic patients are not completing a task such as the Continuous
Performance Test effectively, or are unable to attend to the task as well as normal
subjects, it is not surprising that the brain regions that are activated during test
performance are more randomly distributed, which would result in a statistical
"regression to symmetry," since random signals would follow a symmetrical activation
pattern. One method proposed for dealing with this type of problem is to teach patients
to perform the activation task as well as normal subjects do before collecting imaging
data, or to choose patients who perform as well as normal control subjects (unpublished
1993 paper by C.D. Frith). While this strategy may result in the selection of an atypical
patient group, the findings of these studies could be compared to those from patients
with more typical performance.

Finally, as with studies of brain structure, the reduced activation of a brain region as
assessed by physiological activation studies does not suggest that the underactivated
region is impaired, but only that is it not activated normally. Since there are many
components of a neural network that are activated to perform a specific task, the
impairment of any of these components could lead to dysfunction and/or
underactivation of the region under study. Additionally, with functional imaging, the
usual method for identifying reduced activation in a group of patients is through
calculation of the differences between activation patterns in two separate conditions.
While this method is useful in identifying mean differences between conditions, and
thus in identifying isolated areas of abnormal activity, it limits the conclusions that can
be drawn from the data collected. Identification of specific abnormal patterns of
activation may provide evidence of impaired neural networks associated with specific
psychiatric disorders (73, 80). Statistical path analysis of the physiological activation of
multiple areas corresponding to a putative neural circuit associated with task
performance may provide more convincing support of a physiology-task relationship.
This type of analysis can be applied to determine the functioning of this circuit in
patients who are believed to have an abnormality.

Another strategy for understanding patterns of regional brain activation in psychiatric

patients is to use cognitive activation tasks that are simple enough so that the neural
networks that mediate performance by normal subjects are understood. For example,
data on nonhuman primates suggest that tests of visuospatial working memory activate
neural networks connecting the primary visual centers, parietal cortex, premotor
regions, and dorsolateral prefrontal cortex (41). Preliminary imaging data suggest that
this function is regulated by a similar activation pattern in humans (81). Such a simple
cognitive task may prove to be a useful probe for imaging studies assessing hypotheses
of prefrontal dysfunction in psychiatric disorders. Performance data on tests of
visuospatial working memory tasks under laboratory conditions (82) and typical clinical
conditions (83) suggest that schizophrenic patients may have relatively specific deficits
in this area of functioning. This type of task may help to refine models of specific neural
network dysfunction in schizophrenia and other psychiatric disorders.

PRACTICAL CONTRIBUTIONS OF NEUROPSYCHOLOGICAL EVALUATION

The greatest contribution of the neuropsychological evaluation of patients with

psychiatric disorders may be that it provides important, objective data about the mental
deficiencies that shape our patients' lives. While neuropsychological tests may serve to
inform specific neuropathological models of psychiatric disorders by comparing the
performance of psychiatric patients and patients with brain lesions, this role is likely to
be usurped in the future because of the tremendous advances in imaging technologies
(84). However, images of the structure and regional activation of the brains of our
patients will not provide us with information about their difficulties with their mental
processes or about their cognitive strengths that can be used to facilitate their treatment.
The differential pattern of performance on tests of higher mental processes can be used
to predict the course of psychiatric illnesses, reduce the diagnostic heterogeneity within
disorders and thus improve diagnostic classifications, serve as an aid in the development
of treatment options, and create individualized management of patient care. Three
practical ways in which neuropsychological data may be used to serve these ends are
described below.

PREDICTORS OF COURSE OF ILLNESS. The identification of specific cognitive

deficits in psychiatric disorders may be a powerful predictor of the course of illness.
Naturally, patients of a particular diagnostic group who also have global cognitive
impairment are likely to have worse outcomes than similarly diagnosed patients who
perform normally on neuropsychological tests. In a group of patients assessed in a
psychiatric emergency room, cognitive deficit was the single best predictor of referral
for inpatient hospitalization; it was even superior to the patients' diagnoses (85). In
some disorders, such as schizophrenia and Alzheimer's disease, cognitive deficits as
assessed by a battery of neuropsychological tests may predict the onset of illness. It has
been demonstrated that general impairment on tests assessing information, memory, and
concentration (86) serves as a better tool for the prediction of the eventual development
of Alzheimer's disease in normal elderly volunteers than prior head injury, age of the
mother when the subject was born, smoking, or family history of Alzheimer's disease
(87). While these studies indicate the value of determining cognitive impairment in
general in psychiatric patients, the identification of specific cognitive deficits may be of
even greater value. Patients with major depression and particularly severe memory
deficits may be unable to sustain even simple medication regimens or may be unable to
benefit from previous gains they made in psychotherapy. Schizophrenic patients with
severe learning deficits may never gain the capacity to understand the goals of their
treatment and thus may never be able to function independently. These factors may have
a tremendous effect on the ability of patients to benefit from treatment and may thus
lead to a substantially worse outcome. The ability to identify specific cognitive deficits
in patients may allow better prediction of the course of illness, or, if possible, the
identification of treatment strategies to improve the course of illness (88, 89).

TOOLS FOR IMPROVING DIAGNOSTIC CLASSIFICATION. Similar to the

phenomenology of many major psychiatric disorders, the pattern of cognitive deficits
among patients in a single diagnostic group is heterogeneous. Not all schizophrenic
patients perform poorly on tests of verbal memory or the Wisconsin Card Sorting Test,
and not all depressed patients perform poorly on tests of psychomotor speed. Although
the etiologies of the subtypes of different psychiatric disorders have been presumed to
differ (90, 91), the success of attempts to validate these distinctions made on the basis of
phenomenology has been limited (92, 93). The identification of stable patterns of deficit
on neuropsychological tests within a disorder may contribute to the development of
hypotheses about the differing etiologies of the disorder (94). A subtypology based on
specific differences in cognitive functions among patients with the same diagnosis is
more compelling than a scheme based on phenomenology, since it suggests that
different mechanisms lead to the common end-state of disorder, while
phenomenological distinctions are limited to differences in the appearance of a disorder.

AIDS TO TREATMENT STRATEGIES. There are few empirical data on the relation
between neuropsychological deficit and response to medication, psychotherapy, and
treatment setting. However, baseline cognitive impairment may be an important
predictor of eventual response to treatment (95, 96), and improvement in cognitive
abilities during the very early stages of treatment may herald the amelioration of
symptoms weeks later (95). It seems plausible, and worthy of investigation, that the
heterogeneity of response to treatment among patients with psychotic, affective, and
anxiety disorders may be related to their pretreatment level of cognitive functioning,
and that treatment regimens which are suited specifically to an individual's pattern of
cognitive deficits and abilities may be more effective.

Because of the centrality of cognitive processes in any psychotherapeutic treatment, this

strategy may be particularly applicable for psychiatric disorders that might benefit from
psychotherapy. Historically, psychotherapeutic treatment of many psychiatric disorders,
including most notably schizophrenia, mania, and anorexia nervosa, has not reflected a
consideration of the significance of cognitive deficits. Any attempts to treat
schizophrenia with insight-related or even cognitive therapies that require advanced
concept formation, verbal memory and learning, and normal attentional capacities were
clearly not informed by data suggesting that schizophrenic patients are generally unable
to perform these functions normally. A very basic understanding of the cognitive
deficits of a patient can prevent similar future treatment failures. An example of an
improved psychotherapeutic strategy based on data from cognitive studies can be found
in the treatment of anorexia nervosa. Recent cognitive behavioral and family treatment
approaches, informed by data suggesting that patients with severe forms of anorexia
nervosa are unable to process emotionally laden and internally derived mental events,
have been far more successful than earlier approaches (97).

Recommendations regarding important environmental factors in psychiatric treatment,

such as living arrangements, work schedules, and partial hospitalization programs, may
also benefit from a consideration and assessment of cognitive deficits. This type of
assessment is often performed in an informal manner, such as observing that a patient
does not have sufficient attentional capacity to participate in structured group activity or
to handle independent living circumstances. While these observations can result in
improvements in patient care, a more formalized determination of a patient's profile of
neuropsychological strengths and deficits may provide clinicians and staff members
with a clearer picture of which treatments will be maximally beneficial to their patients.
In this manner, neuropsychological assessment can improve the quality of
individualized therapeutic management.

One of the specific benefits of determining a pattern of cognitive strengths and deficits
associated with a psychiatric disorder is that it may shed light for clinicians onto
possible strategies of cognitive rehabilitation. Since the range of possible cognitive
deficits in psychiatric patients is broad, it is necessary to make several determinations
about the level of intervention that is most likely to be successful and the goals to be
attained. Baseline neuropsychological assessment may contribute to the development of
cognitive rehabilitation programs by laying out the specific deficits that most require
treatment, and indicating whether the focus of rehabilitation should be at the level of
elementary cognitive deficits (such as those in perceptual processes), moderately
complex deficits (such as those in problem solving and verbal memory), or highly
complex deficits (such as those in interpersonal interaction) (98). In addition, since
profiles of neuropsychological performance can make possible a determination of which
cognitive functions remain intact, they can point to the areas of capacity upon which
innovative rehabilitation strategies can rely. For example, one of the primary symptoms
of neurological or psychiatric disorders that involve the frontal cortex is impairment in
the ability to develop alternative coping strategies. Thus, these patients need assistance
in generating such strategies, which should be directed by cognitive rehabilitation
specialists in such a manner that the utilization of identified cognitive strengths can be
optimized. An even more specific application of neuropsychological assessment to
cognitive rehabilitation programs is tailoring treatment to an individual's particular
profile of cognitive deficits (99), which may eventually be keyed to a patient's phase of
illness (98).
CONCLUSIONS

The importance of investigations of the possible neuroanatomic etiology and

pathophysiology of major psychiatric disorders cannot be minimized. Findings
generated from research that relate neuropsychological deficits to neuroanatomic and
neurophysiological abnormalities determined by brain imaging procedures are likely to
be a part of these eventual insights into major psychiatric illness. The important findings
in this research are likely to develop from an understanding of the neural networks that
underlie specific cognitive functions and of how this circuitry can be impaired. If
properly applied, neuropsychological tests may be useful in specifying the relation
between brain function and cognitive performance in individuals with psychiatric
disorders. The abnormal changes in mental processes suggested by the performance of
psychiatric patients on standardized neuropsychological tests should also be considered
as potential tools for improving currently available treatments, especially with regard to
individualizing treatment strategies. Research evidence of abnormal brain metabolism
or structure in groups of patients with a particular psychiatric disorder is a suggestion
that something specific may be wrong with a patient which may be relevant to the
patient's psychiatric disorder. Most importantly, however, cognitive deficits as revealed
by impaired performance on neuropsychological tests are a direct indication that
something quite specific and tangible is wrong with the patient, which is very relevant
to the patient's disorder.

REFERENCES

1. Kolb B, Whishaw IQ (eds): Fundamentals of Human Neuropsychology, 3rd ed. New

York, WH Freeman, 1989

2. Gruzelier J, Seymour K, Wilson L, Jolley A, Hirsch S: Impairments of

neuropsychological tests of temporohippocampal and frontohippocampal functions and
word fluency in remitting schizophrenia and affective disorders. Arch Gen Psychiatry
1988; 45:623-629

3. Katsanis J, Iacono WG: Temporal lobe dysfunction and electrodermal nonresponding

in schizophrenia. Biol Psychiatry 1992; 31:159-170

4. Perlick D, Mattis S, Stastny P, Silverstein B: Negative symptoms are related to both

frontal and nonfrontal neuropsychological measures in chronic schizophrenia. Arch Gen
Psychiatry 1992; 49:245-246

5. Sandman CA, Barron JL, Nackoul K, Goldstein J, Fidler F: Memory deficits

associated with chronic fatigue immune dysfunction syndrome. Biol Psychiatry 1993;
33:618-623

6. Goldberg TE, Gold JM, Greenberg R, Griffin S, Schulz SC, Pickar D, Kleinman JE,
Weinberger DR: Contrasts between patients with affective disorders and patients with
schizophrenia on a neuropsychological test battery. Am J Psychiatry 1993; 150:1355-
1362
7. Marcus JM, Hans SL, Auerbach JG, Auerbach AG: Children at risk for
schizophrenia: the Jerusalem infant development study, II: neurobehavioral deficits at
school age. Arch Gen Psychiatry 1993; 50:797-809

8. Hollander E, Cohen L, Richards M, Mullen L, DeCaria C, Stern Y: A pilot study of

the neuropsychology of obsessive-compulsive disorder and Parkinson's disease: basal
ganglia disorders. J Neuropsychiatry 1993; 5:104-107

9. Gambini O, Abbruzzese M, Scarone S: Smooth pursuit and saccadic eye movements

and Wisconsin Card Sorting Test performance in obsessive-compulsive disorder.
Psychiatry Res 1993; 48:191-200

10. Krystal JH, Karper LP, Seibyl JP, Freeman GK, Delaney R, Bremner DJ, Heninger
GR, Bowers MB, Charney DS: Subanesthetic effects of the noncompetitive NMDA
antagonist, ketamine, in humans: psychotomimetic, perceptual, cognitive, and
neuroendocrine responses. Arch Gen Psychiatry 1994; 51:199-214

11. Stolar N, Berenbaum H, Banich MT, Barch D: Neuropsychological correlates of

alogia and affective flattening in schizophrenia. Biol Psychiatry 1994; 35:164-172

12. Seidman LJ, Yurgelun-Todd D, Kremen WS, Woods BT, Goldstein JM, Faraone
SV, Tsuang MT: Relationship of prefrontal and temporal lobe MRI measures to
neuropsychological performance in chronic schizophrenia. Biol Psychiatry 1994; 35:
235-246

13. Keefe RSE, Silverman J, Lees Roitman SE, Harvey PD, Duncan A, Alroy D, Siever
LJ, Mohs RC, Davis KL: Performance of nonpsychotic relatives of schizophrenic
patients on cognitive tests. Psychiatry Res 1994; 53:1-12

14. Reitan RM, Davison LA: Clinical Neuropsychology: Current Status and
Application. New York, John Wiley & Sons, 1974

15. Golden CJ: A standardized version of Luria's neuropsychological tests, in Handbook

of Clinical Neuropsychology. Edited by Filskov S, Boll TJ. New York, Wiley-
Interscience, 1981

16. Grant DA, Berg EA: A behavioral analysis of degree of reinforcement and ease of
shifting to new responses in a Weigl-type card sorting problem. J Exp Psychol 1948;
38:404-411

17. Heaton RK: The Wisconsin Card Sorting Test Manual. Odessa, Fla, Psychological
Assessment Resources, 1981

18. Van der Does AJW, Van den Bosh RJ: What determines Wisconsin Card Sorting
performance in schizophrenia? Clin Psychol Rev 1992; 12:567-583

19. Milner B: Effects of different brain lesions on card sorting. Arch Neurol 1963;
9:100-110
20. Drewe EA: The effect of type and area of brain lesion on Wisconsin Card Sorting
Test performance. Cortex 1974; 10:159-170

21. Nelson HE: A modified card sorting task sensitive to frontal lobe defects. Cortex
1976; 12:313-324

22. Robinson AL, Heaton RK, Lehman RAW, Stilson DW: The utility of the Wisconsin
Card Sorting Test in detecting and localizing frontal lobe lesions. J Consult Clin
Psychol 1980; 48:605-614

23. Anderson SW, Damasio H, Jones RD, Tranel D: Wisconsin Card Sorting
performance as a measure of frontal lobe damage. J Clin Exp Neuropsychol 1991;
13:909-922

24. Stuss DT, Benson DF, Kaplan EF, Weir WS, Naesser MA, Lieberman I, Ferrill D:
The involvement of orbitofrontal cerebrum in cognitive tasks. Neuropsychologia 1983;
21:235-248

25. Goldberg TE, Weinberger DR, Berman KF, Pliskin NH, Podd MH: Further
evidence for dementia of the prefrontal type in schizophrenia? Arch Gen Psychiatry
1987; 44:1008-1014

26. Chapman LJ, Chapman JP: Disordered Thought in Schizophrenia. New York,
Appleton-Century-Crofts, 1973

27. Sackeim HA, Steif BL: The neuropsychology of depression and mania, in
Depression and Mania. Edited by Georgotas A, Cancro R. New York, Elsevier, 1988

28. Saykin AJ, Gur RC, Gur RE, Mozley PD, Mozley LH, Resnick SM, Kester DB,
Stafiniak P: Neuropsychological function in schizophrenia. Arch Gen Psychiatry 1991;
48:618-624

29. Blanchard JJ, Neale JM: The neuropsychological signature of schizophrenia:

generalized or differential deficit? Am J Psychiatry 1994; 151:40-48

30. Bauman E, Murray DJ: Recognition vs recall in schizophrenia. Can J Psychol 1968;
22:18-25

31. Koh SD: Remembering of verbal materials by schizophrenic young adults, in

Language and Cognition in Schizophrenia. Edited by Schwartz S. Hillsdale, NJ,
Lawrence Erlbaum Associates, 1978

32. Calev A: Recall and recognition in chronic nondemented schizophrenics: use of

matched tasks. J Abnorm Psychol 1984; 93:172-177

33. Nelson HE, Willison JR: The Revised National Adult Reading Test Manual. Slough,
England, National Foundation for Educational Research-Nelson, 1991

34. O'Carroll R, Walker M, Dunan J, Murray C, Blackwood D, Ebmeier KP, Goodwin

GM: Selecting controls for schizophrenia research studies: the use of the National Adult
Reading Test (NART) is a measure of premorbid ability. Schizophrenia Res 1992;
8:137-141

35. Crawford JR, Besson JAO, Parker DM: Estimation of pre-morbid intelligence in
organic conditions. Br J Psychiatry 1988; 153: 178-181

36. Taylor MA, Abrams R: Cognitive impairment patterns in schizophrenia and

affective disorder. J Neurol Neurosurg Psychiatry 1987; 50:895-899

37. Bornstein RA, Schwarzkopf SB, Olson SC, Nasrallah HA: Third-ventricle
enlargement and neuropsychological deficit in schizophrenia. Biol Psychiatry 1993;
31:954-961

38. Nelson E, Early TS, Haller JW: Visual attention in obsessive-compulsive disorder.
Psychiatry Res 1993; 49:183-196

39. Nestor PG, Shenton ME, McCarley RW, Haimson J, Smith RS, O'Donnell B,
Kimble M, Kikinis R, Jolesz FA: Neuropsychological correlates of MRI temporal lobe
abnormalities in schizophrenia. Am J Psychiatry 1993; 150:1849-1855

40. Rosse RB, Schwartz BL, Kim SY, Deutsch SI: Correlation between antisaccade and
Wisconsin Card Sorting Test performance in schizophrenia. Am J Psychiatry 1993;
150:333-335

41. Goldman-Rakic PS: Circuitry of primate prefrontal cortex and regulation of

behavior by representational knowledge, in Handbook of Physiology, Section 1: The
Nervous System, vol 5: Higher Functions of the Brain. Edited by Plum F; Mountcastle
VB, section editor. Bethesda, Md, American Physiological Society, 1987

42. Luria RA: The Working Brain. New York, Basic Books, 1973

43. Geschwind N: Disconnexion syndromes in animals and men, I. Brain 1965; 88:237-
294

44. Geschwind N: Disconnexion syndromes in animals and men, II. Brain 1965;
88:585-644

45. Sperry R: Some effects of disconnecting the cerebral hemispheres. Science 1982;
217:1223-1226

46. John ER, Prichep LS, Fridman J, Easton P: Neurometrics: computer-assisted

differential diagnosis of brain dysfunctions. Science 1988; 239:162-169

47. Rumelhart DE, McClelland JL, PDP Research Group (eds): Parallel Distributed
Processing: Explorations in the Microstructure of Cognition, vol 1. Cambridge, Mass,
MIT Press, 1986

48. McClelland JL, Rumelhart DE, PDP Research Group (eds): Parallel Distributed
Processing: Explorations in the Microstructure of Cognition, vol 2. Cambridge, Mass,
MIT Press, 1986
49. Grossberg S: Neural Networks and Natural Intelligence. Cambridge, Mass, MIT
Press, 1988

50. Hoffman RE: Verbal hallucinations and language production processes in

schizophrenia. Behavioural and Brain Sciences 1986; 9:503-548

51. Cohen JD, Servan-Schreiber D: Context, cortex, and dopamine: a connectionist

approach to behavior and biology in schizophrenia. Psychol Rev 1992; 99:45-77

52. Hoffman RE, McGlashan TH: Parallel distributed processing and the emergence of
schizophrenic symptoms. Schizophr Bull 1993; 19:119-140

53. Keefe RSE: The neurobiology of disturbances of the self: autonoetic agnosia in
schizophrenia, in Insight and Psychosis. Edited by Amador XF, David A. New York,
Oxford University Press (in press)

54. Weinberger DR, Berman KF, Suddath R, Torrey EF: Evidence of dysfunction of a
prefrontal-limbic network in schizophrenia:

magnetic resonance imaging and regional cerebral blood flow study of discordant
monozygotic twins. Am J Psychiatry 1992; 149:890-897

55. Weinberger DR, Berman KF, Zec RF: Physiological dysfunction of the dorsolateral
prefrontal cortex in schizophrenia, I: regional cerebral blood flow (rCBF) evidence.
Arch Gen Psychiatry 1986; 43:114-125

56. Luria AR: Higher Cortical Functions in Man, 2nd ed. New York, Basic Books, 1980

57. Sperry RW, Zaidel E, Zaidel D: Self recognition and social awareness in the
deconnected minor hemisphere. Neuropsychologia 1979; 17:153-166

58. Stevens JR: Abnormal reinnervation as a basis for schizophrenia: a hypothesis. Arch
Gen Psychiatry 1992; 49:238-243

59. Raisman G: Neuronal plasticity in the septal nuclei of the adult rat. Brain Res 1969;
14:25-48

60. Moore RY, Jborklund A, Stenevi U: Plastic changes in the adrenergic innervation of
the rat septal area in response to denervation. Brain Res 1971; 33:13-35

61. Sutin J: Neuronal plasticity induced by electrical stimulation, in Brain Stimulation

and Neuronal Plasticity. Edited by Tsubodawa T. Tokyo, Neuron, 1985

62. Haracz JL: Neural plasticity in schizophrenia. Schizophr Bull 1985; 11:191-299

63. Babb TL: Short-term and long-term modification of neurons and evoked potentials
in the human hippocampal formation. Neurosciences Res Program Bull 1982; 20:729-
739
64. Hubel DH, Wiesel TN, LeVay S: Plasticity of ocular dominance columns in monkey
striate cortex. Philos Trans R Soc Lond Biol 1977; 278:377-409

65. Schneider GE: Is it really better to have your brain lesion early? a revision of the
"Kennard principle." Neuropsychologia 1979; 17:557-583

66. Weinberger DR: Implications of normal brain development for the pathogenesis of
schizophrenia. Arch Gen Psychiatry 1987; 44:660-669

67. Luria AR, Simernitskaya EG, Tubylevich B: The structure of psychological

processes in relation to cerebral organization. Neuropsychologia 1970; 8:13-19

68. Edelman GM, Cunningham BA: Place-dependent cell adhesion, process retraction,
and spatial signaling in neural morphogenesis. Cold Spring Harb Symp Quant Biol
1990; 55:303-318

69. Buchsbaum MS, Nuechterlein KH, Haier RJ, Wu J, Sicotte N, Hazlett E, Asarnow
R, Potkin S, Guich S: Glucose metabolic rate in normals and schizophrenics during the
Continuous Performance Test assessed by positron emission tomography. Br J
Psychiatry 1990; 156:216-227

70. Busatto G, Kerwin RW, Costa DC, David AS, Pilowsky L, Ell PJ: Temporal lobe
activation during verbal memory challenge: a SPET study (abstract). Schizophr Res
1992; 6:148-149

71. Gur RE, Skolnick BE, Gur RC, Caroff S, Rieger W, Walter DO, Younkin D,
Reivich M: Brain function in psychiatric disorders, II: regional cerebral blood flow in
medicated unipolar depressives. Arch Gen Psychiatry 1984; 41:695-699

72. Warren LR, Butler RW, Katholi CR, McFarland CE, Crews EL, Halsey JH Jr: Focal
changes in cerebral blood flow produced by monetary incentive during a mental
mathematics task in normal and depressed subjects. Brain Cogn 1984; 3:71-85

73. Gur RE, Jaggi JL, Shtasel DL, Tagland JD, Gur RC: Cerebral blood flow in
schizophrenia: effects of memory processing on regional activation. Biol Psychiatry
1994; 35:3-15

74. Gur RC, Gur RE, Skolnick BE, Resnick SM, Silver FL, Chawluk JB, Muenz L,
Obrist WD, Reivich M: Effects of task difficulty on regional cerebral blood flow:
relationships with anxiety and performance. Psychophysiology 1988; 25:392-399

75. Allen M: Models of hemispheric specialization. Psychol Bull 1983; 93:73-104

76. Braff DL, Heaton R, Kuck J, Cullum M, Moranville J, Grant I, Zisook S: The
generalized pattern of neuropsychological deficits in outpatients with chronic
schizophrenia with heterogeneous Wisconsin Card Sorting Test results. Arch Gen
Psychiatry 1991; 48:891-898
77. Hoff AL, Riordan H, O'Donnell DW, Morris L, DeLisi LE: Neuropsychological
functioning of first-episode schizophreniform patients. Am J Psychiatry 1992; 149:898-
903

78. Buchsbaum MS, Haier RJ, Potkin SG, Nuechterlein K, Bracha HS, Katz M, Lohr J,
Wu J, Lottenberg S, Jerabeck PA, Trenary M, Tafalla R, Reynolds C, Bunney WE:
Frontostriatal disorder of cerebral metabolism in never-medicated schizophrenics. Arch
Gen Psychiatry 1992; 49:935-942

79. Cornblatt BA, Keilp JG: Impaired attention, genetics, and the pathophysiology of
schizophrenia. Schizophr Bull 1994; 20:31-46

80. Sackeim HA, Prohovnik I, Moeller JR, Brown RP, Apter S, Prudic J, Devanand DP,
Mukherjee S: Regional cerebral blood flow in in mood disorders, I: comparison of
major depressives and normal controls at rest. Arch Gen Psychiatry 1990; 47:60-70

81. Jonides J, Smith EE, Koeppe RA, Awh E, Minoshima S, Mintun MA: Spatial
working memory in humans as revealed by PET. Nature 1993; 363:623-625

82. Park S, Holzman PS: Schizophrenics show working memory deficits. Arch Gen
Psychiatry 1992; 49:975-982

83. Keefe RSE, Lees Roitman SE, Harvey PD, Blum C, DuPre RL, Davidson M, Davis
KL: A pen-and-paper human analogue of a monkey prefrontal cortex activation task:
spatial working memory in patients with schizophrenia. Schizophr Res (in press)

84. Keefe RSE, Harvey PD: Understanding Schizophrenia. New York, Free Press, 1994

85. Galankyer I, Harvey PD: Neuropsychological screening in the psychiatric

emergency room. Compr Psychiatry 1992; 33:291-295

86. Blessed B, Tomlinson E, Roth M: The association between quantitative measures of

dementia and of senile change in the cerebral grey matter of elderly subjects. Br J
Psychiatry 1968; 114:797-811

87. Katzman R, Aronson M, Fuld P, Kawas C, Brown T, Morgenstern H, Frishman W,

Gidez L, Eder H, Ooi WL: Development of dementing illness in an 80-year-old
volunteer cohort. Ann Neurol 1989; 25:317-324

88. Brenner HD, Hodel B, Roder V, Corrigan P: Treatment of cognitive dysfunctions

and behavioral deficits in schizophrenia. Schizophr Bull 1992; 18:21-26

89. Bellack AS: Cognitive rehabilitation in schizophrenia: is it possible? is it necessary?

Schizophr Bull 1992; 18:43-50

90. Bleuler E: Dementia Praecox or the Group of Schizophrenias (1908). Translated by

Zinkin J. New York, International Universities Press, 1950
91. Zimmerman M, Coryell W, Pfohl B, Stangl D: The validity of four definitions of
endogenous depression, II: clinical, demographic, familial, and psychosocial correlates.
Arch Gen Psychiatry 1986; 43:234-244

92. Fenton WS, McGlashan TH: Natural history of schizophrenia subtypes, I:

longitudinal study of paranoid, hebephrenic, and undifferentiated schizophrenia. Arch
Gen Psychiatry 1991; 48: 969-977

93. Guscott R, Grof P: The clinical meaning of refractory depression: a review for the
clinician. Am J Psychiatry 1991; 148:695-704

94. Shallice T, Burgess PW, Frith CD: Can the neuropsychological case-study approach
be applied to schizophrenia Psychol Med 1991; 21:661-673

95. Serper MR, Davidson M, Harvey PD: Attentional predictors of clinical change in
neuroleptic: treatment. Schizophr Res (in press)

96. Marder SR, Asarnow RF, Van Putten T: Information processing and neuroleptic
response in acute and stabilized schizophrenic patients. Psychiatry Res 1984; 13:41-49

97. Garner DM, Garfinkel PE, Bemis KM: A multidimensional psychotherapy for
anorexia nervosa. Int J Eating Disorders 1982; 1:3-46

98. Green MF: Cognitive rehabilitation in schizophrenia: is it time yet? Am J Psychiatry

1993; 150:178-187

99. Spaulding WD, Sullivan M: From the laboratory to the clinic: psychological
methods and principles in psychiatric rehabilitation, in Handbook of Psychiatric
Rehabilitation. Edited by Liberman RP. Elmsford, NY, Pergamon Press, 1991

305 PQ 1254494611