Neuropsychological and Psychophysiological

Correlates of Psychosocial Functioning

in Schizophrenia
by John S. Brekke, Adrian Raine, Mark Ansel, Todd Lencz, and Laura Bird

This study tested hypothesized relationships between

neuropsychological and psychophysiological variables
and concurrent levels of clinical and psychosocial
functioning in schizophrenia. The sample consisted of
40 subjects diagnosed with a chronic schizophrenia
spectrum disorder and living in community-based settings. The psychophysiological variables were tonic
skin conductance (SC) level, SC reactivity to stressors,
and SC response to orienting stimuli. The neuropsychological measures were the Stroop, the Controlled
Word Association Test, and four subtests of the
Wechsler Adult Intelligence ScaleRevised (block
design, digit symbol, digit span, and arithmetic). The
psychosocial variables were measures of symptomatology, independent living, work, and social functioning.
The results suggested that higher symptoms were associated with higher resting arousal, lower stress reactivity, status as an electrodermal responder, and deficits
in verbal fluency and visuo-motor functioning. The
pattern for better social functioning was higher resting
arousal, lower stress reactivity, and more responses to
orienting stimuli. Higher levels of independent living
were associated with better visuo-motor and verbal
processing. Increased work functioning was associated
with better complex visuo-spatial processing. These
findings are discussed in terms of (1) the specificity of
associations between psychosocial, psychophysiological, and neuropsychological variables and (2) a holistic
perspective toward understanding these relationships
and their relevance to rehabilitation in schizophrenia.
Schizophrenia Bulletin, 23(1): 19-28,1997.
Given the large body of research on schizophrenia, few
studies have examined the relationships between neuropsychological and psychophysiological variables and
clinical or psychosocial functioning (Spaulding et al.
1986; Allen 1990; Green 1993). However, for several rea-

Reprint requests should be sent to Dr. J.S. Brekke, School of Social

Work, University Park MC-0411, University of Southern California, Los
Angeles, CA 90089-0411.

19

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on November 20, 2016

sons, this knowledge is important to our understanding of

the disorder and to the development of effective interventions. First, while schizophrenia has been related to
deficits in psychophysiological and neuropsychological
functioning (Levin et al. 1989; Dawson 1990; Zahn et al.
1991; Gold and Harvey 1993), the degree to which these
deficits have significance at the psychosocial level has
been largely unexplored (Spaulding et al. 1986; Hogarty
1988; Spaulding 1992; Green 1993). Second, studies are
needed that will identify the neuropsychological and psychophysiological abnormalities that restrict functioning in
schizophrenia, that could underlie the functional heterogeneity among individuals with schizophrenia, and that
could influence responsiveness to rehabilitative interventions (Spaulding et al. 1986; Carpenter et al. 1988;
Bellack 1992; Brenner et al. 1992; Green 1993). Finally,
Green (1993) has recently argued that identifying the neuropsychological correlates to psychosocial functioning is
a first step toward developing effective cognitive remediation strategies in schizophrenia. In this study, we explored
the degree to which selected neuropsychological and psychophysiological variables were related to clinical and
psychosocial functioning in schizophrenia. To our knowledge, this was the first attempt to examine the relationships between both neuropsychological and psychophysiological variables and distinct aspects of psychosocial
functioning in schizophrenia.
At the outset, we establish a distinction between two
aspects of psychosocial functioning. First are the symptomatic outcomes, and second are the functional outcomes
such as social, work, and independent living. This distinction is useful in the examination of literature on the relationship between psychophysiological or neuropsychological variables and psychosocial functioning and was used
to formulate the hypotheses of this study.

Abstract

Schizophrenia Bulletin, Vol. 23, No. 1, 1997

J.S. Brekke et al.

Several studies have considered psychophysiological predictors of symptomatic outcome in schizophrenia. Zahn et
al. (1981) found that the patients who remained clinically
ill after a 4-month hospital stay were characterized by a
high baseline arousal (heart rate and spontaneous skin
conductance response) and reduced reactivity to demanding stimuli. These findings were only partly replicated by
Straube et al. (1987), who found that patients with minimal improvement over a 28-day treatment period showed
relatively low levels of electrodermal responsivity to
demanding stimuli but more responsivity on cardiovascular measures. In another partial replication of the Zahn et
al. findings, Mussgay et al. (1993) reported that patients
who relapsed over a 1-year followup period tended to
have lower levels of arousal and less responsivity than
patients who did not relapse. Concerning symptomatic
outcomes, Dawson et al. (1992a, 1992>) found that higher
tonic skin conductance (SC) arousal, slower SC habituation to orienting stimuli, and being an electrodermal
responder during an inpatient testing were predictive of
higher symptom scores during a subsequent outpatient
assessment. Although these studies are not entirely consistent (perhaps due to methodological differences in responsivity and outcome measures), they suggest that increased
baseline arousal, slow habituation, and reduced responsivity to stressful or demanding stimuli may predict poor
symptomatic outcomes.
Concerning the relationship between neuropsychological deficits and symptom levels in schizophrenia, most
studies have examined the positive or negative symptom
subtypes (Gold and Harvey 1993). With regard to general
symptom levels, the literature is sparse and inconsistent
(Dickerson et al. 1991). Some studies have found no relationship between neuropsychological deficits and general
symptomatology, and others have found that deficits on
measures of frontal functioning are related to higher overall symptom levels.
Based on the findings cited above, we would predict
that higher symptom levels in schizophrenia will be associated with higher resting SC arousal, lower SC stress reactivity, status as an electrodermal responder, and deficits on
neuropsychological measures of frontal functioning.

Psychophysiological and
Neuropsychological Correlates of
Functional Outcomes
We are aware of only two studies that examine the rela-

20

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on November 20, 2016

tionship between psychophysiological variables and functional outcomes in schizophrenia. Ohman et al. (1989)
used a sample of male schizophrenia subjects and found
that electrodermal nonresponders to orienting stimuli had
worse outcomes over a 2-year period in terms of social
and occupational functioning. Wieselgren et al. (1994)
found that among female schizophrenia subjects poorer
functioning was related to higher tonic SC levels. Two
issues should be noted with regard to these findings. First,
these gender effects co-occur with differences in the specific SC measures used; second, the measure of psychosocial functioning used in these studies was a combined
index of social and occupational variables. Relevant to
this, Tarrier and Turpin (1992) and Dawson et al. (1992a)
speculate that different electrodermal variables (e.g., phasic or tonic measures) might have distinct relationships
with various aspects of clinical or psychosocial functioning. This suggests that it is important to consider specific
electrodermal measures and their relationships to distinct
aspects of psychosocial functioning.
There have been a few studies concerning the neuropsychological predictors of successful outcome for psychiatric patients in the community. Wykes et al. (1990)
and Wykes (1994) administered a complex reaction time
test to a group of 49 chronic psychiatric patients and followed them for 3 to 6 years. Better performance on the
response measure was a strong predictor of higher levels
of independent living during the followup period.
Silverstein et al. (1991) found that better performance on
cognitive measures of reading, writing, expressive speech,
and arithmetic were associated with better work outcomes
over a 2-year period. Buchanan et al. (1994) found that
changes in verbal memory were correlated with changes
in social and occupational functioning. Lysaker et al.
(1995) found a relationship between better performance
on the Wisconsin Card Sorting Task and amount of work.
Two other studies found that poorer performance on neuropsychological measures of frontal functioning was
related to decrements in social and work functioning
among chronic schizophrenia sufferers (Allen 1990;
Breier et al. 1991). Unlike the studies just cited, Johnstone
et al. (1990) failed to find a significant relationship
between two cognitive measures (digit symbol from the
Wechsler Adult Intelligence Scale-Revised [WAIS-R;
Wechsler 1981] and a vocabulary test) and occupational
outcome at 2 years in a sample of first-episode schizophrenia subjects.
Even though it is difficult to draw clear conclusions
across these studies because of the wide variability in the
selection of measures, we would predict that neuropsychological deficits will be related to poorer work, independent living, and social functioning.
In addition to these empirical findings, there has been

Psychophysiological and
Neuropsychological Correlates of
Symptomatic Outcomes

Schizophrenia Bulletin, Vol. 23, No. 1, 1997

Psychosocial Functioning

1. Higher symptom levels in schizophrenia will be

associated with higher resting SC arousal, lower SC stress
reactivity, status as an electrodermal responder, and
deficits on neuropsychological measures of frontal functioning.
2. Higher levels of social functioning will be associated with lower SC stress reactivity, higher electrodermal
orienting, and better neuropsychological performance.
3. Poorer functioning in work and independent living
will be associated with neuropsychological deficits.

Measures. Psychophysiological and neuropsychological

measures were taken to index resting SC arousal, psychophysiological SC reactivity to stressors, SC response
to orienting stimuli (SCOR), and neuropsychological
functioning. SC testing took place in a sound-attenuated
testing room. Experimenters were blind to scores on the
psychosocial measures. Space limitations preclude
detailed descriptions of psychophysiological and neuropsychological methodology in this laboratory; these
may be found in Raine et al. (1990, 1992). The symptom
and psychosocial ratings were made by a trained research
interviewer who was blind to the results of the cognitive
or psychophysiological testing.
Psychophysiological measures. Tonic SC arousal
was indicated by (1) SC level measured at the beginning
and end of the rest period before the neuropsychological
or other psychophysiological testing and (2) the number
of spontaneous fluctuations (SF) in SC during the rest
period. SF was defined as an SC with an amplitude of
more than 0.05 microsiemens.
SC orienting stimuli consisted of a series of six tones
with an intensity of 75 decibels, a frequency of 1311 hertz
[Hz], a 25-ms rise time, and 1-second duration. Interstimulus interval was randomized between 35 and 50 seconds. SC was recorded from the first and second fingers
of the nonpreferred hand (Scerbo et al. 1992). SC was
measured using a Grass Model 7D Polygraph. A constant
voltage system (Venables and Christie 1973) was used in
configuration with Grass 7P1 preamplifier and 7DA driver
amplifiers. The gain was set at 0.1 mV/cm and highfrequency cut-off at 3 Hz. Amplification allowed resolution of all SCORs greater than 0.05 microsiemens within
a latency window of 1-3 seconds poststimulus. Two key
SCOR variables were used in this study. The first SCOR
variable consisted of the responding/nonresponding
dichotomy. Nonresponding was defined as the absence of

In addition to testing these hypotheses, exploratory

analyses were conducted to examine other relationships
between neuropsychological and psychophysiological
variables and aspects of psychosocial functioning.

Methods
A cross-sectional design was used to assess the relationships between the neuropsychological and psychophysiological variables and concurrent levels of clinical and psychosocial functioning. Symptomatology ratings were
completed on the same day as the neuropsychological and
psychophysiological testing, and psychosocial data were
gathered on the same day or generally within 2 weeks of
the testing date as part of an existing protocol.
Subjects. Forty subjects were selected from a study of
193 individuals diagnosed with schizophrenia or schizoaffective disorder and living in community-based care.
Ongoing checks of subject characteristics during sample
recruitment yielded 40 subjects who did not differ from
the larger sample on psychosocial functioning or demographic characteristics (p > 0.05 on all variables). The
sample was made up of 15 women and 25 men. The ethnic composition of the sample was 19 whites, 9 African
Americans, 7 Latinos, 1 Asian, and 4 other. The mean age
was 33.2 (standard deviation [SD] = 7.4); the mean years

21

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on November 20, 2016

of education was 12.5 (SD - 2.9). The mean age of the

first psychiatric admission was 21.4 (SD = 5.6), and the
average lifetime cumulative months of hospitalization was
10.5 (SD = 16.3, range = 0-73.2). After an initial clinical
screening for schizophrenia, study diagnoses came from
the Schedule for Affective Disorders and Schizophrenia
(Endicott and Spitzer 1978), administered by a licensed
Ph.D.-level clinician trained in the use of the instrument.
Diagnoses were made using interview data and clinical
records. Twenty-three subjects were diagnosed with schizophrenia, and 17 were diagnosed with schizoaffective disorder. All but four subjects were on maintenance dosages
of neuroleptic medication. The average dosage of neuroleptic medication (in chlorpromazine equivalents) was
400 mg.

theoretical speculation concerning social functioning in

schizophrenia. Brenner et al. (1992) and Bellack (1992)
have suggested that poor social functioning in schizophrenia might be related to attentional and cognitive deficits in
combination with dysfunctional reactions to social or
environmental stressors. Therefore, we would predict that
higher levels of social functioning will be associated with
lower SC stress reactivity, higher electrodermal orienting,
and better neuropsychological performance.
As can be seen, the literature cited above suggests
that there are patterns of relationships among neuropsychological, psychophysiological, and psychosocial variables that have not been tested. The following hypotheses
attempt to address these patterns:

Schizophrenia Bulletin, Vol. 23, No. 1, 1997

J.S. Brckkeetal.

Neuropsychological measures.

Results
Descriptive data on the study variables are presented in
table 1. The indices of skewness and kurtosis were within
acceptable levels except on the number of SCORs. This
measure reflects that 21 (53%) of the subjects were electrodermal nonresponders, having zero orienting responses.
On the BPRS, higher scores reflect higher symptom levels.
On the social, work, and independent living items, higher
scores reflect better functioning. These results suggest that
this was a moderately symptomatic outpatient sample,
with moderate residual impairment in social functioning
and independent living and greater deficits in work functioning. Table 2 presents the correlations between the functional variables; they suggest that the psychosocial variables were largely independent of one another.
Table 3 presents the correlations that were used to
test the hypotheses of this study. Given the directionality
of the hypotheses, one-tailed tests of significance were
used. In contrast, when exploratory analyses were done,
two-tailed significance tests were reported. Overall in this
study, the number of statistically significant findings far
exceeds what would be expected by chance.

Six measures of

neuropsychological functioning were used: Stroop, verbal

fluency, digit span, digit symbol, block design, and arithmetic. These measures tapped a variety of neuropsychological functions (Lezak 1983). Two of the tests (verbal
fluency and the Stroop) are more related to frontal functioning. The measure of verbal fluency consisted of the
Controlled Oral Word Association Test, which requires
the subject to name as many words as possible beginning
with C, F, and L in three separate 1-minute segments. The
key variable was the total number of acceptable words
produced across the three 1-minute conditions. On the
Stroop, the key variable consisted of the difference score
between number of color stimuli identified in 60 seconds
relative to the number of color words reported in the
Stroop interference condition. The remaining four measures, digit span, arithmetic, digit symbol, and block
design, are WAIS-R subtests that were administered using

Study Hypotheses. There was support for hypothesis 1,

which predicted that higher symptom levels in schizo-

22

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on November 20, 2016

guidelines outlined by Wechsler (1981). The digit span

and arithmetic tasks partly reflect verbal short-term working memory. Digit symbol largely reflects visuo-motor
functioning, while block design largely reflects visuo-spatial construction skills (Lezak 1983). Although not primary measures of visual information processing, three of
the tasks (digit symbol, Stroop, and block design) partly
involve visual processing. Two other measures (digit span
and arithmetic) are tests presented in the auditory domain.
Symptom and psychosocial measures. The indicators of psychosocial functioning came from two instruments: the Strauss and Carpenter Outcome Scale (SCOS;
Strauss and Carpenter 1974) and the Role Functioning
Scale (RFS; McPheeters 1984; Goodman et al. 1993).
Symptom ratings came from a 22-item version of the
Brief Psychiatric Rating Scale (BPRS; Overall and
Gorham 1962). These instruments, the training protocols,
and their current psychometric performance are discussed
elsewhere in detail (Brekke 1992; Brekke et al. 1993). In
brief, the instruments showed high interrater and interitem
reliabilities. The symptom variable was the total score
from the BPRS. The variables used as indicators of psychosocial functioning were an index of the independence
of living situation from the RFS and ratings of both work
and social functioning from the SCOS.

an SC response to all six tones. This is a more strict definition of nonresponding than the more conventional threetone criterion. We used it because it yielded a better distribution of nonresponding in this sample (53% nonresponders) than when a three-tone criterion was used (70%
nonresponders). Recently, Dawson et al. (1994) used a 12tone criterion as a strict definition of nonresponding. The
second SCOR variable was the number of SC responses
given to the orienting stimuli. Low scores on this frequency measure indicate hyporesponsivity, which is a
dimensional analog to the responding-nonresponding
dichotomy. Both of the SCOR variables were indicators
of the allocation of attention resources to external stimuli.
SC and SF measures of stress reactivity were recorded
during two stressors. The first consisted of administration
of the arithmetic subscale of the WAIS-R, which reliably
elicits strong psychophysiological responses and leads to
significant increases in heart rate (e.g., Linden 1991). This
test was administered using manual guidelines (Wechsler
1981). The second paradigm consisted of the subject being
asked to prepare a 2-minute speech about his or her faults
(Rozanski et al. 1988). This manipulation produces significantly greater physiological responsiveness than other
mental stressors (e.g., Stroop task).
The SC level index of stress reactivity was measured
as the increase in SC assessed at the end of the stress
manipulations relative to the lowest value during the resting periods. The index of SF stress reactivity was measured as the frequency of nonspecific SC responses during
the stressor period minus the frequency during rest.
Higher values indicate greater stress reactivity. A
repeated-measures analysis of variance found that the SC
stress manipulations resulted in statistically significant
increases in SC (F = 44; df = 3,34; p < 0.0001) and SF
(F = 23; df= 3,36; p < 0.0001) from resting rates.

Schizophrenia Bulletin, Vol. 23, No. 1, 1997

Psychosocial Functioning

Table 3. Correlations between

neuropsychological and psychophysiological
variables and clinical and psychosocial
functioning

Table 1. Descriptive statistics on

psyche-physiological, neuropsychological, and
functional variables
Mean

Variable

3.3
0.55

4.5

3.9
0.15

4.7

0.63

1.2

29.2
36.5

13.3
11.9

7.8
6.9
7.5
6.4

2.5
2.3
2.7
2.3

45.6

14.1

2.5
1.2
4.1

1.7

Variable

Resting arousal
SC
SF
0.302
Stress reactivity
SC
-0.39 3
SF
-0.29 2
Orienting
Responder
0.352
SCOR
Neuropsychological
Verbal fluency -0.27 2
Stroop
Digit symbol -0.45 3
Block design
Digit span
Arithmetic

1.2
1.3

BPRS

-0.04

(0.8)
Social
Work

Work
-0.28
(0.08)
-0.03
(0.8)

Work

Independent
living

0.361
0.30

-0.30 2

0.393
0.30 2
0.513
0.42 3
0.35 2

V < 0.05 (two-tailed).

2

p< 0.05 (one-tailed).

p< 0.01 (one-tailed).

There was partial support for this hypothesis in that

higher social functioning was related to lower stress reactivity and to more SC responses to orienting stimuli. In
addition, higher SC resting arousal, but none of the neuropsychological measures, was associated with better
social functioning.
Hypothesis 3 predicted that neuropsychological
deficits would be associated with poorer functioning in
work and independent living. There was some support for
this generic hypothesis. Specifically, lower levels of independent living were associated with poorer functioning on
measures of visuo-motor (digit symbol) and frontal functioning (verbal fluency and Stroop). Poorer work functioning was associated with deficits in complex visuo-spatial construction skills.
In summary, the pattern for higher symptoms was
greater tonic SC arousal, lower stress reactivity, status as
an SC responder, and deficits in verbal fluency and one
measure of visuo-motor processing. The pattern for better
social functioning was higher tonic SC arousal, lower SC
stress reactivity, and more responses to SC orienting stimuli. This similarity in the electrodermal patterns for both
better social functioning and higher symptoms will be
addressed in the next section. Poorer levels of work and

Table 2. Pearson correlations among the

clinical and psychosocial variables
Social

Social

Note.BPRS = Brief Psychiatric Rating Scale (Overall and

Gorham 1962); SC = skin conductance; SF = spontaneous fluctuations; SCOR = skin conductance orienting response.

Note.SD = standard deviation; SF = spontaneous fluctuations;

SCL = skin conductance level; SCOR = skin conductance orienting responses; BPRS = Brief Psychiatric Rating Scale (Overall
and Gorham 1962).

Variable

BPRS

Independent
living
-0.22
(0.16)
-0.06
(0.7)
0.14
(0.4)

Note.BPRS = Brief Psychiatric Rating Scale (Overall and

Gorham 1962); p values are in parentheses.

phrenia would be associated with higher resting arousal,

lower stress reactivity, status as an electrodermal responder, and deficits on neuropsychological measures of
frontal functioning. Specifically, the results indicated that
higher symptoms on the BPRS were related to higher resting SC arousal, lower SC stress reactivity, being an SC
responder, and deficits on verbal fluency and visuo-motor
processing.
Hypothesis 2 predicted that better social functioning
would be related to lower stress reactivity, better attentional allocation, and fewer neuropsychological deficits.

23

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on November 20, 2016

Psychophysiological measures
SF rest
SCL rest
0.34
SF reactivity
SCL reactivity
0.19
SCOR
Neuropsychological measures
Verbal fluency
Stroop
Digit span
Arithmetic
Block design
Digit symbol
Functional measures
BPRS
Social
Work
Independent living

Schizophrenia Bulletin, Vol. 23, No. 1, 1997

J.S. Brekkectal.

Potential Confounds. Additional analyses were done to

assess the impact of several potential confounds. First,
clinical status as reflected in the BPRS was not significantly correlated with any of the psychosocial variables;
therefore, it did not appear to be confounding the relationships between functional status and the neuropsychological or psychophysiological variables. Second, a diagnosis
of schizophrenia versus schizoaffective disorder was not
related to differences on any of the neuropsychological or
psychophysiological variables (p > 0.05 in all cases).
Third, medication dosage as reflected in chlorpromazine
equivalents (Davis et al. 1989) was not significantly
related to the neuropsychological or psychophysiological
variables (p > 0.05 in all cases). Fourth, there is evidence
to suggest that the anticholinergic potency of certain neuroleptic medications can impact electrodermal measures
(Green et al. 1989). In this sample nine subjects were on
high-potency anticholinergic neuroleptics (as defined by
Green et al. 1989). The subjects on high- or low-potency
anticholinergic neuroleptics showed no statistically significant differences on symptoms or levels of psychosocial
functioning; however, there was a significant difference
on one electrodermal measure of orienting, the number of
orienting responses, SCOR (/ = 2.5; df = 38; p < 0.05).
Finally, 27 subjects were on anticholinergic medication.
There was no difference between the subjects on or off
anticholinergics on any of the neuropsychological measures (p > 0.05 in all cases).

These results also indicated some specificity of associations between the psychosocial and the neuropsychological or psychophysiological variables. Specifically,
social functioning was related to a pattern of electrodermal variables but not to any of the neuropsychological
variables. On the other hand, work and independent living
were related only to the neuropsychological variables and
not to any electrodermal measures. This might suggest
that the capacities necessary for adequate functioning in
these areas are distinct. For example, the more intensive,
interpersonal, and reciprocally interactive nature of social

Discussion
To our knowledge, this is the first study to examine the
relationships between both neuropsychological and psychophysiological variables and distinct aspects of psychosocial functioning in schizophrenia. The results provided support for hypothesized patterns of relationships

24

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on November 20, 2016

between these variables. First, higher symptom levels

were associated with higher resting SC arousal, lower SC
stress reactivity, and status as an electrodermal responder.
Although this is the first study to examine all three of
these variables in one sample, our findings corroborate
several previous studies that examined these variables
separately (Frith et al. 1978; Zahn et al. 1981; Straube et
al. 1987; Dawson et al. 1992a, 1992b). We also corroborated Dawson et al. (1992a, 19926) in that these findings
occur in outpatient samples. We extend this previous work
by suggesting that deficits in verbal fluency (which
reflects frontal functioning) and visuo-motor processing
are possible correlates of increased symptomatology. This
pattern of findings on symptomatology suggests that
increased autonomic arousal in combination with heightened responsivity to incoming stimuli and neuropsychological deficits is clinically debilitating. This might lead to
a protective withdrawal from the environment, which is
reflected in reduced reactivity to environmental stressors.
Our findings on social functioning indicated that
lower responsivity to environmental stressors and greater
responses to SC orienting stimuli were related to higher
levels of social functioning. These findings partially support the conjecture of Brenner et al. (1992) and Bellack
(1992), who suggested that decrements in social functioning in schizophrenia are related to a combination of poor
attention, overreactivity to environmental stressors, and
cognitive deficits. There were also two unanticipated findings. First, higher resting SC arousal was related to better
social functioning; second, none of the neuropsychological variables was associated with social functioning.
Overall, this pattern of findings suggested that better
social functioning was related to higher resting arousal,
increased attentional allocation to the environment, and
lower reactivity to its stimulation. If we speculate that
higher resting arousal might reflect a readiness or ability
to engage the social environment, these results could suggest that better social functioning in schizophrenia is
related to a capacity for engaging with the environment
and good attentional allocation to its stimuli, but not overreacting to its stressors.

independent living functioning were related to deficits on

neuropsychological measures of visuo-motor, verbal, and
visuo-spatial functioning. None of the psychosocial variables was related to neuropsychological measures presented in the auditory domain.
Given the findings cited earlier that suggested that
gender influenced the relationship between electrodermal
variables and social outcomes (Ohman et al. 1989;
Wieselgren et al. 1994), we explored the impact of gender
in this sample. Although we did not have adequate sample
size to rigorously examine gender effects, gender was not
related to any of the electrodermal variables and was significantly related to only one psychosocial variable, work
functioning. This argues against a pervasive gender effect
in this sample.

Schizophrenia Bulletin, Vol. 23, No. 1, 1997

Psychosocial Functioning

25

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on November 20, 2016

phrenia (Hogarty 1988; Hogarty and Flesher 1992;

Liberman and Green 1992; Spring and Ravdin 1992;
Green 1993). This study supports Green's (1993) notion
that there are rate-limiting cognitive factors that restrict
the psychosocial functioning of patients, and that different
neuropsychological deficits might be related to distinct
aspects of psychosocial functioning. This clearly has relevance for the development of specific cognitive remediation strategies. These findings also support the contention
that cognitive remediation in schizophrenia ought to
include variables such as attention and psychophysiological arousal (Spaulding 1992).
Second, these findings can also help target the specific neuropsychological or psychophysiological deficits
that need intervention in order to facilitate gains in particular psychosocial areas (Spaulding et al. 1986; Carpenter
et al. 1988; Green 1993). For example, these results suggest that certain self-control or stress management techniques might be used to regulate autonomic arousal and
stress reactivity, and in combination with strategies for
improving attentional allocation, they could improve
aspects of social functioning. Strategies for improving
visuo-motor processing might be critical to greater independent living, while more complex visuo-spatial
processes might need to be targeted to achieve work outcomes. However, these results also suggest that remediation in schizophrenia could involve processes that interact
in complex and sometimes oppositional ways. Therefore,
rehabilitative interventions will need to be based on a
holistic perspective toward these cognitive and psychophysiological mechanisms. In this context, Spaulding
et al. (1986) and Granholm (1992) discuss specific cognitive interventions that hold promise for schizophrenia.
While we did not find that any of our neuropsychological measures were related to social functioning, there
have been several recent lab-based studies using inpatient
samples and assessments of social functioning based on
role-play (Mueser et al. 1991; Corrigan et al. 1992,
1994a, 19946; Kern et al. 1992; Penn et al. 1993, 1995;
Bowen et al. 1994). These studies have found that neuropsychological factors such as visual processing, verbal
recall, and early information processing are positively
correlated with social competence and social skills acquisition. Although this difference from our present findings
could be related to a variety of factors such as sample differences (inpatient vs. community-based), the type of
social assessments (naturalistic/global vs. role-play), and
the neuropsychological measures selected, it would be
useful at this point to investigate whether the findings
from the lab-based studies will generalize to community
settings using assessments of social functioning that
reflect day-to-day social life and competencies.

functioning might be keyed to autonomic responses and

performance. Conversely, work and independent living
performance might be tied more to instrumental tasks and
skills that require cognitive mastery.
This specificity of associations might also be indicated by the separate results on independent living and
work functioning. Higher independent living was related
to measures of visuo-motor functioning (digit symbol)
and verbal processing (verbal fluency and Stroop), with
the latter tests also partly reflecting frontal and left frontal
lobe functioning. Work functioning was related to a measure of more complex visuo-spatial processing (block
design), which in part reflects right parietal lobe functioning. Interpretations of this specificity are speculative, but
it is possible that successful work functioning requires
more complex visuo-spatial abilities than independent living performance. It should also be noted, however, that
minimal variation on the work functioning variable could
account for its association with only one of the neuropsychological measures. A more consistent finding was that
the psychosocial variables were not related to either of the
neuropsychological measures presented in the auditory
domain, whereas they did relate to the measures that
partly reflected visual processing.
One other aspect of these findings deserves attention.
The same pattern of electrodermal findings was associated
with both higher symptom levels and better social functioningnamely higher resting arousal, low stress reactivity, and more responsivity to orienting stimuli. Because
symptoms and social functioning were not inversely
related (see table 2), these findings cannot be explained in
this way. Seen in a more dynamic context, these results
suggest that the psychophysiological mechanisms related
to clinical and social improvement might be in opposition
to one another. Therefore, one could predict that increases
in social functioning might co-occur with more symptom
relapse. This prediction is consistent with findings cited in
Test (1984) and Tarrier and Turpin (1992), which suggest
that improvements in functional outcomes over time have
co-occurred with higher rates of symptomatological exacerbation. Thus, it is possible that maintaining better clinical status and improving social functioning in schizophrenia involve a balance between conflicting psychophysiological processes. This also illustrates how the complex
and sometimes divergent relationships between clinical
and psychosocial outcomes in schizophrenia might be
partially explained at the psychophysiological and neuropsychological levels. Longitudinal studies will be necessary to test these speculations.
These results have implications for psychosocial
rehabilitation in schizophrenia. First, there have been several recent discussions of cognitive remediation in schizo-

Schizophrenia Bulletin, Vol. 23, No. 1, 1997

J.S. Brekke ct al.

chronic schizophrenia: Prognosis and predictors of outcome. Archives of General Psychiatry, 48:239-245, 1991.

Concerning this study, the following caveats need to

be considered. First, although we found support for the
hypotheses using a sample of 40, these findings must be
replicated on larger samples. This will allow for multivariate tests of these findings, which can decompose the
shared variance among the variables and allow for more
rigorous tests of unique and shared effects in the patterns
we found. The variation on the work functioning item was
also limited, which could account for the paucity of neuropsychological correlates to this aspect of functioning.
There is also a need for specific theories to guide the
selection of the cognitive and psychophysiological variables and to specify which ones will be related to specific
psychosocial domains (Green 1993). At present these theories are not well developed (Spaulding et al. 1986),
although these findings could contribute to their specification. Finally, given the cross-sectional nature of this study,
it is not possible to determine the specific causal linkages
between the neuropsychological and psychophysiological
variables and psychosocial functioning in schizophrenia.
Although we have implicitly assumed that neuropsychological and psychophysiological variables influence the
levels of psychosocial functioning, it is entirely possible
that the causal direction is reversed or even reciprocal
(Green 1993).

Brekke, J. An examination of the relationships among

three outcome scales in schizophrenia. Journal of Nervous
and Mental Disease, 188:162-167, 1992.
Brekke, J.; Levin, S.; Wolkon, G.H.; Sobel, E.; and Slade,
E. Psychosocial functioning and subjective experience in
schizophrenia. Schizophrenia Bulletin, 19(3):599-608,
1993.

Buchanan, R.W.; Holstein, G; and Breier, A. The comparative efficacy and long-term effect of clozapine treatment
on neuropsychological test performance. Biological
Psychiatry, 36:717-725, 1994.
Carpenter, W.T., Jr.; Schooler, N.R.; Wise, S.S.; Goldman,
H.; Goldstein, M.J.; Hogarty, G.E.; Jeste, D.; Kane, J.M.;
Klerman, G.L.; Liberman, R.P.; Paul, S.M.; Robinson,
D.S.; Spring, B.; Stahl, S.M.; and Tamminga, C.A.
Treatment, services, and environmental factors.
Schizophrenia Bulletin, 14(3):427^t37, 1988.
Corrigan, P.W.; Green, M.F.; and Toomey, R. Cognitive
correlates to social cue perception in schizophrenia.
Psychiatry Research, 53:141-151, 1994a.

In the future, longitudinal studies are needed to test

the stability of the associations found in this study.
Prospective studies could also explore the nature of the
causal relationships between neuropsychological and psychophysiological variables and levels of psychosocial
functioning. Finally, studies that examine the relationship
of neuropsychological and psychophysiological variables
to the effectiveness of psychosocial rehabilitation in
schizophrenia will also have important implications for
the design and effectiveness of those interventions.

Corrigan, P.W.; Wallace, C.J.; and Green, M.F. Deficits in

social schemata in schizophrenia.
Schizophrenia
Research, 8:129-135, 1992.
Corrigan, P.W.; Wallace, C.J.; Green, M.F.; and Schade,
M.L. Cognitive dysfunctions and psychosocial skills
learning in schizophrenia. Behavior Therapy, 25:5-15,
1994b.
Davis, J.M.; Barter, J.T.; and Kane, J.M. Antipsychotic
drugs. In: Kaplan, H.I., and Sadock, B., eds.
Comprehensive Textbook of Psychiatry. 5th ed., Vol. 2.
Baltimore, MD: Williams & Wilkins Company, 1989.
pp. 1591-1626.

References
Allen, H. Cognitive processing and its relationship to
symptoms and social functioning in schizophrenia. British
Journal of Psychiatry, 56:201-203, 1990.

Dawson, M.E. Psychophysiology at the interface of clinical science, cognitive science, and neuroscience.
Psychophysiology, 27(3):243-255, 1990.

Bellack, A.S. Cognitive rehabilitation for schizophrenia:

Is it possible? Is it necessary? Schizophrenia Bulletin,
18(l):43-50, 1992.

Dawson, M.E.; Nuechterlein, K.H.; and Schell, A.M.

Electrodermal anomalies in recent-onset schizophrenia:
Relationships to symptoms and prognosis. Schizophrenia
Bulletin, 18(2): 295-311, 1992a.

Bowen, L.; Wallace, C.J.; Glynn, S.M.; and Nuechterlein,

K.H. Schizophrenics' cognitive functioning and performance in interpersonal interactions and skills training procedures. Journal of Psychiatric Research, 28:289-301,
1994.

Dawson, M.E.; Nuechterlein, K.H.; Schell, A.M.; Gitlin,

M.; and Ventura, J. Autonomic abnormalities in schizophrenia: State or trait indicators? Archives of General
Psychiatry, 51 (10):813-824, 1994.

Breier, A.; Schreiber, J.L.; Dyer, J.; and Pickar, D.

National Institute of Mental Health longitudinal study of

26

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on November 20, 2016

Brenner, H.D.; Hodel, B.; Roder, V.; and Corrigan, P.

Treatment of cognitive dysfunctions and behavioral
deficits in schizophrenia. Schizophrenia
Bulletin,
18(l):21-26, 1992.

Schizophrenia Bulletin, Vol. 23, No. 1, 1997

Psychosocial Functioning

of clinical neuropsychology to the study of schizophrenia.

Journal of Abnormal Psychology, 98(4):34l-356, 1989.
Lezak, M.D. Neuropsychological Assessment. 2nd ed.
New York, NY: Oxford University Press, 1983.

Dawson, M.E.; Nuechterlein, K.H.; Schell, A.M.; and

Mintz, J. Concurrent and predictive electrodermal correlates of symptomatology in recent-onset schizophrenic

patients.

Journal

of

Abnormal

Psychology,

101(1):153-164, 19926.

Liberman, R.P., and Green, M.F. Whither cognitivebehavioral therapy for schizophrenia? Schizophrenia
Bulletin, l8(l):27-35, 1992.

Dickerson, F.B.; Ringel, N.B.; and Boronow, J.J.

Neuropsychological deficits in chronic schizophrenics:
Relationships with symptoms and behavior. Journal of
Nervous and Mental Disease, 179(12): 744-749, 1991.

Linden, W. What do arithmetic stress tests measure?

Protocol variations and cardiovascular responses.
Psychophysiology, 28:91-102, 1991.

Endicott, J., and Spitzer, R. A diagnostic interview: The

schedule of affective disorders and schizophrenia.
Archives of General Psychiatry, 35:837-844, 1978.
Frith, CD.; Stevens, M.; Johnstone, E.C.; and Crow, T.J.
Skin conductance responsivity during acute episodes of
schizophrenia as a predictor of symptomatic improvement. Psychological Medicine, 8:1-6, 1978.
Gold, J.M., and Harvey, P.D. Cognitive deficits in schizo-

Mueser, K.T.; Bellack, A.S.; Douglas, M.S.; and Wade,

J.H. Prediction of social skill acquisition in schizophrenic
and major affective disorder patients from memory and
symptomatology. Psychiatry Research, 37:281-296, 1991.

phrenia. Psychiatric Clinics of North America,

16(2): 295-312, 1993.
Goodman, S.H.; Sewell, D.R.; Cooley, E.L.; and Leavitt,
N. Assessing levels of adaptive functioning: The role
functioning scale. Community Mental Health Journal,
29:119-131,1993.

Mussgay, L.; Voss, E.; Pfeiffer, H.; and Olbrich, R.

Psychophysiological reactivity to cognitive demands and
its relevance in predicting schizophrenic relapse. Journal
of Psychophysiology, 7:209-216, 1993.

Granholm, E. Processing resource limitations in schizophrenia: Implications for predicting medication response
and planning attentional training. In: Margolin, D.I., ed.
Cognitive Neuropsychology in Clinical Practice. New
York, NY: Oxford University Press, 1992. pp. 119-141.

Ohman, A.; Ohlund, L.S.; Aim, T.; Wieselgren, I.; Ost, L.;
and Lindstrom, L.H. Electrodermal nonresponding, premorbid adjustment, and symptomatology as predictors of
long-term social functioning in schizophrenics. Journal of
Abnormal Psychology, 98(4):426-435, 1989.

Green, M.F. Cognitive remediation in schizophrenia: Is it

time yet? American Journal of Psychiatry, 150:178-187,
1993.

Overall, J.E., and Gorham, D.R. The Brief Psychiatric

Rating Scale. Psychological Reports, 10:799-812, 1962.

Green, M.F.; Nuechterlein, K.H.; and Satz, P. The role of

symptomatology and medication on electrodermal activity
in schizophrenia. Psychophysiology, 26:148-157, 1989.

Penn, D.L.; Mueser, K.T.; Spaulding, W.; Hope, D.A.; and

Reed, D. Information processing and social competence
in chronic schizophrenia. Schizophrenia
Bulletin,
21(2):269-281, 1995.

Hogarty, G.E. Resistance of schizophrenic patients to

social and vocational rehabilitation. In: Dencerk, S.J., and
Kulhanek, F., eds. Treatment Resistance in Schizophrenia.
Wiesbaden, Germany: Springer-Verlag, 1988. pp. 83-97.

Penn, D.L.; Van der Does, W.; Spaulding, W.D.; Garbin,

C.P.; Linszen, D.; and Dingemans, P. Information processing and social cognitive problem solving in schizophrenia.
Journal of Nervous and Mental Disease, 181:13-20,
1993.
Raine, A.; Lencz, T.; Reynolds, G.P.; Harrison, G.;
Sheard, S.; Medley, I.; Reynolds, L.M.; and Cooper, J.E.
An evaluation of structural and functional prefrontal
deficits in schizophrenia: MRI and neuropsychological
measures. Psychiatry Research, 45:123-137, 1992.

Hogarty, G.E., and Flesher, S. Cognitive remediation in

schizophrenia: Proceed . . . with caution! Schizophrenia
Bulletin, 18(l):51-57, 1992.
Johnstone, E.C.; Macmillan, J.F.; Frith, CD.; Benn, D.K.;
and Crow, T.J. Further investigation of the predictors of
outcome following first schizophrenic episodes. British
Journal of Psychiatry, 157:182-189, 1990.
Kern, R.S.; Green, M.F.; and Satz, P. Neuropsychological
predictors of skills training for chronic psychiatric
patients. Psychiatry Research, 43:223-230, 1992.

Raine, A.; Venables, PH.; and Williams, M. Relationships

between CNS and ANS measures of arousal at age 15 and
criminality at age 24. Archives of General Psychiatry,
47:1003-1007, 1990.

Levin, S.; Yurgelun-Todd, D.; and Craft, S. Contributions

27

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on November 20, 2016

Lysaker, PH.; Bell, M.D.; and Goulet, J.L. The Wisconsin

Card Sorting Test and work performance in schizophrenia. Schizophrenia Research, 11:45-51, 1995.
McPheeters, H.L. Statewide mental health outcome evaluation: A perspective of two southern states. Community
Mental Health Journal, 20:44-55, 1984.

J.S. Brckke et al.

Schizophrenia Bulletin, Vol. 23, No. 1, 1997

Wechsler, D. Wechsler Adult Intelligence ScaleRevised.

New York, NY: The Psychological Corporation, 1981.

Rozanski, A.; Bairey, C.N.; Kranyz, D.S.; Friedman, J.;

Resser, K.J.; Morell, M.; Hilton-Chalfen, S.; Hestrin, L.;
Bistendorf, J.; and Berman, D.S. Mental stress and the
induction of silent myocardial ischemia in patients with
coronary artery disease. New England Journal of
Medicine, 318:1005-1012, 1988.

Wieselgren, I.; Ohlund, L.S.; Lindstrom, L.H.; and

Ohman, A. Electrodermal activity as a predictor of social
functioning in female schizophrenics. Journal of
Abnormal Psychology, 103:570-575, 1994.
Wykes, T. Predicting symptomatic and behavioral outcomes of community care. British Journal of Psychiatry,
165:486-492, 1994.

Scerbo, A.; Freedman, R.; Raine, A.; Dawson, M.; and

Venables, P.H. A major effect of recording site on measurement of electrodermal activity. Psychophysiology,
29:241-246, 1992.

Zahn, T.P.; Carpenter, W.T., Jr.; and McGlashan, T.H.

Autonomic nervous system activity in acute schizophrenia. Archives of General Psychiatry, 38:260-266, 1981.

Spaulding, W.D. Design prerequisites for research on cognitive therapy for schizophrenia. Schizophrenia Bulletin,
18(l):39-42, 1992.

Zahn, T.P.; Christopher, D.F.; and Steinhauser, S.R.

Autonomic functioning in schizophrenia: Electrodermal
activity, heart rate, pupillography. In: Steinhauser, S.R.;
Gruzelier, J.H.; and Zubin, J., eds. Handbook of
Schizophrenia: Neuropsychology, Psychophysiology, and
Information Processing. Vol. 5. Amsterdam, The
Netherlands: Elsevier Science Publishers, 1991.
pp. 185-224.

Spaulding, W.D.; Storms, L; Goodrich, V.; and Sullivan,

M. Applications of experimental psychopathology in psychiatric rehabilitation. Schizophrenia
Bulletin,
12(4):560-577, 1986.
Spring, B.J., and Ravdin, L. Cognitive remediation in
schizophrenia: Should we attempt it? Schizophrenia
Bulletin, 18(1): 15-20, 1992.
Straube, E.R.; Schied, H.W.; Rein, W.; and Breyer-Pfaff,
U. Autonomic nervous system differences as predictors of
short-term outcome in schizophrenics. Pharmacopsychiatry, 20:105-110, 1987.

Acknowledgments
This research was supported in part by grant MH-43640
awarded to the first author, and a Research Scientist
Development Award MH-01114 to the second author from
the National Institute of Mental Health.

Strauss, J.S., and Carpenter, W.T., Jr. The prediction of

outcome in schizophrenia: I. Characteristics of outcome.
Archives of General Psychiatry, 27:739-746, 1974.
Tarrier, N., and Turpin, G. Psychosocial factors, arousal
and schizophrenic relapse: The psychophysiological data.
British Journal of Psychiatry, 161:3-11, 1992.

The Authors

Test, M.A. Community support programs. In: Bellack,

A.S., ed. Treatment and Care for Schizophrenia. New
York, NY: Grune & Stratton, 1984. pp. 347-373.

John S. Brekke, Ph.D., is Associate Professor, School of

Social Work; Adrian Raine, D.Phil., is Professor,
Department of Psychology; Mark Ansel, M.S.W., is
Doctoral Candidate, School of Social Work; Todd Lencz,
Ph.D., is Doctoral Candidate, Department of Psychology;
and Laura Bird, B.A., is Research Assistant, Department
of Psychology, University of Southern California, Los
Angeles, CA.

Venables, P.H., and Christie, M.J. Mechanisms, instrumentation, recording techniques and quantification of
responses. In: Prokasy, W.F., and Raskin, D.C., eds.
Electrodermal Activity in Psychological Research. New
York, NY: John Wiley & Sons, 1973. pp. 1-124.

28

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ by guest on November 20, 2016

Wykes, T; Sturt, E.; and Katz, R. The prediction of rehabilitative success after three years: The use of social,
symptom and cognitive variables. British Journal of
Psychiatry, 157:865-870, 1990.

Silverstein, M.L.; Fogg, L.; and Harrow, M. Prognostic

significance of cerebral status: Dimensions of clinical outcome. Journal of Nervous and Mental
Disease,
179:534-539,1991.