Elsevier

Psychiatry Research

Volume 88, Issue 1, 18 October 1999, Pages 25-39
Psychiatry Research

Electrodermal activation in first-episode psychotic patients and their first-degree relatives

https://doi.org/10.1016/S0165-1781(99)00071-2Get rights and content

Abstract

We hypothesized that electrodermal deviations evident in patients with schizophrenia would also be present in their biological relatives and examined the specificity of abnormal EDA to schizophrenia patients and their families. One hundred and thirty-five first-episode psychotic patients with either schizophrenia or other psychotic disorders; 104 non-psychiatric comparison subjects; 178 relatives of these subjects; and a comparison group of 61 patients with chronic schizophrenia had their EDA monitored while they listened to auditory stimuli. Electrodermal non-responding, regardless of the nature of the stimulus, was common to all patient groups and tended to run in families. However, non-responding did not differentiate the relatives of the psychotic patients from those of non-psychiatric subjects. Responders in both the chronic and first-episode schizophrenia patients showed an excessively high rate of non-specific fluctuations (NSFs), as did the first-degree relatives of the first-episode patients. Patients with major depression had more NSFs than normal, but significantly so only during one of the tone series. Their relatives, however, had a high NSF rate in both tone series. The results indicate that a high NSF rate may represent a psychophysiological marker of risk for schizophrenia and psychotic depression. Electrodermal non-responding is not specific to schizophrenia and is not likely to be useful as an indicator of genetic risk.

Introduction

One of the most consistently observed biological deviations associated with schizophrenia involves disturbed electrodermal activation. Electrodermal activity (EDA) is typically studied using habituation paradigms in which subjects are repeatedly presented with a simple stimulus such as a pure tone while their electrodermal orienting response to the tones is monitored. Although various electrodermal anomalies have been identified, two findings in particular have shown a robust association with schizophrenia. One, termed ‘non-responding,’ is indicated by a failure of the stimuli to elicit an electrodermal reaction. Iacono et al. (1993) identified 20 published studies examining this phenomenon in schizophrenia. All but one found at least a trend indicating higher rates of non-responsiveness in schizophrenia patients than in comparison subjects. Pooling data across studies revealed that on average 43% of schizophrenia patients failed to respond to orienting stimuli. This figure was over three times higher than the rate of non-responding observed in comparison subjects (14%).

The second anomaly that has shown a robust association to schizophrenia is a high rate of spontaneous electrodermal responses. These events, often referred to as ‘non-specific fluctuations’ (NSFs), consist of electrodermal responses that occur at times other than those when stimuli are presented. Although there are exceptions (e.g. Levinson, 1991), schizophrenia is typically associated with excess NSFs, especially in those patients who are electrodermal responders (Gruzelier and Venables, 1972, Rubens and Lapidus, 1978, Straube, 1979, Zahn et al., 1981, Iacono, 1982, Öhman et al., 1989, Dawson et al., 1994).

Although the precise significance of these findings is not known, EDA is believed to provide an index of arousal, affect, attention, and information processing, all of which have been hypothesized to be abnormal in schizophrenia. Various parts of the brain implicated in the etiology of schizophrenia appear to govern EDA. Lesion studies with non-human primates have shown that the amygdala, hippocampus, and frontal cortex play a role in the production of electrodermal responses (Bagshaw et al., 1965). Recent studies using neuropsychological tests (Katsanis and Iacono, 1992), neurological patients with well characterized brain lesions (Tranel and Damasio, 1994), positron emission tomography (Hazlett et al., 1993), and magnetic resonance imaging (Raine et al., 1991, Lencz et al., 1996) have provided evidence that the frontal and temporal lobes play a role in electrodermal responding and non-responding. In addition, attentional dysfunction (Hazlett et al., 1997) and early life exposure to adverse environmental conditions likely to affect CNS development have also been associated with electrodermal hyporesponsivity (Öhlund et al., 1990, Öhlund et al., 1991, Katsanis et al., 1992, Schnur et al., 1995, Venables, 1997, Hultman and Öhman, 1998).

Several lines of evidence suggest that EDA has the potential to identify genetic vulnerability for schizophrenia. Hollister et al. (1994) examined NSFs in the children of schizophrenia patients who were part of the Copenhagen Schizophrenia High Risk Study. These children underwent an electrodermal assessment when the mean age of the group was 15.1 years. Based on their parents’ psychopathology, they were divided into risk groups composed of those with normal parents, a mother afflicted with schizophrenia, or two parents with schizophrenia spectrum diagnoses. The high risk subjects, especially those with two affected parents, produced more NSFs than the low risk subjects and also generated more NSFs over subsequent blocks of recording time than the low risk subjects. This latter effect was interpreted as a failure of the high risk subjects to show NSF habituation over time. These same subjects were reclassified into groups based on their psychiatric status as adults. Those who developed schizophrenia, compared to those with no mental disorder, tended to have a higher overall NSF rate and showed significantly less NSF habituation when they were tested as children. These findings thus indicate both that deviant EDA can be observed in the non-adult children of schizophrenia patients and that it predicts their subsequent development of schizophrenia.

Further evidence of the potential of EDA to serve as a biological marker stems from studies of normal twins, all of which indicate that genes play a role in determining EDA. In the only study (Lykken et al., 1987) to include a relatively large twin sample (n=121 pairs), including twins reared apart, genes were estimated to account for more than 50% of the variance in three-fourths of the EDA measures examined. Because 37 of the twins were classified as EDA non-responders, it was possible to estimate the monozygotic and dizygotic twin concordances for non-responding. They were, respectively, 75% and 13%, indicating that non-responding has a genetic basis.

Although EDA seems to hold considerable promise as a vulnerability indicator, it is unclear whether the vulnerability is specific to schizophrenia or applies as well to other severe forms of psychopathology. Hollister et al. (1994) found that the EDA of high-risk subjects who ultimately developed non-schizophrenic Axis I disorders failed to differentiate them from those who developed schizophrenia. This psychiatric group generated patterns of NSF activity that placed them between the schizophrenia and normal groups. Examining the adolescent and young adult offspring of bipolar probands, Zahn et al. (1989) failed to uncover strong support that EDA may index genetic risk for this disorder. However, Kugelmass et al. (1995) found that reduced skin conductance responsivity in adolescence predicted ‘affective spectrum disorder’ in young adulthood, suggesting that EDA may have the potential to predict the development of mood disorders. Further study of the relatives of those with non-schizophrenic psychoses thus appears important, especially in light of findings that mood disorder patients, in particular those with major depression, have been found to have EDA deviations similar to those seen in schizophrenia patients, including excessive rates of non-responding (Mirken and Coppen, 1980, Iacono et al., 1983, Iacono et al., 1984, Bernstein et al., 1988, Levinson, 1991).

We extended this line of research in the present investigation by conducting a comprehensive examination of EDA in psychotic patients, their siblings, and their parents. To insure that our results were broadly representative, instead of relying on a single source to recruit patients, we examined an epidemiologically based sample drawn from a large metropolitan area. We were able to address the specificity of EDA dysfunction to schizophrenia by including patients with bipolar and major depressive disorder with psychotic features in our study sample. To our knowledge, ours is the first study to examine EDA in bipolar patients when they are acutely ill. By focusing on first episode cases, we were able to avoid some of the pitfalls that arise when studying chronic patients whose EDA could reflect the cumulative effects of living with schizophrenia and its treatments. However, because most studies have relied on chronic patients, to make our investigation comparable to others, we included a group of chronic patients who were similar in age to our first-episode subjects. Because the rate of electrodermal non-responding in normal comparison subjects has been found to vary with the source from which they were drawn (e.g. see Iacono, 1991), we took special care to recruit normal subjects and their relatives from the same community from which the patients and their relatives were drawn.

Bernstein et al., 1988, Bernstein et al., 1995 and others (see Iacono et al., 1993, for a review) have presented data indicating that stimulus intensity and meaningfulness may influence the electrodermal orienting response and non-responding in schizophrenia, and that this effect is quite different from that observed in major depression. This line of investigation suggests that the rate of non-responding is high and very similar in patients with schizophrenia and major depression when they are presented with stimuli that have little significance. However, when the psychological impact of the stimuli is augmented by making them more intense or meaningful (e.g. by making them louder or having the stimuli signal a task requirement), patients with schizophrenia have been found to respond normally while non-responding persists in those with major depression. Bernstein et al., 1988, Bernstein et al., 1995 have interpreted these results as indicating that schizophrenia patients have an information-processing deficit that can be corrected by making the stimuli more salient. In contrast, depressed patients have been hypothesized to possess a structural defect, perhaps reflecting dysfunctional neurotransmission in the cholinergically mediated EDA system. Given these findings and their apparent importance to the differentiation of schizophrenia from major depression, we measured EDA using stimuli that differed both in intensity and in psychological meaningfulness.

A preliminary report examining the electrodermal non-responding of some of the participants in the present study can be found elsewhere (Iacono et al., 1993). Iacono et al. (1993) examined only one electrodermal variable, non-responding, and only first-episode psychotic patients. The current investigation extends that of Iacono et al. (1993) by: (a) examining EDA in the relatives of the first-episode and non-psychiatric comparison subjects; (b) including a chronic schizophrenia comparison sample; and (c) providing a comprehensive analysis of electrodermal measures, including NSF rate, habituation rate, response amplitude, half-recovery time, and tonic skin conductance levels. In addition, the results of the Iacono et al. report were based on the diagnoses of the first-episode patients that were made from an interview conducted at the time of their intake into the study. For the current investigation, the diagnoses were based on additional information obtained from structured interviews conducted 9 and 18 months after study intake. Hence, these diagnoses were derived from clinical data aggregated over the initial 18-month course of disorder. During this time, most of the patients with a provisional diagnosis of schizophreniform disorder could be confirmed as having schizophrenia (see Beiser et al., 1988) and many of those diagnosed as having major depression, because they subsequently had a manic episode, were re-diagnosed as having bipolar disorder.

Section snippets

First-episode cases

All of the first-episode psychotic subjects were recruited as part of the Markers and Predictors (MAP) project described in more detail elsewhere (Iacono and Beiser, 1989). A major goal of the MAP study was to identify all individuals experiencing a first episode of non-organic psychosis in the Vancouver, Canada, metropolitan area over a 2.5-year interval. Subjects were recruited from a wide variety of sources, including psychiatric hospitals, psychiatric wards of general hospitals, counseling

Results

To simplify the presentation of the data, preliminary analyses were carried out to determine how age was correlated with non-responding, habituation rate, amplitude and recovery time of the first tone-elicited SCR, the mean SCL, and the frequency of NSFs separately for the two tone series. Because psychopathology is confounded with age in the patient groups, we examined these correlations only in the 104 normal subjects. None of the correlations was significant.

Discussion

Our results are consistent with the hypothesis that EDA abnormalities may indicate genetic vulnerability for schizophrenia. The NSF rate of subjects who produced orienting responses to the auditory stimuli has obvious marker potential. This variable differentiated both the chronic and first-episode schizophrenia patients from the normal subjects. It also differentiated the first-degree relatives of the first-episode patients from the relatives of the normal subjects. Our findings are thus

References (44)

  • A. Bernstein et al.

    Schizophrenia is associated with altered orienting activity; depression with electrodermal (cholinergic?) deficit and normal orienting response

    Journal of Abnormal Psychology

    (1988)
  • A.S. Bernstein et al.

    Differing patterns of electrodermal and finger pulse responsivity in schizophrenia and depression

    Psychological Medicine

    (1995)
  • M. Buchsbaum et al.

    Stimulus intensity control and the cortical evoked response

    Psychosomatic Medicine

    (1968)
  • M. Dawson et al.

    Electrodermal anomalies in recent-onset schizophrenia, relationships to symptoms and prognosis

    Schizophrenia Bulletin

    (1992)
  • M.E. Dawson et al.

    Autonomic abnormalities in schizophrenia. State or trait indicators?

    Archives of General Psychiatry

    (1994)
  • D.C. Fowles et al.

    Committee report. Publication recommendations for electrodermal measurements

    Psychophysiology

    (1981)
  • J. Gruzelier et al.

    Skin conductance orienting activity in a heterogeneous sample of schizophrenics: possible evidence of limbic dysfunction

    Journal of Nervous and Mental Disease

    (1972)
  • E.A. Hazlett et al.

    Reduced regional brain glucose metabolism assessed by positron emission tomography in electrodermal non-responder schizophrenics: a pilot study

    Journal of Abnormal Psychology

    (1993)
  • E.A. Hazlett et al.

    Autonomic orienting and the allocation of processing resources in schizophrenia patients and putatively at-risk individuals

    Journal of Abnormal Psychology

    (1997)
  • J.M. Hollister et al.

    Impaired autonomic nervous system-habituation in those at genetic risk for schizophrenia

    Archives of General Psychiatry

    (1994)
  • W. Iacono

    Bilateral electrodermal habituation–dishabituation and resting EEG in remitted schizophrenics

    Journal of Nervous and Mental Disease

    (1982)
  • W.G. Iacono

    Control groups in schizophrenia research: a neglected source of variability

  • Cited by (0)

    View full text