- Email: [email protected]
Electrodermal responsivity distinguishes ERP activity and symptom profile in schizophrenia
Schizophrenia Research 59 (2002) 115 – 125 www.elsevier.com/locate/schres
Electrodermal responsivity distinguishes ERP activity and symptom profile in schizophrenia Leanne (Lea) M. Williams a,b,*, Homayoun Bahramali b,c, David R. Hemsley d, Anthony W.F. Harris b,c, Kerri Brown b,e, Evian Gordon b,c a
Cognitive Neuroscience Unit, Department of Psychology, University of Sydney, Sydney, NSW 2006, Australia b The Brain Dynamics Centre, Westmead Hospital, Westmead, NSW 2145, Australia c Department of Psychological Medicine, University of Sydney, Sydney, NSW 2006, Australia d Department of Psychology, Institute of Psychiatry, De Crespigny Park, London, SE5 8AZ, UK e Department of Psychology, University of Wollongong, Wollongong, NSW, Australia Received 22 February 2001; accepted 1 October 2001
Abstract Background: Traditional averaging of late component Event Related Potentials (ERPs) might obscure important psychophysiological subprocesses underlying schizophrenia disturbances in cognitive functioning. One such subprocess could be the active orientation of attention to significant or novel stimuli. In this study, we used skin conductance responses (SCRs) to index orienting responses (ORs). ERP activity was examined in relation to concomitant ORs in a schizophrenia and nonpsychiatric control group. Schizophrenia responses were considered with respect to the Reality Distortion, Disorganisation and Psychomotor Poverty syndromes. Method: Forty schizophrenia and 40 age and sex matched control subjects were tested. The three schizophrenia syndromes were derived from a principal component analysis of Positive and Negative Syndrome Scale (PANSS) ratings. Auditory ERPs (N100, N200, P200, P300) were elicited using a conventional auditory oddball paradigm, and electrodermal SCR data were acquired simultaneously. Results: ERP data were sub-averaged according to the presence/absence of an OR. For both ‘with-’ and ‘without-OR’ ERPs, schizophrenia subjects as a group showed reduced N100 (associated with vigilance level) and N200 (associated with response selection) amplitude, and for with-OR responses, they showed an additional reduction in P300 (context processing). Concerning schizophrenia syndromes, Reality Distortion was related primarily to frontal disturbances (earlier N100/N200 latency and decreased P200/P300 amplitude), and Psychomotor Poverty to a generally delayed P300 latency. Similarly delayed P300 in Disorganisation was explained by medication effects. There were no associations with syndromes for without-OR ERPs. Conclusion: These results suggest that schizophrenia syndromes are dissociated with regard to both the direction and nature of speed of information processing disturbances, in relation to taskrelevant information that produces active orientation. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Electrodermal responsivity; ERP activity; Symptom profile; Schizophrenia
*
Corresponding author. Cognitive Neuroscience Unit, Department of Psychology, University of Sydney, NSW 2006, Australia. Tel.: +61-2-9351-5750; fax: +61-2-9351-2603. E-mail address: [email protected] (L.L.M. Williams).
1. Introduction A number of studies have observed cognitive disturbances in schizophrenia using Event Related Poten-
0920-9964/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 0 - 9 9 6 4 ( 0 1 ) 0 0 3 6 8 - 1
116
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125
tial (ERP) recording (e.g., Brown et al., 2000; Ford, 1999; Matsumoto et al., 2001; McCarley et al., 1991; Williams et al., 2000). In these studies, the traditional averaging of multiple single-trial ERPs is a soundly based method for determining the late components of the ERP. However, it is likely that in paradigms designed to explore aspects of information processing, multiple psychophysiological subprocesses are engaged across the trials. These subprocesses might be obscured by an average ERP measure. One such subprocess is likely to be the orienting response (OR) to task relevant (target) stimuli. In this study, we draw on Sokolov’s (1990) definition of the OR as a distinctive index of the active orientation of attention towards significant or novel events, which can be elicited by the differentiation of relevant (target) versus irrelevant (background) information. Simultaneous measurement of ERPs and ORs would allow us to compare responses that elicit an OR to those that do not, thereby providing complementary information to the overall average. In studies of ORs using electrodermal measures, schizophrenia subjects have displayed reduced responsivity to task-relevant stimuli (Boucsein, 1992; Zahn et al., 1981), although some studies also provide evidence for hyperresponsivity (Bernstein et al., 1982), suggesting that the core impairment might be regulation of autonomic orienting. Several studies indicate that disturbances in phasic electrodermal orienting are related to cognitive-attentional factors that mediate the vulnerability to schizophrenia (Dawson et al., 1994; Hazlett et al., 1997). Neuroimaging research has suggested that reduced orienting in schizophrenia is associated with frontal brain deficits (Lencz et al., 1996), and may be particularly apparent in relation to negative symptoms (Perry et al., 1985), consistent with previously observed associations between hypofrontality and negative psychomotor poverty symptoms (Liddle, 1996). Several studies have proposed an association between the OR and the late components, including N100, N200, P200 and P300 (Hillyard et al., 1971; Na¨a¨ta¨nen and Gaillard, 1983; Polich, 1989; Roth and Kopell, 1973; Squires et al., 1975). Only a few studies have examined ERPs acquired simultaneously with ORs, assessed via skin conductance responses or SCRs (Lyytinen and Na¨a¨ta¨nen, 1987; Pritchard et al., 1986; Rosler et al., 1987; Roth and Kopell, 1973; Verbaten,
1983). These studies have used a relatively long duration (seconds) interstimulus interval (ISI) to avoid overlap of the SCRs, which have a latency of 1– 2 s and last 4– 6 s (Boucsein, 1992). Accurate scoring of SCR overlap in short ISI cognitive paradigms requires a robust quantitative model to disentangle the overlap. Our group (Lim et al., 1997) have developed a sigmoid exponential model to allow accurate scoring of overlapping SCRs in conventional short ISI paradigms, which has recently been applied in OR-ERP and ORneuroimaging studies of healthy control subjects (Bahramali et al., 1997; Lim et al., 1999a; Williams et al., 2000), and in an SCR study of schizophrenia (Lim et al., 1999b). In this study, we used the simultaneous assessment of SCRs and ERPs (N100, P200, N200, P300) in a short ISI conventional oddball paradigm to examine OR related disturbances in processing task-relevant information in individuals with schizophrenia. ERP component data were subaveraged to delineate ‘withOR’ ERPs and ‘without-OR’ ERPs. The primary aim of this study was to conduct a within-sample investigation of the associations between three syndromes of schizophrenia OR-ERPs. In an initial study of schizophrenia as a single diagnostic group, our group found that, compared to controls, schizophrenia subjects have disturbed N100, N200 and P300 amplitude for both with- and without-ORs, and additional with-OR P300 latency and without-OR P200 disturbances (Bahramali et al., 2001). Related research by our group suggests that consideration of symptom profiles unlocks significant associations with brain function that are not revealed through these traditional analyses of group averages (Harris et al., 1999; Gordon et al., 2001; Williams, 1996; Williams et al., 2000; Williams and Gordon, 2000). Three main factors have been shown to adequately account for the heterogeneity of core schizophrenic symptoms (e.g., Arndt et al., 1991; Gur et al., 1991; Liddle, 1987a; Kawasaki et al., 1994; Thompson and Meltzer, 1993). The three factors may be characterised as follows. Positive symptoms of hallucinations and delusions contribute primarily to Liddle’s (1987a) ‘Reality Distortion’ syndrome. A second factor includes positive and negative aspects of thought disorder, attentional problems (Kawasaki et al., 1994; Liddle, 1987a) and sometimes bizarre behaviour (Gur et al., 1991; Thompson and Meltzer, 1993), reflecting Lid-
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125
117
dle’s ‘Disorganisation’ syndrome. The third factor is defined by ‘deficit’ negative symptoms, such as affective flattening, emotional and social withdrawal, and avolition, representing a ‘Psychomotor Poverty’ syndrome. Liddle’s syndromes are distinguished by different patterns of performance on cognitive, neuropsychological and neuroimaging measures (Psychomotor Poverty—left prefrontal malfunction, impaired long-term memory, object naming and conceptual thinking; Reality Distortion—medial temporal lobe dysfunction, poor figure-ground perception and reduced cognitive inhibition; Disorganisation—diffuse and severe dysfunctions including impaired orientation, concentration, immediate recall, word learning and cognitive inhibition; Liddle, 1987a,b, 1996; Liddle et al., 1992; Williams, 1996). Our past qEEG and ERP studies have also differentiated these syndromes. Psychomotor Poverty was found to be associated with delayed N100 and increased beta and delta power, whereas Reality Distortion showed an opposing pattern of earlier N200 latency and increased alpha activity. Disorganisation exhibited the most global delays in N100, P200 and P300 latency to both target and background stimuli (Harris et al., 1999; Williams et al., 2000). We drew upon the above evidence for different patterns of functioning in the three schizophrenic syndromes, together with Gray et al.’s (1991) neuropsychological model of schizophrenia to formulate working hypotheses. We predicted that, for ‘with-OR’ ERPs, Psychomotor Poverty would be associated with slowed processing, shown in increased later component ERP latency and diminished amplitude (reflecting a reduced orientation to task-relevant stimuli), wherea Reality Distortion would be associated with reduced ERP latencies and amplitudes (reflecting a rapid processing of information and concomitant diminished inhibition of responses). Disorganisation would manifest the most severe dysfunctions in ERPs both with- and without-ORs.
University of Sydney and 40 gender and age (within 5 years) matched nonpsychiatric control subjects were drawn from the general population. Inclusion criteria for both groups were age of 18– 60 years and exclusion criteria were left-handedness, recent history of substance abuse, epilepsy or other neurological disorders, and mental retardation or head injury [assessed using Section M from the Composite International Diagnostic Interview (CIDI; Robins et al., 1988) and the Westmead Hospital Clinical Information Base questionnaire (WHCIB)]. Diagnosis of schizophrenia was confirmed using CIDI Section G, or by two independent psychiatrists, according to DSM-IV (American Psychiatric Association, 1994) and ICD-10 criteria. All schizophrenia subjects were in a stable, chronic phase of their illness. After interview, schizophrenic symptoms were rated using the Positive and Negative Syndrome Scale (PANSS; Kay et al., 1986). Control subjects were screened for history of psychiatric illness (themselves or first-degree relative) and treatment with psychiatric medication. The WHCIB was also used to obtain demographic information for both groups, and additional clinical data for the schizophrenia group. The schizophrenia sample comprised 29 males and 11 females, with a mean age of 35.4 years (S.D.=8.0 years). The mean chlorpromazine equivalent level of medication (Hollister, 1987; van Kammen and Marder, 1995) was 661 mg (S.D.= 637 mg) and the mean duration of illness was 12.6 years (S.D.=8.4 years). The mean PANSS Positive subscale score was 20.4 (S.D.=6.8), the mean Negative subscale score was 20.6 (S.D.= 7.1), and the mean General Psychopathology subscale score was 32.6 (S.D.= 8.4). The mean age of control subjects was 34.0 years (S.D.= 9.4). All subjects were asked to refrain from smoking or drinking caffeine for 3 h prior to the recording session. Written consent was obtained from all subjects prior to testing in accordance with National Health and Medical Research Council guidelines.
2. Methods
2.2. Data acquisition
2.1. Subjects
Subjects were seated in a dental chair in a quiet and dimly lit room with ambient temperature. A conventional auditory ‘oddball’ paradigm was employed, consisting of 40 target tones (1500 Hz) with 15%
Forty subjects with schizophrenia comprised patients or former patients from teaching hospitals of the
118
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125
probability of occurring and the remainder, background (1000 Hz) tones lasting 50 ms (with 10 ms rise and fall). The tone intensity was 80 (dBSPL) and ISI was 1.3 s. Subjects were asked to fixate on a dot on the computer screen, ignore the low (background) tones and press two reaction time buttons with the index fingers of each hand (to counterbalance motor activity) when they identified a target tone. ‘Speed and accuracy’ of response were emphasized equally. Skin conductance level, ERP and electro-oculogram (EOG) were recorded simultaneously and continuously using a 32-channel PC based system. The signals were continuously digitised at 256 Hz and stored onto a hard disk. ERPs were recorded as part of an electrophysiological battery, from 19 scalp electrode sites (F8, T4, T6, Fp2, F4, C4, P4, O2, Fz, CZ, Pz, Fp1, F3, C3, P3, O1, F7, T3 and T5). Only midline site data were used in the current analyses, since topographical differences were not the focus of the study. Data were collected according to the 10 – 20 international system (Bloom and Anneveldt, 1982) in reference to linkedear electrodes, with an amplification of 50,000 and a band pass from 0.07 to 50 Hz. Horizontal EOG was recorded via electrodes placed at the outer canthus of each eye and vertical EOG was recorded via two electrodes placed on the middle of the supraorbital and infraorbital regions of the left eye. EOG correction was carried out using a technique based on Gratton et al. (1983). Skin conductance level was recorded via a pair of silver – silver chloride electrodes with 0.05 M sodium chloride gel placed on the distal phalanges of digits II and III of the left hand. The electrode pairs as part of the input circuit were supplied by a constant voltage and therefore the current change representing conductance was recorded using a DC amplifier. The SCL (AS) was measured for each 500 ms epoch prior to each target tone. 2.3. Data scoring Average amplitude and latency for target ERP components were measured relative to a pre-stimulus (200 ms) baseline. Scoring was performed baseline to peak by an automated system based on the based on the detection of a change in the sign of the gradient of
each ERP component (Haig et al., 1995). Preset scoring criteria were that N100 occurred between 50 and 100 ms, P200 between 150 and 200 ms, N200 between 200 and 280 ms and P300 between 250 and 500 ms. Scoring revealed that one control and two schizophrenia subjects showed remaining artifact for N200 and P300 components. These data were therefore removed for these subjects and replaced with the group mean for amplitude and latency for these components. The presence of an electrodermal OR was determined by our customised software (Skin Conductance Response Evaluation System, SCORES), according to the method of Lim et al. (1997). Orienting was defined as an unambiguous increase ( > 0.05 microSiemens) in skin conductance with respect to each prestimulus baseline and occuring 1– 2 s after the stimulus (Barry, 1984; Barry and Sokolov, 1993). The segments of electrodermal data containing composite skin conductance signals and often overlapping responses were decomposed into phasic skin conductance responses (SCRs or Orienting) and tonic skin conductance level (SCL) components using our sigmoid-exponent SCR model implemented by SCORES. This model represents the SCR in mathematical form, which enables curve-fitting and analytical separation and scoring of the SCR/SCL components (see Fig. 1). 2.4. Selection of ERPS with and without associated SCRs The same 40 single epoch ERPs that were used to determine the traditional average ERP were used to derive two ERP sub-averages in each of the subjects, using the sigmoid exponential SCR model (SCORES; Lim et al., 1997). One sub-average was generated from those single-trial ERPs that were associated with an SCR to the target stimulus (the ‘with-OR’ ERP). A second ERP sub-average consisted of target ERPs that did not elicit an SCR (‘without-OR’ ERP). Where possible, we matched the number of with-OR to without-OR trials. Where there were a relatively greater number of with-OR trials, the limiting ‘ratio’ we used was a maximum of 7:1 for with-OR/withoutOR trials. That is, we excluded subjects from subsequent analyses if they had less than five or greater than or equal to 35 with-OR trials (and therefore greater than 35 or less than or equal to five without-
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125
Fig. 1. The sigmoid-exponential SCR model method (Lim et al., 1997) allows overlapping SCRs (as depicted above) to be decomposed and scored for both peak amplitude and latency (indicated by arrows). The modelled SCRs are curve-fitted to the actual SCRs using the parameters; time at which reponse starts, gain factor, rise time, time constant of the decay, constant (skin conductance level) and size of tail of previous SCR.
119
separately for ‘with-OR’ and ‘without-OR’ using twoway, repeated measures MANOVAs with group as the between-group factor, and site as the within-subjects factor. The effects of potential between-group differences in the number or amplitude of SCRs were examined in subsequent MANCOVAs. For the within-schizophrenia group analyses, independent syndromes were derived from a principal components analysis (with varimax rotation) of the PANSS Positive and Negative subscale ratings. The General Psychopathology subscale was omitted because the focus was on the relationship between performance and dimensions of positive and negative symptoms specifically, in correspondence with previous factor analyses of the primary symptom dimensions (e.g., Liddle, 1987a). Factor scores produced by the PCA were then correlated (using two-tailed Pearson analysis) with the ERP data (with- and without-OR). Given that a priori hypotheses were specified, an alpha level of 0.05 was used for correlation analyses. Post hoc partial correlation analyses were conducted to ensure that significant associations between symptom profiles and ERP data were not due to the confounding effects of medication or duration of illness.
OR trials, respectively) to constrain any potential effects of differential with-OR versus without-OR ratios. To ensure that remaining differences in these ratios did not produce different ERP patterns within the subaveraged data, we assessed the analogue ERP graphs across each individual subject within each group. In each group, 75 – 80% of subjects showed a consistent pattern of both with- and without-OR ERP responses, consistent with the typical degree of within-subject variation. 2.5. Data analysis MANOVA was used to first directly compare the with- and without-OR conditions, with both ERP latency and amplitude as the dependent measures, with group (schizophrenia versus controls) as the between-group factor, with- versus without-OR as a within-subjects factor, midline site (Fz,Cz,Pz) as an additional within-subjects factor. The amplitude and latency of the four ERP components were analysed
Fig. 2. Number of SCRs (electrodermal orienting responses) to target stimuli, depicted in bands of five, for schizophrenia and control groups.
120
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125
3. Results 3.1. Between-group analysis of ERPs with- and without-ORs Our criteria of a minumum of six and maximum of 35 SCRs to target stimuli for the formulation of withand without-OR ERP subaverages, resulted in the exclusion of 15 control subjects and eight schizophrenia subjects. The distribution of number of SCRs for the final 25 control and 32 schizophrenia subjects used in the subaveraged analyses are depicted in Fig. 2. The observation that schizophrenia subjects tended to evoke fewer SCRs than (control mean = 26.3, S.D. = 8.2; schizophrenia mean = 20.8, S.D.=8.6) was taken into account in the subsequent MANCOVA analyses. There was little difference in mean SCR amplitude for
control (mean = 0.17, S.D.= 0.09) and schizophrenia (mean= 0.14, S.D.=0.09) groups, but amplitude was also covaried in subsequent analyses. The ERP subaverages derived from the presence (with-OR) versus absence (without-OR) of SCRs are presented in analogure form in Fig. 3 for both groups across the three midline sites. 3.1.1. With- versus without-OR An interaction was observed for with- versus without-OR by site for P300 ( F(2,110) =3.45, P= 0.035), due in particular to the decreased parietal (Pz) amplitude for without-OR P300 responses (see Fig. 1). There was a significant group by site interaction for N200 ( F(2,54) = 47.86, P=0.0001), due to significantly reduced amplitude centrally Cz; ( F(1,55) = 7.99, P=0.007) within both with- and without-OR subave-
Fig. 3. ERP data for both with- and without-ORs in schizophrenia and control groups. In the with-OR ERPs, schizophrenia subjects showed reduced amplitude for N100 at Fz (A), Cz (B) and N200 at Cz (B). N200 latency was delayed for schizophrenia subjects across all sites (A, B, C). In without-OR, schizophrenia subjects showed decreased N100 and N200 amplitude at Cz (E) and Pz (F) but, P200 amplitude was increased at Cz (E).
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125
rages. A significant group main effect was also observed for N100 ( F(1,55) = 7.80, P= 0.007) and a group effect of borderline significance was revealed for P300 ( F(1,55) = 3.50, P= 0.05), reflecting reduced N100 and P300 amplitude across both with-OR and without-OR subaverages. 3.1.2. With-OR For ERP amplitude, there were significant group by midline site interactions for N100 ( F(2,55) = 6.01, P= 0.004), N200 ( F(2,50) =6.17, P=0.004) and P300 ( F(2,54) = 3.42, P= 0.036). Schizophrenic subjects showed significantly reduced amplitude fronto-centrally for both N100 (Fz; F(1,55) =7.93, P=0.007; Cz; F ( 1 , 5 5 ) = 11 . 0 8 , P = 0 . 0 0 2 ) a n d N 2 0 0 ( F z ; F(1,55) = 4.85, P = 0.032; Cz; F(1,55) = 9.73, P = 0.003), and P300 amplitude was significantly reduced in at the frontal (Fz) site ( F(1,55) = 13.43, P= 0.001) — see Fig. 3A – C. MANCOVAs showed that these group differences were not due to the counfounding effects
121
of either number of SCRs (N100, F(1,54) = .86, P= 0.357; N200, F(1,54) = 1.41, P= 0.240); P300, F(1,54) =0.27, P= 0.606) or to the SCR amplitude (N100, F(1,54) = 1.03, P= 0.315; N200, F(1,54) = 2.44, P=0.124); P300, F(1,54) = 1.08, P= 0.302). There were no significant effects involving group for ERP latency. 3.1.3. Without-OR For ERP amplitude, there were significant group by site interactions for N100 ( F(2,54) = 38.22, P= 0.0001), P200 ( F(2,54) = 5.44, P= 0.007) and N200 ( F(2,54) = 7.44, P= 0.001) — see Fig. 3D – F. N100 amplitude was significantly reduced in the schizophrenia group, relative to controls, both frontally (N100, F(1,55) = 4.85, P= 0.032) and centrally (N100, F(1,55) = 9.73, P=0.003), and N200 amplitude was reduced centrally ( F(1,55) = 5.97, P= 0.018) — see Fig. 3E. By contrast, there was a trend for P200 amplitude to be increased centrally ( F(1,55) =3.26,
Table 1 Pearson correlations (and p values) for factor scores on Psychomotor Poverty, Reality Distortion and Disorganisation dimensions and ‘with-OR’ N100, P200, N200 and P300 amplitude and latency for midline sites: significant results only With-OR ERP
Psychomotor poverty
Reality distortion
Disorganisation
Amplitude
Latency
Amplitude
Latency
N100 Fz Cz Pz
– – –
– –
– – –
P200 Fz Cz Pz
– – –
– – –
– –
N200 Fz Cz Pz
– – –
– – –
– – –
P300 Fz Cz Pz
– – –
0.48 (0.004)a 0.57 (0.000)a 0.36 (0.037)c
a
Latency
– –
– – –
– – –
– – –
– – –
– – –
–
– – –
– – –
– – –
– – –
0.35 (0.049)b – –
0.37 (0.034)a
0.45 (0.008)a
0.40 (0.021)a 0.43 (0.015)a
0.41 (0.018)a – –
Amplitude
Remained significant when CPZ equiv. medication and duration of illness partialled out. Remained significant when duration of illness partialled out, but not when CPZ equiv. medication controlled for. c Remained at borderline significance ( p = 0.051) when CPZ equiv. medication partialled out, but no longer significant when duration of illness was controlled for. b
122
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125
P = 0.07). There were no significant effects for P300 amplitude. There were also no significant effects for latency. 3.2. Within-group analysis of schizophrenia symptoms for ERPs with- and without-ORs The PCA of PANSS subscale ratings confirmed solutions revealed in our previous studies (Harris et al., 1999; Gordon et al., 2001; Williams et al., 2000). Three factors with eigenvalues >1 emerged, which explained 68.8% of the variance (Factor 1, 38.8%; Factor 2, 19.7%; Factor 3, 10.3%). Factor 1, defined primarily by deficit negative symptoms (blunted affect, emotional and social withdrawal, poor rapport), represented a Psychomotor Poverty syndrome. Factor 2 was defined by high loadings from both positive and negative aspects of thought disorder and grandiosity, and reflected the Disorganisation syndrome. The third factor, defined primarily by delusions, hallucinations and paranoia items (suspiciousness, hostility) delineated a Reality Distortion syndrome. Factor scores for each syndrome were relatively normally distributed with skewness scores of less than 1.0. Significant two-tailed correlations between factor scores and with-OR ERP amplitude and latency are presented in Table 1. As shown in Table 1, with respect to evoking an orienting response, the Reality Distortion and Psychomotor Poverty syndromes showed a distinct (and opposing) pattern of results. By contrast, when there was no orienting response, the three syndromes were not differentiated by ERP responses. For the ‘without-OR’ ERP sub-averages there were no significant correlations between ERP components and factor scores.
4. Discussion The primary aim of this study was to explore whether independent syndromes of schizophrenia would show a different pattern of neurophysiological functioning, when ERPs are subaveraged according to the presence or absence of a skin conductance orienting response (OR). The syndrome findings were considered against the background of a conventional between-group analysis of schizophrenia and control subjects. The ERP data were first considered in terms
of a traditional ERP average (evoked using an auditory oddball paradigm). With and without-OR ERPs were determined using our technique for concomitant measurement of electrodermal activity in short interstimulus interval paradigms (Bahramali et al., 1997; Lim et al., 1997, 1999a). Direct comparison of with- and without-OR subaverages showed that the presence of SCRs was associated with a relatively larger parietal P300 response, consistent with the view that the parietal P300 is an index of active orientation of attention and context processing of task-relevant mismatches (e.g., targets in the context of background stimuli; Duncan-Johnson and Donchin, 1982). Nevertheless, schizophrenia subjects exhibited a reduction in N100 (all sites), N200 (Cz) and P300 (all sites), regardless of the presence of orienting. However, more specific orienting-related disturbances in schizophrenia were fleshed out in the more fine-grained analyses of with- and withoutOR subaverages and associations with symptom profile. Decreases in P300 amplitude have been reported consistently in schizophrenia, and interpreted as a disturbance in context processing and updating of working memory (see Ford et al., 1992; Ford, 1999 for a review). Our individual analyses of with- and withoutOR subaverages indicated that the schizophrenia reduction in P300 was linked specifically to the presence of SCRs (with-OR responses), and localised to the frontal (Fz) site. As indicated in Fig. 1, control subjects showed a differential pattern of with-OR (relatively large P300) versus without-OR (smaller P300) responses, which accords with the ‘active orientation of attention’ conceptualisation of P300 modulation. This distinctive with-OR pattern could not be explained by group differences in either the number or amplitude of SCRs. By contrast, schizophrenia subjects failed to show the increase in frontal P300 amplitude in the presence of SCR orienting, consistent with previous neuroimaging evidence for an association between prefrontal lobe disturbances and orienting in schizophrenia (Lencz et al., 1996). We speculate that there is a dysregulation in schizophrenia between autonomic orienting (SCRs), which normally signals the evaluation of a target stimulus as significant or novel, and the central nervous system ({P300) contextual evaluation and working memory processing of the target stimulus.
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125
In contrast to the P300 finding, the schizophrenia reduction in N100 and N200 amplitude was observed in both the presence and absence of SCR orienting, consistent with our earlier group study (Bahramali et al., 2001). These reductions were observed at the fronto-central sites. Reductions in N100 and N200 have also been observed in previous studies of unmedicated patients, suggesting that they may be enduring traits of this disorder, along with the reduced P300 (Laurent et al., 1999). It has also been proposed that reduced frontal N100 responses may be valuable in classifying schizophrenia subjects with frontal lobe dysfunction (Winterer et al., 2000). Since the N100 has been associated with vigilance level (or initial attention) and the N200 with response selection (Ford et al., 1992; McCarley et al., 1991; Pritchard et al., 1986), these findings suggests that schizophrenia subjects have a general disturbance in sustaining attention and making responses to task-relevant information, that might arise from a trait-based frontal impairment. In the absence of an OR, schizophrenia subjects showed a trend towards increased P200 amplitude that was not apparent in the with-OR data. The reduction of the SCR reflects an inhibition of activity due to repeated novel (i.e., familiar) stimuli. Since positive components may also represent inhibitory processes (Roberts et al., 1994) the without-OR P200 effect may point to a lack of inhibition of responses to familiar or repeated stimuli in schizophrenia. This proposal is consistent with latent inhibition evidence for a failure to inhibit ‘irrelevant’ stimuli that have been pre-exposed in schizophrenia (Gray et al., 1991). The overall group average findings were extended in this study to within-group schizophrenic syndromes of Reality Distortion, Psychomotor Poverty and Disorganisation (Liddle, 1987a), derived from principal components analysis of symptom ratings. Significant results were revealed only for with-OR responses, suggesting that these syndromes might reflect distinct manifestations of a fundamental dysregulation of autonomic orienting and target stimulus processing. The with-OR ERP results for Reality Distortion and Psychomotor Poverty revealed a double dissociation in the data, suggesting that these syndromes may be associated with opposing information processing strategies (Wright and Kydd, 1986).
123
The Reality Distortion syndrome showed a consistent reduction in frontal latency for the negative ERP components (N100 and N200), and a consistent reduction in frontal amplitude for the positive components (P200 and P300). This pattern of performance could not be explained by the effects of either medication or length of illness. The finding of reduced latency in this syndrome suggests that is associated with an unusually rapid response to sensory input, consistent with Hemsley’s (1993) proposal that positive symptoms involving reality distortion reflect a tendency to ‘jump to conclusions/decisions’ on the basis of minimal information. A ‘jump to conclusions’ account would be consistent with the subsequently reduced P200 amplitude, given that the P200 component has been associated with decision making. The decrease in P300 amplitude (which parallels that for the schizophrenia group in total), points to a resulting breakdown in stimulus context. By contrast, the Psychomotor Poverty syndrome was associated with increased latency for with-OR P300 at both frontal and central sites, suggesting that early processing is normal, but that there is a retardation of the later processes of context and working memory in this syndrome. This suggestion is consistent with Hemsley’s (1994) model that negative symptoms represent an adaptive strategy, develop to deal with a long-term state of ‘information overload’. The differentiation of Reality Distortion and Psychomotor Poverty syndromes was consistent with our predictions. The observation that these syndromes were dissociated for with-OR ERPS (but not without-OR ERPs) indicates that the use of concommittant autonomic measures, to index physiological subprocesses, may be valuable in elucidating specific patterns of schizophrenia disturbance. The only significant finding for the Disorganisation syndrome (increased frontal with-OR P300 latency) was explained by medication effects. Contrary to our predictions, therefore, this syndrome was not associated with a widespread pattern of ERP disturbances. Nevertheless, since other neurophysiological and cognitive studies by our group point to a particularly severe deficit in this syndrome (Gordon et al., 2001; Williams, 1996; Williams et al., 2000), it remains possible there is a generalised disturbance in Disorganisation that is not revealed by skin conductance orienting subaverages.
124
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125
This preliminary study has demonstrated the potential to extract ERP information beyond the traditional averaged data, by simultaneously assessing electrodermal activity, allowing for a more integrated evaluation of central and autonomic abnormalities in schizophrenia. The results highlight the potential utility of symptom profiles in elucidating specific patterns of brain dysfunction in this disorder. We have recently applied the SCR subaveraging technique to functional neuroimaging (Williams et al., 2000). In future studies, neuroimaging measures might be used to help illuminate the spatio-temporal mechanisms underlying different information processing strategies in independent syndromes of schizophrenia.
References American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders, 3rd edn. revised (DSM-IV) APA, Washington, DC. Arndt, S., Alliger, R.J., Andreasen, N.C., 1991. The distinction of positive and negative symptoms. The failure of a two-dimensional model. British Journal of Psychiatry 158, 317 – 322. Bahramali, H., Gordon, E., Lim, C., Li, W., Lagopolous, J., Leslie, J., Rennie, C., Meares, R.A., 1997. Evoked related potentials associated with and without an orienting reflex. NeuroReport 8, 2665 – 2669. Bahramali, H., Lim, C., Rennie, C., Meares, R.A., Gordon, E., 2001. ERPs associated with an ‘orienting reflex’ in patients with schizophrenia. Int. J. Neurosci. 92, 44 – 55. Barry, R.J., 1984. Preliminary process in O – R elicitation. Acta Psychologica 55, 109 – 142. Barry, R.J., Sokolov, E.N., 1993. Habituation of phasic and tonic components of the orienting reflex. International Journal of Psychophysiology 15, 30 – 42. Bernstein, A.S., Frith, C.D., Gruzelier, J.H., Patterson, T., Straube, E., Venables, P.H., Zahn, T.P., 1982. An analysis of the skin conductance orienting response in samples of American, British, and German schizophrenics. Biological Psychology 14, 155 – 211. Bloom, J.L., Anneveldt, M., 1982. An electrode cap tested. Electroencephalography and Clinical Neurophysiology 54, 591 – 594. Boucsein, W., 1992. Electrodermal Activity. Plenum, New York. Brown, K., Gordon, E., Williams, L., Bahramali, H., Harris, A., Gray, J., Meares, R., 2000. Misattribution of sensory input reflected in dysfunctional target: nontarget ERPs in schizophrenia. Psychological Medicine 30, 1443 – 1449. Dawson, M.E., Nuechterlein, K.H., Schell, A.M., Gitlin, M., Ventura, J., 1994. Autonomic abnormalities in schizophrenia. State or trait indicators? Archives of General Psychiatry 51, 813 – 824. Duncan-Johnson, C.C., Donchin, E., 1982. The P300 component of
the event-related brain potential as an index of information processing. Biological Psychology 14, 1 – 52. Ford, J.M., 1999. Schizophrenia: the broken P300 and beyond. Psychophysiology 36, 667 – 682. Ford, J.M., Pfefferbaum, A., Roth, W., 1992. P3 and schizophrenia. Annals of the New York Academy of Sciences 658, 146 – 162. Gordon, E., Williams, L.M., Haig, A.R., Wright, J., Meares, R.A., 2001. Symptom profile and ‘Gamma’ processing in schizophrenia. Cognitive Neuropsychiatry 6, 7 – 19. Gratton, G., Coles, M.G., Donchin, E., 1983. A new method for offline removal of ocular artifact. Electroencephalography and Clinical Neurophysiology 554, 468 – 484. Gray, J.A., Feldon, J., Rawlins, J.N.P., Hemsley, D.R., Smith, A.D., 1991. The neuropsychology of schizophrenia. Behavioural and Brain Sciences 14, 1 – 84. Gur, R.E., Mozley, P.D., Resnik, S.M., Levick, S., Erwin, R., Saykin, A.J., Gur, R.C., 1991. Relations among clinical scales in schizophrenia. American Journal of Psychiatry 148, 472 – 478. Haig, A.R., Gordon, E., Rogers, G., Anderson, J., 1995. Classification of single-trial ERP sub-types: application of globally optimal vector quantization using simulated annealing. Electroencephalography and Clinical Neurophysiology 94, 288 – 297. Harris, A.W.F., Williams, L., Gordon, E., Bahramali, H., SlewaYounan, S., 1999. Different psychopathological models and quantified electroencephalography in schizophrenia. Psychological Medicine 29, 1175 – 1182. Hazlett, E.A., Dawson, M.E., Filion, D.L., Schell, A.M., Nuechterlein, K.H., 1997. Autonomic orienting and the allocation of processing resources in schizophrenia patients and putatively at-risk individuals. [Journal Article] Journal of Abnormal Psychology 106 (2), 171 – 181 May. Hemsley, D.R., 1993. A simple (or simplistic?) cognitive model for schizophrenia. Behaviour Research and Therapy 31, 633 – 645. Hemsley, D.R., 1994. Cognitive disturbances as the link between schizophrenic symptoms and their biological bases. Neurology, Psychiatry and Brain Research 2, 163 – 170. Hillyard, S.A., Squires, K.C., Bauer, J., Lindsay, P.H., 1971. Evoked potential correlates of auditory signal detection. Science 172, 1357 – 1360. Hollister, L.E., 1987. Psychiatric disorders. In: Speight, T.M. (Ed.), Avery’s Drug Treatment, 3rd edn. ADIS Press, Auckland. Kawasaki, Y., Maeda, Y., Sakai, N., Higashima, M., Urata, K., Yamaguchi, N., Kurachi, M., 1994. Evaluation and interpretation of symptom structures in patients with schizophrenia. Acta Psychiatrica Scandinavica 89, 399 – 404. Kay, S.R., Opler, L.A., Fiszbein, A., 1986. Positive and Negative Syndrome Scale (PANSS). Multi-Health Systems, North Tonawanda, NY. Laurent, A., Garcia-Larrea, L., d’Amato, R., Bosson, J.L., Saoud, M., Marie-Cardine, M., Maugiere, F., Dalery, J., 1999. Auditory event-related potentials and clinical scores in unmedicated schizophrenic patients. Psychiatry Research 86, 229 – 238. Lencz, T., Raine, A., Sheard, C., 1996. Neuroanatomical bases of electrodermal hypo-responding: a cluster analytic study. International Journal of Psychophysiology 22, 141 – 153. Liddle, P.F., 1987a. The symptoms of chronic schizophrenia. A re-
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125 examination of the positive – negative dichotomy. British Journal of Psychiatry 151, 145 – 151. Liddle, P.F., 1987b. Schizophrenic syndromes, cognitive performance and neurological dysfunction. Psychological Medicine 17, 49 – 57. Liddle, P.F., 1996. Syndromes in schizophrenia and thier neuropsychological and neuroanatomical correlates. In: Pantelis, C., Nelson, H.E., Barnes, T.R.E. (Eds.), Schizophrenia: A Neuropsychological Perspective. Wiley, Chichester. Liddle, P.F., Friston, K.J., Frith, C.D., Hirsch, S.R., Jones, T., Frackowiak, R.S.J., 1992. Patterns of cerebral blood flow in schizophrenia. British Journal of Psychiatry 160, 179 – 186. Lim, C.L., Rennie, C., Barry, R.J., Bahramali, H., Lazzaro, I., Manor, B., Gordon, E., 1997. Decomposing skin conductance into tonic and phasic components. International Journal of Psychophysiology 25, 97 – 109. Lim, C.L., Gordon, E., Rennie, C., Wright, J.J., Bahramali, H., Li, W., Clouston, P., Morris, J.G.L., 1999a. The dynamics of SCR, EEG and ERP activity in an oddball paradigm with short interstimulus intervals. Psychophysiology 36, 543 – 551. Lim, C.L., Gordon, E., Harris, A., Bahramali, H., Li, W.M., Manor, B., Rennie, C., 1999b. Electrodermal activity in schizophrenia: a quantitative study using a short interstimulus paradigm. Biological Psychiatry 45, 127 – 135. Lyytinen, H., Na¨a¨ta¨nen, R., 1987. Autonomic and ERP responses to deviant stimuli: analysis of covariation. In: Johnson, R., Rohrbaugh, J.W., Parasuraman, R. (Eds.), Current Trends in EventRelated Potential Research (EEG Suppl. 40). Elsevier, Amsterdam, pp. 108 – 117. Matsumoto, K., Matsuoka, H., Yamazaki, H., Sakai, H., Kato, T., Miura, N., Nakamura, M., Osakabe, K., Saito, H., Ueno, T., Sato, M., 2001. Impairment of an event-related potential correlate of memory in schizophrenia: effects of immediate and delayed word repetition. Clinical Neurophysiology 112, 662 – 673. McCarley, R.W., Faux, S.F., Shenton, M.E., Nestor, P.G., Adams, J., 1991. Event-related potentials in schizophrenia: their biological and clinical correlates and new model of schizophrenic pathophysiology. Schizophrenia Research 4, 209 – 231. Na¨a¨ta¨nen, R., Gaillard, A.W.K., 1983. The orienting reflex and the N2 deflection of the event-related potential (ERP). In: Gaillard, W., Ritter, W. (Eds.), Tutorials in ERP Research: Endogenous Components. Elsevier, Amsterdam, pp. 119 – 141. Perry, W., Felger, T., Braff, D., 1985. The relationship between skin conductance hyporesponsivity and perseverations in schizophrenia patients. Biological Psychiatry 44, 459 – 465. Polich, J., 1989. Habituation of P300 from auditory stimuli. Psychobiology 17, 19 – 28. Pritchard, W.S., Brandt, M.E., Shappell, S.A., O’Dell, T.J., Barratt, E.S., 1986. No decrement in visual P300 amplitude during extended performance of the oddball task. International Journal of Neuroscience 29, 199 – 204. Roberts, L.E., Rau, H., Lutzenberger, W., Birbaumer, N., 1994. Mapping P300 waves onto inhibition: Go/No-Go discrimination. Electroencephalography and Clinical Neurophysiology 92, 44 – 55.
125
Robins, L.N., Wing, J., Wittchen, H.-U., Helzer, J.E., Babor, T.F., Burke, J., Farmer, A., Jablenski, A., Pickens, R., Reiger, D.A., Sartorius, N., Towle, L.H., 1988. The composite international diagnostic interview. Archives of General Psychiatry 45, 1069 – 1077. Rosler, F., Hasselmann, D., Sojka, B., 1987. Central and peripheral correlates of orienting and habituation current trends in eventrelated potential research (EEG Suppl. 40). In: Johnson, R., Rohrbough, J.W., Parasuraman, R. (Eds.), Current Trends in Event-Related Potential Research (EEG Suppl. 40). Elsevier, Amsterdam, pp. 366 – 372. Roth, W.T., Kopell, B.S., 1973. P300 — an orienting reaction in the human auditory evoked response. Perceptual and Motor Skills 36, 219 – 225. Sokolov, E.N., 1990. The orienting response, and future directions of its development. Pavlovian Journal of Biological Science 25, 142 – 150. Squires, N.K., Squires, K.C., Hillyard, S., 1975. Two varieties of long-latency positive waves evoked by unpredictable auditory stimuli in man. Electroencephalography and Clinical Neurophysiology 38, 387 – 401. Thompson, P.A., Meltzer, H.Y., 1993. Positive, negative, and disorganisation factors from the schedule for affective disorders and schizophrenia and the present state examination. British Journal of Psychiatry 163, 344 – 351. van Kammen, D.P., Marder, S.R., 1995. In: Kaplan, H.I., Sadock, B.J. (Eds.), Dopamine Receptor Antagonists, 6th edn. Comprehensive Textbook of Psychiatry, vol. 2. Williams & Wilkins, Baltimore. Verbaten, M.N., 1983. The influence of information on habituation of cortical, autonomic and behavioural components of the orienting response (OR). In: Gaillard, A.W.K., Ritter, W. (Eds.), Tutorials in ERP Research: Endogenous Components. Elsevier, Amsterdam, pp. 201 – 216. Williams, L.M., 1996. ‘Cognitive inhibition’ and schizophrenic symptom subgroups. Schizophrenia Bulletin 22, 139 – 151. Williams, L.M., Gordon, E., 2000. ‘Masked’ span of apprehension performance in schizophrenic subgroups. Cognitive Neuropsychiatry 5, 37 – 52. Williams, L.M., Gordon, E., Bahramali, H., Wright, J., Meares, R., 2000. Late component ERPs are associated with distinct schizophrenic syndromes. International Journal of Neuroscience 105, 37 – 52. Winterer, G., Ziller, M., Dorn, H., Frick, K., Mulert, C., Wuebben, Y., Herrmann, W.M., 2000. Frontal dysfunction in schizophrenia – a new electrophysiological classifier for research and clinical applications. European Archives of Psychiatry and Clinical Neuroscience 250, 207 – 214. Wright, J.J., Kydd, R.R., 1986. Schizophrenia as a disorder of cerebral state-transition. Australia, New Zealand, Journal of Psychiatry 20, 167 – 178. Zahn, T.P., Carpenter Jr., W.T., McGlashan, T.H., 1981. Autonomic nervous system activity in acute schizophrenia: I. method and comparison with normal controls. Archives of General Psychiatry 38, 251 – 258.
Electrodermal responsivity distinguishes ERP activity and symptom profile in schizophrenia Leanne (Lea) M. Williams a,b,*, Homayoun Bahramali b,c, David R. Hemsley d, Anthony W.F. Harris b,c, Kerri Brown b,e, Evian Gordon b,c a
Cognitive Neuroscience Unit, Department of Psychology, University of Sydney, Sydney, NSW 2006, Australia b The Brain Dynamics Centre, Westmead Hospital, Westmead, NSW 2145, Australia c Department of Psychological Medicine, University of Sydney, Sydney, NSW 2006, Australia d Department of Psychology, Institute of Psychiatry, De Crespigny Park, London, SE5 8AZ, UK e Department of Psychology, University of Wollongong, Wollongong, NSW, Australia Received 22 February 2001; accepted 1 October 2001
Abstract Background: Traditional averaging of late component Event Related Potentials (ERPs) might obscure important psychophysiological subprocesses underlying schizophrenia disturbances in cognitive functioning. One such subprocess could be the active orientation of attention to significant or novel stimuli. In this study, we used skin conductance responses (SCRs) to index orienting responses (ORs). ERP activity was examined in relation to concomitant ORs in a schizophrenia and nonpsychiatric control group. Schizophrenia responses were considered with respect to the Reality Distortion, Disorganisation and Psychomotor Poverty syndromes. Method: Forty schizophrenia and 40 age and sex matched control subjects were tested. The three schizophrenia syndromes were derived from a principal component analysis of Positive and Negative Syndrome Scale (PANSS) ratings. Auditory ERPs (N100, N200, P200, P300) were elicited using a conventional auditory oddball paradigm, and electrodermal SCR data were acquired simultaneously. Results: ERP data were sub-averaged according to the presence/absence of an OR. For both ‘with-’ and ‘without-OR’ ERPs, schizophrenia subjects as a group showed reduced N100 (associated with vigilance level) and N200 (associated with response selection) amplitude, and for with-OR responses, they showed an additional reduction in P300 (context processing). Concerning schizophrenia syndromes, Reality Distortion was related primarily to frontal disturbances (earlier N100/N200 latency and decreased P200/P300 amplitude), and Psychomotor Poverty to a generally delayed P300 latency. Similarly delayed P300 in Disorganisation was explained by medication effects. There were no associations with syndromes for without-OR ERPs. Conclusion: These results suggest that schizophrenia syndromes are dissociated with regard to both the direction and nature of speed of information processing disturbances, in relation to taskrelevant information that produces active orientation. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Electrodermal responsivity; ERP activity; Symptom profile; Schizophrenia
*
Corresponding author. Cognitive Neuroscience Unit, Department of Psychology, University of Sydney, NSW 2006, Australia. Tel.: +61-2-9351-5750; fax: +61-2-9351-2603. E-mail address: [email protected] (L.L.M. Williams).
1. Introduction A number of studies have observed cognitive disturbances in schizophrenia using Event Related Poten-
0920-9964/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 0 - 9 9 6 4 ( 0 1 ) 0 0 3 6 8 - 1
116
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125
tial (ERP) recording (e.g., Brown et al., 2000; Ford, 1999; Matsumoto et al., 2001; McCarley et al., 1991; Williams et al., 2000). In these studies, the traditional averaging of multiple single-trial ERPs is a soundly based method for determining the late components of the ERP. However, it is likely that in paradigms designed to explore aspects of information processing, multiple psychophysiological subprocesses are engaged across the trials. These subprocesses might be obscured by an average ERP measure. One such subprocess is likely to be the orienting response (OR) to task relevant (target) stimuli. In this study, we draw on Sokolov’s (1990) definition of the OR as a distinctive index of the active orientation of attention towards significant or novel events, which can be elicited by the differentiation of relevant (target) versus irrelevant (background) information. Simultaneous measurement of ERPs and ORs would allow us to compare responses that elicit an OR to those that do not, thereby providing complementary information to the overall average. In studies of ORs using electrodermal measures, schizophrenia subjects have displayed reduced responsivity to task-relevant stimuli (Boucsein, 1992; Zahn et al., 1981), although some studies also provide evidence for hyperresponsivity (Bernstein et al., 1982), suggesting that the core impairment might be regulation of autonomic orienting. Several studies indicate that disturbances in phasic electrodermal orienting are related to cognitive-attentional factors that mediate the vulnerability to schizophrenia (Dawson et al., 1994; Hazlett et al., 1997). Neuroimaging research has suggested that reduced orienting in schizophrenia is associated with frontal brain deficits (Lencz et al., 1996), and may be particularly apparent in relation to negative symptoms (Perry et al., 1985), consistent with previously observed associations between hypofrontality and negative psychomotor poverty symptoms (Liddle, 1996). Several studies have proposed an association between the OR and the late components, including N100, N200, P200 and P300 (Hillyard et al., 1971; Na¨a¨ta¨nen and Gaillard, 1983; Polich, 1989; Roth and Kopell, 1973; Squires et al., 1975). Only a few studies have examined ERPs acquired simultaneously with ORs, assessed via skin conductance responses or SCRs (Lyytinen and Na¨a¨ta¨nen, 1987; Pritchard et al., 1986; Rosler et al., 1987; Roth and Kopell, 1973; Verbaten,
1983). These studies have used a relatively long duration (seconds) interstimulus interval (ISI) to avoid overlap of the SCRs, which have a latency of 1– 2 s and last 4– 6 s (Boucsein, 1992). Accurate scoring of SCR overlap in short ISI cognitive paradigms requires a robust quantitative model to disentangle the overlap. Our group (Lim et al., 1997) have developed a sigmoid exponential model to allow accurate scoring of overlapping SCRs in conventional short ISI paradigms, which has recently been applied in OR-ERP and ORneuroimaging studies of healthy control subjects (Bahramali et al., 1997; Lim et al., 1999a; Williams et al., 2000), and in an SCR study of schizophrenia (Lim et al., 1999b). In this study, we used the simultaneous assessment of SCRs and ERPs (N100, P200, N200, P300) in a short ISI conventional oddball paradigm to examine OR related disturbances in processing task-relevant information in individuals with schizophrenia. ERP component data were subaveraged to delineate ‘withOR’ ERPs and ‘without-OR’ ERPs. The primary aim of this study was to conduct a within-sample investigation of the associations between three syndromes of schizophrenia OR-ERPs. In an initial study of schizophrenia as a single diagnostic group, our group found that, compared to controls, schizophrenia subjects have disturbed N100, N200 and P300 amplitude for both with- and without-ORs, and additional with-OR P300 latency and without-OR P200 disturbances (Bahramali et al., 2001). Related research by our group suggests that consideration of symptom profiles unlocks significant associations with brain function that are not revealed through these traditional analyses of group averages (Harris et al., 1999; Gordon et al., 2001; Williams, 1996; Williams et al., 2000; Williams and Gordon, 2000). Three main factors have been shown to adequately account for the heterogeneity of core schizophrenic symptoms (e.g., Arndt et al., 1991; Gur et al., 1991; Liddle, 1987a; Kawasaki et al., 1994; Thompson and Meltzer, 1993). The three factors may be characterised as follows. Positive symptoms of hallucinations and delusions contribute primarily to Liddle’s (1987a) ‘Reality Distortion’ syndrome. A second factor includes positive and negative aspects of thought disorder, attentional problems (Kawasaki et al., 1994; Liddle, 1987a) and sometimes bizarre behaviour (Gur et al., 1991; Thompson and Meltzer, 1993), reflecting Lid-
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125
117
dle’s ‘Disorganisation’ syndrome. The third factor is defined by ‘deficit’ negative symptoms, such as affective flattening, emotional and social withdrawal, and avolition, representing a ‘Psychomotor Poverty’ syndrome. Liddle’s syndromes are distinguished by different patterns of performance on cognitive, neuropsychological and neuroimaging measures (Psychomotor Poverty—left prefrontal malfunction, impaired long-term memory, object naming and conceptual thinking; Reality Distortion—medial temporal lobe dysfunction, poor figure-ground perception and reduced cognitive inhibition; Disorganisation—diffuse and severe dysfunctions including impaired orientation, concentration, immediate recall, word learning and cognitive inhibition; Liddle, 1987a,b, 1996; Liddle et al., 1992; Williams, 1996). Our past qEEG and ERP studies have also differentiated these syndromes. Psychomotor Poverty was found to be associated with delayed N100 and increased beta and delta power, whereas Reality Distortion showed an opposing pattern of earlier N200 latency and increased alpha activity. Disorganisation exhibited the most global delays in N100, P200 and P300 latency to both target and background stimuli (Harris et al., 1999; Williams et al., 2000). We drew upon the above evidence for different patterns of functioning in the three schizophrenic syndromes, together with Gray et al.’s (1991) neuropsychological model of schizophrenia to formulate working hypotheses. We predicted that, for ‘with-OR’ ERPs, Psychomotor Poverty would be associated with slowed processing, shown in increased later component ERP latency and diminished amplitude (reflecting a reduced orientation to task-relevant stimuli), wherea Reality Distortion would be associated with reduced ERP latencies and amplitudes (reflecting a rapid processing of information and concomitant diminished inhibition of responses). Disorganisation would manifest the most severe dysfunctions in ERPs both with- and without-ORs.
University of Sydney and 40 gender and age (within 5 years) matched nonpsychiatric control subjects were drawn from the general population. Inclusion criteria for both groups were age of 18– 60 years and exclusion criteria were left-handedness, recent history of substance abuse, epilepsy or other neurological disorders, and mental retardation or head injury [assessed using Section M from the Composite International Diagnostic Interview (CIDI; Robins et al., 1988) and the Westmead Hospital Clinical Information Base questionnaire (WHCIB)]. Diagnosis of schizophrenia was confirmed using CIDI Section G, or by two independent psychiatrists, according to DSM-IV (American Psychiatric Association, 1994) and ICD-10 criteria. All schizophrenia subjects were in a stable, chronic phase of their illness. After interview, schizophrenic symptoms were rated using the Positive and Negative Syndrome Scale (PANSS; Kay et al., 1986). Control subjects were screened for history of psychiatric illness (themselves or first-degree relative) and treatment with psychiatric medication. The WHCIB was also used to obtain demographic information for both groups, and additional clinical data for the schizophrenia group. The schizophrenia sample comprised 29 males and 11 females, with a mean age of 35.4 years (S.D.=8.0 years). The mean chlorpromazine equivalent level of medication (Hollister, 1987; van Kammen and Marder, 1995) was 661 mg (S.D.= 637 mg) and the mean duration of illness was 12.6 years (S.D.=8.4 years). The mean PANSS Positive subscale score was 20.4 (S.D.=6.8), the mean Negative subscale score was 20.6 (S.D.= 7.1), and the mean General Psychopathology subscale score was 32.6 (S.D.= 8.4). The mean age of control subjects was 34.0 years (S.D.= 9.4). All subjects were asked to refrain from smoking or drinking caffeine for 3 h prior to the recording session. Written consent was obtained from all subjects prior to testing in accordance with National Health and Medical Research Council guidelines.
2. Methods
2.2. Data acquisition
2.1. Subjects
Subjects were seated in a dental chair in a quiet and dimly lit room with ambient temperature. A conventional auditory ‘oddball’ paradigm was employed, consisting of 40 target tones (1500 Hz) with 15%
Forty subjects with schizophrenia comprised patients or former patients from teaching hospitals of the
118
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125
probability of occurring and the remainder, background (1000 Hz) tones lasting 50 ms (with 10 ms rise and fall). The tone intensity was 80 (dBSPL) and ISI was 1.3 s. Subjects were asked to fixate on a dot on the computer screen, ignore the low (background) tones and press two reaction time buttons with the index fingers of each hand (to counterbalance motor activity) when they identified a target tone. ‘Speed and accuracy’ of response were emphasized equally. Skin conductance level, ERP and electro-oculogram (EOG) were recorded simultaneously and continuously using a 32-channel PC based system. The signals were continuously digitised at 256 Hz and stored onto a hard disk. ERPs were recorded as part of an electrophysiological battery, from 19 scalp electrode sites (F8, T4, T6, Fp2, F4, C4, P4, O2, Fz, CZ, Pz, Fp1, F3, C3, P3, O1, F7, T3 and T5). Only midline site data were used in the current analyses, since topographical differences were not the focus of the study. Data were collected according to the 10 – 20 international system (Bloom and Anneveldt, 1982) in reference to linkedear electrodes, with an amplification of 50,000 and a band pass from 0.07 to 50 Hz. Horizontal EOG was recorded via electrodes placed at the outer canthus of each eye and vertical EOG was recorded via two electrodes placed on the middle of the supraorbital and infraorbital regions of the left eye. EOG correction was carried out using a technique based on Gratton et al. (1983). Skin conductance level was recorded via a pair of silver – silver chloride electrodes with 0.05 M sodium chloride gel placed on the distal phalanges of digits II and III of the left hand. The electrode pairs as part of the input circuit were supplied by a constant voltage and therefore the current change representing conductance was recorded using a DC amplifier. The SCL (AS) was measured for each 500 ms epoch prior to each target tone. 2.3. Data scoring Average amplitude and latency for target ERP components were measured relative to a pre-stimulus (200 ms) baseline. Scoring was performed baseline to peak by an automated system based on the based on the detection of a change in the sign of the gradient of
each ERP component (Haig et al., 1995). Preset scoring criteria were that N100 occurred between 50 and 100 ms, P200 between 150 and 200 ms, N200 between 200 and 280 ms and P300 between 250 and 500 ms. Scoring revealed that one control and two schizophrenia subjects showed remaining artifact for N200 and P300 components. These data were therefore removed for these subjects and replaced with the group mean for amplitude and latency for these components. The presence of an electrodermal OR was determined by our customised software (Skin Conductance Response Evaluation System, SCORES), according to the method of Lim et al. (1997). Orienting was defined as an unambiguous increase ( > 0.05 microSiemens) in skin conductance with respect to each prestimulus baseline and occuring 1– 2 s after the stimulus (Barry, 1984; Barry and Sokolov, 1993). The segments of electrodermal data containing composite skin conductance signals and often overlapping responses were decomposed into phasic skin conductance responses (SCRs or Orienting) and tonic skin conductance level (SCL) components using our sigmoid-exponent SCR model implemented by SCORES. This model represents the SCR in mathematical form, which enables curve-fitting and analytical separation and scoring of the SCR/SCL components (see Fig. 1). 2.4. Selection of ERPS with and without associated SCRs The same 40 single epoch ERPs that were used to determine the traditional average ERP were used to derive two ERP sub-averages in each of the subjects, using the sigmoid exponential SCR model (SCORES; Lim et al., 1997). One sub-average was generated from those single-trial ERPs that were associated with an SCR to the target stimulus (the ‘with-OR’ ERP). A second ERP sub-average consisted of target ERPs that did not elicit an SCR (‘without-OR’ ERP). Where possible, we matched the number of with-OR to without-OR trials. Where there were a relatively greater number of with-OR trials, the limiting ‘ratio’ we used was a maximum of 7:1 for with-OR/withoutOR trials. That is, we excluded subjects from subsequent analyses if they had less than five or greater than or equal to 35 with-OR trials (and therefore greater than 35 or less than or equal to five without-
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125
Fig. 1. The sigmoid-exponential SCR model method (Lim et al., 1997) allows overlapping SCRs (as depicted above) to be decomposed and scored for both peak amplitude and latency (indicated by arrows). The modelled SCRs are curve-fitted to the actual SCRs using the parameters; time at which reponse starts, gain factor, rise time, time constant of the decay, constant (skin conductance level) and size of tail of previous SCR.
119
separately for ‘with-OR’ and ‘without-OR’ using twoway, repeated measures MANOVAs with group as the between-group factor, and site as the within-subjects factor. The effects of potential between-group differences in the number or amplitude of SCRs were examined in subsequent MANCOVAs. For the within-schizophrenia group analyses, independent syndromes were derived from a principal components analysis (with varimax rotation) of the PANSS Positive and Negative subscale ratings. The General Psychopathology subscale was omitted because the focus was on the relationship between performance and dimensions of positive and negative symptoms specifically, in correspondence with previous factor analyses of the primary symptom dimensions (e.g., Liddle, 1987a). Factor scores produced by the PCA were then correlated (using two-tailed Pearson analysis) with the ERP data (with- and without-OR). Given that a priori hypotheses were specified, an alpha level of 0.05 was used for correlation analyses. Post hoc partial correlation analyses were conducted to ensure that significant associations between symptom profiles and ERP data were not due to the confounding effects of medication or duration of illness.
OR trials, respectively) to constrain any potential effects of differential with-OR versus without-OR ratios. To ensure that remaining differences in these ratios did not produce different ERP patterns within the subaveraged data, we assessed the analogue ERP graphs across each individual subject within each group. In each group, 75 – 80% of subjects showed a consistent pattern of both with- and without-OR ERP responses, consistent with the typical degree of within-subject variation. 2.5. Data analysis MANOVA was used to first directly compare the with- and without-OR conditions, with both ERP latency and amplitude as the dependent measures, with group (schizophrenia versus controls) as the between-group factor, with- versus without-OR as a within-subjects factor, midline site (Fz,Cz,Pz) as an additional within-subjects factor. The amplitude and latency of the four ERP components were analysed
Fig. 2. Number of SCRs (electrodermal orienting responses) to target stimuli, depicted in bands of five, for schizophrenia and control groups.
120
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125
3. Results 3.1. Between-group analysis of ERPs with- and without-ORs Our criteria of a minumum of six and maximum of 35 SCRs to target stimuli for the formulation of withand without-OR ERP subaverages, resulted in the exclusion of 15 control subjects and eight schizophrenia subjects. The distribution of number of SCRs for the final 25 control and 32 schizophrenia subjects used in the subaveraged analyses are depicted in Fig. 2. The observation that schizophrenia subjects tended to evoke fewer SCRs than (control mean = 26.3, S.D. = 8.2; schizophrenia mean = 20.8, S.D.=8.6) was taken into account in the subsequent MANCOVA analyses. There was little difference in mean SCR amplitude for
control (mean = 0.17, S.D.= 0.09) and schizophrenia (mean= 0.14, S.D.=0.09) groups, but amplitude was also covaried in subsequent analyses. The ERP subaverages derived from the presence (with-OR) versus absence (without-OR) of SCRs are presented in analogure form in Fig. 3 for both groups across the three midline sites. 3.1.1. With- versus without-OR An interaction was observed for with- versus without-OR by site for P300 ( F(2,110) =3.45, P= 0.035), due in particular to the decreased parietal (Pz) amplitude for without-OR P300 responses (see Fig. 1). There was a significant group by site interaction for N200 ( F(2,54) = 47.86, P=0.0001), due to significantly reduced amplitude centrally Cz; ( F(1,55) = 7.99, P=0.007) within both with- and without-OR subave-
Fig. 3. ERP data for both with- and without-ORs in schizophrenia and control groups. In the with-OR ERPs, schizophrenia subjects showed reduced amplitude for N100 at Fz (A), Cz (B) and N200 at Cz (B). N200 latency was delayed for schizophrenia subjects across all sites (A, B, C). In without-OR, schizophrenia subjects showed decreased N100 and N200 amplitude at Cz (E) and Pz (F) but, P200 amplitude was increased at Cz (E).
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125
rages. A significant group main effect was also observed for N100 ( F(1,55) = 7.80, P= 0.007) and a group effect of borderline significance was revealed for P300 ( F(1,55) = 3.50, P= 0.05), reflecting reduced N100 and P300 amplitude across both with-OR and without-OR subaverages. 3.1.2. With-OR For ERP amplitude, there were significant group by midline site interactions for N100 ( F(2,55) = 6.01, P= 0.004), N200 ( F(2,50) =6.17, P=0.004) and P300 ( F(2,54) = 3.42, P= 0.036). Schizophrenic subjects showed significantly reduced amplitude fronto-centrally for both N100 (Fz; F(1,55) =7.93, P=0.007; Cz; F ( 1 , 5 5 ) = 11 . 0 8 , P = 0 . 0 0 2 ) a n d N 2 0 0 ( F z ; F(1,55) = 4.85, P = 0.032; Cz; F(1,55) = 9.73, P = 0.003), and P300 amplitude was significantly reduced in at the frontal (Fz) site ( F(1,55) = 13.43, P= 0.001) — see Fig. 3A – C. MANCOVAs showed that these group differences were not due to the counfounding effects
121
of either number of SCRs (N100, F(1,54) = .86, P= 0.357; N200, F(1,54) = 1.41, P= 0.240); P300, F(1,54) =0.27, P= 0.606) or to the SCR amplitude (N100, F(1,54) = 1.03, P= 0.315; N200, F(1,54) = 2.44, P=0.124); P300, F(1,54) = 1.08, P= 0.302). There were no significant effects involving group for ERP latency. 3.1.3. Without-OR For ERP amplitude, there were significant group by site interactions for N100 ( F(2,54) = 38.22, P= 0.0001), P200 ( F(2,54) = 5.44, P= 0.007) and N200 ( F(2,54) = 7.44, P= 0.001) — see Fig. 3D – F. N100 amplitude was significantly reduced in the schizophrenia group, relative to controls, both frontally (N100, F(1,55) = 4.85, P= 0.032) and centrally (N100, F(1,55) = 9.73, P=0.003), and N200 amplitude was reduced centrally ( F(1,55) = 5.97, P= 0.018) — see Fig. 3E. By contrast, there was a trend for P200 amplitude to be increased centrally ( F(1,55) =3.26,
Table 1 Pearson correlations (and p values) for factor scores on Psychomotor Poverty, Reality Distortion and Disorganisation dimensions and ‘with-OR’ N100, P200, N200 and P300 amplitude and latency for midline sites: significant results only With-OR ERP
Psychomotor poverty
Reality distortion
Disorganisation
Amplitude
Latency
Amplitude
Latency
N100 Fz Cz Pz
– – –
– –
– – –
P200 Fz Cz Pz
– – –
– – –
– –
N200 Fz Cz Pz
– – –
– – –
– – –
P300 Fz Cz Pz
– – –
0.48 (0.004)a 0.57 (0.000)a 0.36 (0.037)c
a
Latency
– –
– – –
– – –
– – –
– – –
– – –
–
– – –
– – –
– – –
– – –
0.35 (0.049)b – –
0.37 (0.034)a
0.45 (0.008)a
0.40 (0.021)a 0.43 (0.015)a
0.41 (0.018)a – –
Amplitude
Remained significant when CPZ equiv. medication and duration of illness partialled out. Remained significant when duration of illness partialled out, but not when CPZ equiv. medication controlled for. c Remained at borderline significance ( p = 0.051) when CPZ equiv. medication partialled out, but no longer significant when duration of illness was controlled for. b
122
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125
P = 0.07). There were no significant effects for P300 amplitude. There were also no significant effects for latency. 3.2. Within-group analysis of schizophrenia symptoms for ERPs with- and without-ORs The PCA of PANSS subscale ratings confirmed solutions revealed in our previous studies (Harris et al., 1999; Gordon et al., 2001; Williams et al., 2000). Three factors with eigenvalues >1 emerged, which explained 68.8% of the variance (Factor 1, 38.8%; Factor 2, 19.7%; Factor 3, 10.3%). Factor 1, defined primarily by deficit negative symptoms (blunted affect, emotional and social withdrawal, poor rapport), represented a Psychomotor Poverty syndrome. Factor 2 was defined by high loadings from both positive and negative aspects of thought disorder and grandiosity, and reflected the Disorganisation syndrome. The third factor, defined primarily by delusions, hallucinations and paranoia items (suspiciousness, hostility) delineated a Reality Distortion syndrome. Factor scores for each syndrome were relatively normally distributed with skewness scores of less than 1.0. Significant two-tailed correlations between factor scores and with-OR ERP amplitude and latency are presented in Table 1. As shown in Table 1, with respect to evoking an orienting response, the Reality Distortion and Psychomotor Poverty syndromes showed a distinct (and opposing) pattern of results. By contrast, when there was no orienting response, the three syndromes were not differentiated by ERP responses. For the ‘without-OR’ ERP sub-averages there were no significant correlations between ERP components and factor scores.
4. Discussion The primary aim of this study was to explore whether independent syndromes of schizophrenia would show a different pattern of neurophysiological functioning, when ERPs are subaveraged according to the presence or absence of a skin conductance orienting response (OR). The syndrome findings were considered against the background of a conventional between-group analysis of schizophrenia and control subjects. The ERP data were first considered in terms
of a traditional ERP average (evoked using an auditory oddball paradigm). With and without-OR ERPs were determined using our technique for concomitant measurement of electrodermal activity in short interstimulus interval paradigms (Bahramali et al., 1997; Lim et al., 1997, 1999a). Direct comparison of with- and without-OR subaverages showed that the presence of SCRs was associated with a relatively larger parietal P300 response, consistent with the view that the parietal P300 is an index of active orientation of attention and context processing of task-relevant mismatches (e.g., targets in the context of background stimuli; Duncan-Johnson and Donchin, 1982). Nevertheless, schizophrenia subjects exhibited a reduction in N100 (all sites), N200 (Cz) and P300 (all sites), regardless of the presence of orienting. However, more specific orienting-related disturbances in schizophrenia were fleshed out in the more fine-grained analyses of with- and withoutOR subaverages and associations with symptom profile. Decreases in P300 amplitude have been reported consistently in schizophrenia, and interpreted as a disturbance in context processing and updating of working memory (see Ford et al., 1992; Ford, 1999 for a review). Our individual analyses of with- and withoutOR subaverages indicated that the schizophrenia reduction in P300 was linked specifically to the presence of SCRs (with-OR responses), and localised to the frontal (Fz) site. As indicated in Fig. 1, control subjects showed a differential pattern of with-OR (relatively large P300) versus without-OR (smaller P300) responses, which accords with the ‘active orientation of attention’ conceptualisation of P300 modulation. This distinctive with-OR pattern could not be explained by group differences in either the number or amplitude of SCRs. By contrast, schizophrenia subjects failed to show the increase in frontal P300 amplitude in the presence of SCR orienting, consistent with previous neuroimaging evidence for an association between prefrontal lobe disturbances and orienting in schizophrenia (Lencz et al., 1996). We speculate that there is a dysregulation in schizophrenia between autonomic orienting (SCRs), which normally signals the evaluation of a target stimulus as significant or novel, and the central nervous system ({P300) contextual evaluation and working memory processing of the target stimulus.
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125
In contrast to the P300 finding, the schizophrenia reduction in N100 and N200 amplitude was observed in both the presence and absence of SCR orienting, consistent with our earlier group study (Bahramali et al., 2001). These reductions were observed at the fronto-central sites. Reductions in N100 and N200 have also been observed in previous studies of unmedicated patients, suggesting that they may be enduring traits of this disorder, along with the reduced P300 (Laurent et al., 1999). It has also been proposed that reduced frontal N100 responses may be valuable in classifying schizophrenia subjects with frontal lobe dysfunction (Winterer et al., 2000). Since the N100 has been associated with vigilance level (or initial attention) and the N200 with response selection (Ford et al., 1992; McCarley et al., 1991; Pritchard et al., 1986), these findings suggests that schizophrenia subjects have a general disturbance in sustaining attention and making responses to task-relevant information, that might arise from a trait-based frontal impairment. In the absence of an OR, schizophrenia subjects showed a trend towards increased P200 amplitude that was not apparent in the with-OR data. The reduction of the SCR reflects an inhibition of activity due to repeated novel (i.e., familiar) stimuli. Since positive components may also represent inhibitory processes (Roberts et al., 1994) the without-OR P200 effect may point to a lack of inhibition of responses to familiar or repeated stimuli in schizophrenia. This proposal is consistent with latent inhibition evidence for a failure to inhibit ‘irrelevant’ stimuli that have been pre-exposed in schizophrenia (Gray et al., 1991). The overall group average findings were extended in this study to within-group schizophrenic syndromes of Reality Distortion, Psychomotor Poverty and Disorganisation (Liddle, 1987a), derived from principal components analysis of symptom ratings. Significant results were revealed only for with-OR responses, suggesting that these syndromes might reflect distinct manifestations of a fundamental dysregulation of autonomic orienting and target stimulus processing. The with-OR ERP results for Reality Distortion and Psychomotor Poverty revealed a double dissociation in the data, suggesting that these syndromes may be associated with opposing information processing strategies (Wright and Kydd, 1986).
123
The Reality Distortion syndrome showed a consistent reduction in frontal latency for the negative ERP components (N100 and N200), and a consistent reduction in frontal amplitude for the positive components (P200 and P300). This pattern of performance could not be explained by the effects of either medication or length of illness. The finding of reduced latency in this syndrome suggests that is associated with an unusually rapid response to sensory input, consistent with Hemsley’s (1993) proposal that positive symptoms involving reality distortion reflect a tendency to ‘jump to conclusions/decisions’ on the basis of minimal information. A ‘jump to conclusions’ account would be consistent with the subsequently reduced P200 amplitude, given that the P200 component has been associated with decision making. The decrease in P300 amplitude (which parallels that for the schizophrenia group in total), points to a resulting breakdown in stimulus context. By contrast, the Psychomotor Poverty syndrome was associated with increased latency for with-OR P300 at both frontal and central sites, suggesting that early processing is normal, but that there is a retardation of the later processes of context and working memory in this syndrome. This suggestion is consistent with Hemsley’s (1994) model that negative symptoms represent an adaptive strategy, develop to deal with a long-term state of ‘information overload’. The differentiation of Reality Distortion and Psychomotor Poverty syndromes was consistent with our predictions. The observation that these syndromes were dissociated for with-OR ERPS (but not without-OR ERPs) indicates that the use of concommittant autonomic measures, to index physiological subprocesses, may be valuable in elucidating specific patterns of schizophrenia disturbance. The only significant finding for the Disorganisation syndrome (increased frontal with-OR P300 latency) was explained by medication effects. Contrary to our predictions, therefore, this syndrome was not associated with a widespread pattern of ERP disturbances. Nevertheless, since other neurophysiological and cognitive studies by our group point to a particularly severe deficit in this syndrome (Gordon et al., 2001; Williams, 1996; Williams et al., 2000), it remains possible there is a generalised disturbance in Disorganisation that is not revealed by skin conductance orienting subaverages.
124
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125
This preliminary study has demonstrated the potential to extract ERP information beyond the traditional averaged data, by simultaneously assessing electrodermal activity, allowing for a more integrated evaluation of central and autonomic abnormalities in schizophrenia. The results highlight the potential utility of symptom profiles in elucidating specific patterns of brain dysfunction in this disorder. We have recently applied the SCR subaveraging technique to functional neuroimaging (Williams et al., 2000). In future studies, neuroimaging measures might be used to help illuminate the spatio-temporal mechanisms underlying different information processing strategies in independent syndromes of schizophrenia.
References American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders, 3rd edn. revised (DSM-IV) APA, Washington, DC. Arndt, S., Alliger, R.J., Andreasen, N.C., 1991. The distinction of positive and negative symptoms. The failure of a two-dimensional model. British Journal of Psychiatry 158, 317 – 322. Bahramali, H., Gordon, E., Lim, C., Li, W., Lagopolous, J., Leslie, J., Rennie, C., Meares, R.A., 1997. Evoked related potentials associated with and without an orienting reflex. NeuroReport 8, 2665 – 2669. Bahramali, H., Lim, C., Rennie, C., Meares, R.A., Gordon, E., 2001. ERPs associated with an ‘orienting reflex’ in patients with schizophrenia. Int. J. Neurosci. 92, 44 – 55. Barry, R.J., 1984. Preliminary process in O – R elicitation. Acta Psychologica 55, 109 – 142. Barry, R.J., Sokolov, E.N., 1993. Habituation of phasic and tonic components of the orienting reflex. International Journal of Psychophysiology 15, 30 – 42. Bernstein, A.S., Frith, C.D., Gruzelier, J.H., Patterson, T., Straube, E., Venables, P.H., Zahn, T.P., 1982. An analysis of the skin conductance orienting response in samples of American, British, and German schizophrenics. Biological Psychology 14, 155 – 211. Bloom, J.L., Anneveldt, M., 1982. An electrode cap tested. Electroencephalography and Clinical Neurophysiology 54, 591 – 594. Boucsein, W., 1992. Electrodermal Activity. Plenum, New York. Brown, K., Gordon, E., Williams, L., Bahramali, H., Harris, A., Gray, J., Meares, R., 2000. Misattribution of sensory input reflected in dysfunctional target: nontarget ERPs in schizophrenia. Psychological Medicine 30, 1443 – 1449. Dawson, M.E., Nuechterlein, K.H., Schell, A.M., Gitlin, M., Ventura, J., 1994. Autonomic abnormalities in schizophrenia. State or trait indicators? Archives of General Psychiatry 51, 813 – 824. Duncan-Johnson, C.C., Donchin, E., 1982. The P300 component of
the event-related brain potential as an index of information processing. Biological Psychology 14, 1 – 52. Ford, J.M., 1999. Schizophrenia: the broken P300 and beyond. Psychophysiology 36, 667 – 682. Ford, J.M., Pfefferbaum, A., Roth, W., 1992. P3 and schizophrenia. Annals of the New York Academy of Sciences 658, 146 – 162. Gordon, E., Williams, L.M., Haig, A.R., Wright, J., Meares, R.A., 2001. Symptom profile and ‘Gamma’ processing in schizophrenia. Cognitive Neuropsychiatry 6, 7 – 19. Gratton, G., Coles, M.G., Donchin, E., 1983. A new method for offline removal of ocular artifact. Electroencephalography and Clinical Neurophysiology 554, 468 – 484. Gray, J.A., Feldon, J., Rawlins, J.N.P., Hemsley, D.R., Smith, A.D., 1991. The neuropsychology of schizophrenia. Behavioural and Brain Sciences 14, 1 – 84. Gur, R.E., Mozley, P.D., Resnik, S.M., Levick, S., Erwin, R., Saykin, A.J., Gur, R.C., 1991. Relations among clinical scales in schizophrenia. American Journal of Psychiatry 148, 472 – 478. Haig, A.R., Gordon, E., Rogers, G., Anderson, J., 1995. Classification of single-trial ERP sub-types: application of globally optimal vector quantization using simulated annealing. Electroencephalography and Clinical Neurophysiology 94, 288 – 297. Harris, A.W.F., Williams, L., Gordon, E., Bahramali, H., SlewaYounan, S., 1999. Different psychopathological models and quantified electroencephalography in schizophrenia. Psychological Medicine 29, 1175 – 1182. Hazlett, E.A., Dawson, M.E., Filion, D.L., Schell, A.M., Nuechterlein, K.H., 1997. Autonomic orienting and the allocation of processing resources in schizophrenia patients and putatively at-risk individuals. [Journal Article] Journal of Abnormal Psychology 106 (2), 171 – 181 May. Hemsley, D.R., 1993. A simple (or simplistic?) cognitive model for schizophrenia. Behaviour Research and Therapy 31, 633 – 645. Hemsley, D.R., 1994. Cognitive disturbances as the link between schizophrenic symptoms and their biological bases. Neurology, Psychiatry and Brain Research 2, 163 – 170. Hillyard, S.A., Squires, K.C., Bauer, J., Lindsay, P.H., 1971. Evoked potential correlates of auditory signal detection. Science 172, 1357 – 1360. Hollister, L.E., 1987. Psychiatric disorders. In: Speight, T.M. (Ed.), Avery’s Drug Treatment, 3rd edn. ADIS Press, Auckland. Kawasaki, Y., Maeda, Y., Sakai, N., Higashima, M., Urata, K., Yamaguchi, N., Kurachi, M., 1994. Evaluation and interpretation of symptom structures in patients with schizophrenia. Acta Psychiatrica Scandinavica 89, 399 – 404. Kay, S.R., Opler, L.A., Fiszbein, A., 1986. Positive and Negative Syndrome Scale (PANSS). Multi-Health Systems, North Tonawanda, NY. Laurent, A., Garcia-Larrea, L., d’Amato, R., Bosson, J.L., Saoud, M., Marie-Cardine, M., Maugiere, F., Dalery, J., 1999. Auditory event-related potentials and clinical scores in unmedicated schizophrenic patients. Psychiatry Research 86, 229 – 238. Lencz, T., Raine, A., Sheard, C., 1996. Neuroanatomical bases of electrodermal hypo-responding: a cluster analytic study. International Journal of Psychophysiology 22, 141 – 153. Liddle, P.F., 1987a. The symptoms of chronic schizophrenia. A re-
L.L.M. Williams et al. / Schizophrenia Research 59 (2002) 115–125 examination of the positive – negative dichotomy. British Journal of Psychiatry 151, 145 – 151. Liddle, P.F., 1987b. Schizophrenic syndromes, cognitive performance and neurological dysfunction. Psychological Medicine 17, 49 – 57. Liddle, P.F., 1996. Syndromes in schizophrenia and thier neuropsychological and neuroanatomical correlates. In: Pantelis, C., Nelson, H.E., Barnes, T.R.E. (Eds.), Schizophrenia: A Neuropsychological Perspective. Wiley, Chichester. Liddle, P.F., Friston, K.J., Frith, C.D., Hirsch, S.R., Jones, T., Frackowiak, R.S.J., 1992. Patterns of cerebral blood flow in schizophrenia. British Journal of Psychiatry 160, 179 – 186. Lim, C.L., Rennie, C., Barry, R.J., Bahramali, H., Lazzaro, I., Manor, B., Gordon, E., 1997. Decomposing skin conductance into tonic and phasic components. International Journal of Psychophysiology 25, 97 – 109. Lim, C.L., Gordon, E., Rennie, C., Wright, J.J., Bahramali, H., Li, W., Clouston, P., Morris, J.G.L., 1999a. The dynamics of SCR, EEG and ERP activity in an oddball paradigm with short interstimulus intervals. Psychophysiology 36, 543 – 551. Lim, C.L., Gordon, E., Harris, A., Bahramali, H., Li, W.M., Manor, B., Rennie, C., 1999b. Electrodermal activity in schizophrenia: a quantitative study using a short interstimulus paradigm. Biological Psychiatry 45, 127 – 135. Lyytinen, H., Na¨a¨ta¨nen, R., 1987. Autonomic and ERP responses to deviant stimuli: analysis of covariation. In: Johnson, R., Rohrbaugh, J.W., Parasuraman, R. (Eds.), Current Trends in EventRelated Potential Research (EEG Suppl. 40). Elsevier, Amsterdam, pp. 108 – 117. Matsumoto, K., Matsuoka, H., Yamazaki, H., Sakai, H., Kato, T., Miura, N., Nakamura, M., Osakabe, K., Saito, H., Ueno, T., Sato, M., 2001. Impairment of an event-related potential correlate of memory in schizophrenia: effects of immediate and delayed word repetition. Clinical Neurophysiology 112, 662 – 673. McCarley, R.W., Faux, S.F., Shenton, M.E., Nestor, P.G., Adams, J., 1991. Event-related potentials in schizophrenia: their biological and clinical correlates and new model of schizophrenic pathophysiology. Schizophrenia Research 4, 209 – 231. Na¨a¨ta¨nen, R., Gaillard, A.W.K., 1983. The orienting reflex and the N2 deflection of the event-related potential (ERP). In: Gaillard, W., Ritter, W. (Eds.), Tutorials in ERP Research: Endogenous Components. Elsevier, Amsterdam, pp. 119 – 141. Perry, W., Felger, T., Braff, D., 1985. The relationship between skin conductance hyporesponsivity and perseverations in schizophrenia patients. Biological Psychiatry 44, 459 – 465. Polich, J., 1989. Habituation of P300 from auditory stimuli. Psychobiology 17, 19 – 28. Pritchard, W.S., Brandt, M.E., Shappell, S.A., O’Dell, T.J., Barratt, E.S., 1986. No decrement in visual P300 amplitude during extended performance of the oddball task. International Journal of Neuroscience 29, 199 – 204. Roberts, L.E., Rau, H., Lutzenberger, W., Birbaumer, N., 1994. Mapping P300 waves onto inhibition: Go/No-Go discrimination. Electroencephalography and Clinical Neurophysiology 92, 44 – 55.
125
Robins, L.N., Wing, J., Wittchen, H.-U., Helzer, J.E., Babor, T.F., Burke, J., Farmer, A., Jablenski, A., Pickens, R., Reiger, D.A., Sartorius, N., Towle, L.H., 1988. The composite international diagnostic interview. Archives of General Psychiatry 45, 1069 – 1077. Rosler, F., Hasselmann, D., Sojka, B., 1987. Central and peripheral correlates of orienting and habituation current trends in eventrelated potential research (EEG Suppl. 40). In: Johnson, R., Rohrbough, J.W., Parasuraman, R. (Eds.), Current Trends in Event-Related Potential Research (EEG Suppl. 40). Elsevier, Amsterdam, pp. 366 – 372. Roth, W.T., Kopell, B.S., 1973. P300 — an orienting reaction in the human auditory evoked response. Perceptual and Motor Skills 36, 219 – 225. Sokolov, E.N., 1990. The orienting response, and future directions of its development. Pavlovian Journal of Biological Science 25, 142 – 150. Squires, N.K., Squires, K.C., Hillyard, S., 1975. Two varieties of long-latency positive waves evoked by unpredictable auditory stimuli in man. Electroencephalography and Clinical Neurophysiology 38, 387 – 401. Thompson, P.A., Meltzer, H.Y., 1993. Positive, negative, and disorganisation factors from the schedule for affective disorders and schizophrenia and the present state examination. British Journal of Psychiatry 163, 344 – 351. van Kammen, D.P., Marder, S.R., 1995. In: Kaplan, H.I., Sadock, B.J. (Eds.), Dopamine Receptor Antagonists, 6th edn. Comprehensive Textbook of Psychiatry, vol. 2. Williams & Wilkins, Baltimore. Verbaten, M.N., 1983. The influence of information on habituation of cortical, autonomic and behavioural components of the orienting response (OR). In: Gaillard, A.W.K., Ritter, W. (Eds.), Tutorials in ERP Research: Endogenous Components. Elsevier, Amsterdam, pp. 201 – 216. Williams, L.M., 1996. ‘Cognitive inhibition’ and schizophrenic symptom subgroups. Schizophrenia Bulletin 22, 139 – 151. Williams, L.M., Gordon, E., 2000. ‘Masked’ span of apprehension performance in schizophrenic subgroups. Cognitive Neuropsychiatry 5, 37 – 52. Williams, L.M., Gordon, E., Bahramali, H., Wright, J., Meares, R., 2000. Late component ERPs are associated with distinct schizophrenic syndromes. International Journal of Neuroscience 105, 37 – 52. Winterer, G., Ziller, M., Dorn, H., Frick, K., Mulert, C., Wuebben, Y., Herrmann, W.M., 2000. Frontal dysfunction in schizophrenia – a new electrophysiological classifier for research and clinical applications. European Archives of Psychiatry and Clinical Neuroscience 250, 207 – 214. Wright, J.J., Kydd, R.R., 1986. Schizophrenia as a disorder of cerebral state-transition. Australia, New Zealand, Journal of Psychiatry 20, 167 – 178. Zahn, T.P., Carpenter Jr., W.T., McGlashan, T.H., 1981. Autonomic nervous system activity in acute schizophrenia: I. method and comparison with normal controls. Archives of General Psychiatry 38, 251 – 258.