- Email: [email protected]
Emotional Stroop performance predicts disorganization in schizophrenia
Schizophrenia Research 77 (2005) 141 – 149 www.elsevier.com/locate/schres
Emotional Stroop performance predicts disorganization in schizophrenia Laura K. Phillipsa,T, Patricia J. Deldinb, Martina M. Voglmaierc, Sarah Rabbittd a Department of Psychology, Harvard University, 33 Kirkland Street Rm. 1205, Cambridge, MA, 02138, USA Department of Psychology, University of Michigan, Ann Arbor, 525 E. University, 2252 East Hall, Ann Arbor, MI, 48109, USA c Department of Psychiatry, Harvard Medical School, Cambridge Health Alliance, 1493 Cambridge St., Cambridge, MA 02139, USA d Clinical and Research Program in Pediatric Psychopharmacology, Massachusetts General Hospital, 185 Alewifebrook Parkway, Cambridge, MA 02138 USA b
Received 29 October 2004; received in revised form 6 February 2005; accepted 16 February 2005 Available online 1 April 2005
Abstract An examination of emotional processing in individuals with schizophrenia may aid in understanding the heterogeneous disease states of schizophrenia. An emotional Stroop test, comprising conditions separated by both emotional valence (positive, negative) and arousal (low, high), was administered to disorganized (N = 12) and non-disorganized (N = 15) schizophrenic and schizoaffective participants, and 22 non-patient controls. Results indicate that the performance of disorganized versus nondisorganized participants differed particularly on the negative, high arousal condition. Understanding which symptom dimensions accompany aberrant emotion processing might be useful in understanding the mechanisms involved in the exacerbation of symptoms, particularly disorganization. Such information may lead to improved treatment and prevention strategies for schizophrenia spectrum diagnoses. D 2005 Elsevier B.V. All rights reserved. Keywords: Disorganization; Schizophrenia; Stroop; Emotion
1. Introduction Rather than comprising a single disease entity, schizophrenia may be a cluster of heterogeneous syndromes that differ in underlying structure (e.g.,
T Corresponding author. Tel.: +1 617 496 8563; fax: +1 617 495 3728. E-mail address: [email protected] (L.K. Phillips). 0920-9964/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2005.02.009
Andreasen et al., 1995). Efforts have been made toward the clarification of subtypes, encompassing more homogenous manifestations of the illness and perhaps common pathophysiology. At the descriptive level, the positive and negative symptom dimensions of schizophrenia can be broken down further into three syndromes. Negative symptoms, such as affective flattening, withdrawal, and avolition are included in a bpsychomotor povertyQ dimension. Positive symptoms can be divided into breality distortion,Q comprising
142
L.K. Phillips et al. / Schizophrenia Research 77 (2005) 141–149
hallucinations and delusions, and bdisorganization,Q describing attentional, conceptual, and behavioral disorganization (see Liddle, 1987; Andreasen et al., 1995, for review). A number of theorists have attempted to refine the construct of disorganization. Based on the results of semantic priming studies, disorganization has been explained as a deficit in the ability to integrate contextual information (Hardy-Bayle et al., 2003) and in the capacity to maintain a mental set (Rosenthal et al., 1960). In addition, the phenomena associated with disorganization have been explained by deficits in working memory (Daban et al., 2003), sensory motor ability, IQ (Basso et al., 1998), selective attention (Ngan and Liddle, 2000), and theory of mind (Hardy-Bayle et al., 2003). Lastly, individuals diagnosed with higher levels of disorganization have displayed higher levels of arousal (Lapidus and Schmolling, 1975) and emotional reactivity (Burbridge and Barch, 2002). Evidence for the last relationship, between the disorganization syndrome and abnormalities of arousal and emotional reactivity, is becoming increasingly evident. Docherty et al. (1994) categorize individuals as affectively reactive if they produce more speech errors, in the form of referential communication failure, when discussing affectively negative topics than when discussing affectively positive topics. This disruption in language production is speculated to reflect a differential responsivity to negative valence (Docherty et al., 1994). Specifically, language reactivity, during the discussion of stressful events, is greater in participants who scored higher on positive symptoms (includes both the positive and disorganization dimensions) (Docherty and Herbert, 1997), while unrelated to deficit symptoms (Cohen and Docherty, 2003). Similarly, Burbridge and Barch (2002) found that symptoms under the disorganization dimension predicted affective reactivity in language. Finally, in a study where individuals with schizophrenia rated slides on pleasantness and arousal, higher levels of disorganization was associated with lower ratings on the pleasant slides and with a more negative mood post study participation (Quirk et al., 1998). Thus, both aberrant emotional processing and emotional reactivity may be characteristic of individuals meeting criteria for the disorganized dimension of schizophrenia.
1.1. The emotional Stroop In the current study, we examined whether aberrant emotional processing is evident in individuals with schizophrenia or schizoaffective disorder who score higher on levels of disorganization. Specifically, we compared the emotional Stroop performance of those within the schizophrenia spectrum group who scored higher in disorganization with those who scored lower. The emotional Stroop is a variant of the standard color Stroop (see Williams et al., 1996, for review). In this task, words with positive or negative emotional valence are printed in colored ink, and the participant is asked to name the color of the word while attempting to ignore the meaning of the word. Interference is measured by the delay in response time and errors in naming the color of the word. Increased response time in color naming reflects increased attention allocated toward the meaning of the word, as well as difficulty disengaging attention from it. Few studies have examined emotional Stroop effects in schizophrenia, and none has specifically addressed the Stroop in the disorganized subtype. This is a surprising oversight given that a possible dysfunctional affective system, as suggested by the associated differential responsivity to negative valence (as cited above), may be more likely to be associated with aberrant processing on this task. Epstein et al. (1999) employed an emotional Stroop test to measure responsivity to threatening verses neutral information. Although controls showed increased mesolimbic activity to threat but not neutral stimuli, individuals with schizophrenia showed increased mesolimbic activity in both threatening and neutral contexts, reflecting heightened responsivity to neutral information. Bentall and Kaney (1989) found that individuals with persecutory delusions showed an increased response time to words with hostile content versus neutral words. Thus, it remains unclear whether a Stroop effect in schizophrenia is an effect of emotionality, negativity, arousal, or specific to hostile information. 1.2. Purpose of the current study The purpose of the current study was to determine whether the disorganization symptom dimension is associated with abnormalities in emotional Stroop
L.K. Phillips et al. / Schizophrenia Research 77 (2005) 141–149
performance. Further, we aimed to disentangle responsivity to valence from that of arousal to determine which specific components of emotion are most related to disorganization. It is unknown whether individuals with schizophrenia process arousal differently, regardless of valence; or if instead, they process valence differently, regardless of the level of the arousal of the stimulus. Although past work (cited above) has shown that individuals with schizophrenia as a group show increased interference with threat related words, relative to neutral words, no work to date involving the Stroop and schizophrenia has examined the effect of more generally positively and negatively valenced emotional stimuli, separated by level of arousal. Further, no study has addressed group differences, defined by level of disorganization on an emotional Stroop task.
diagnoses of schizophrenia or schizoaffective disorder. A phone screen served as a preliminary test for study eligibility. A doctoral-level clinical psychologist (PJD) or a trained clinical psychology graduate student used the Structured Clinical Interview for the DSM-IV Axis I Disorders Patient Edition (SCID-I/ P: First et al., 1995) to interview all participants. All subjects were right handed as determined by the Annett Handedness Inventory (Annett, 1970). The demographic variables collected include: gender, age, and education (Table 1). Reading ability, relative to other cognitive domains, is presumed to be less affected by the pathophysiology associated with schizophrenia; thus, the WTAR was also administered Table 1 Demographic and clinical characteristics Characteristic
1.3. Hypotheses Quirk et al. (1998) showed that disorganized schizophrenics interpreted negative stimuli as more salient than positive stimuli compared to non-disorganized schizophrenics. Based on these findings, we hypothesized that individuals with a schizophrenic disorder who were more disorganized would display increased attention allocated to the content of the negative words relative to words that were positive in content. Specifically, we predicted that individuals diagnosed with a schizophrenic disorder who show higher levels of disorganization, relative to those with lower levels of disorganization, would show longer response time and decreased accuracy in the negatively valenced, high arousal condition, relative to the neutral condition.
2. Methods 2.1. Participants The group consisting of individuals diagnosed with schizophrenia (SZ) or schizoaffective disorder (SA) included 32 (18 SZ [13 males and 5 females] and 14 SA [6 males and 8 females]) individuals who met DSM-IV criteria. These participants were recruited through newspaper advertisements and flyers, announcing a study looking for individuals with
143
Control group Patient group Mean (SD)
Gender 11 F; 11 M Age 42.16 (9.06) Education 16.86 (1.83) (in years) Wechsler test 41.9 (8.96) of adult reading Age of onset (in years) Length of illness (in years) Taking atypical antipsychotics (%) Taking typical antipsychotics (%) Taking antiparkinsonians (%) Taking antidepressants (%) Taking moodstabilizers (%) Taking benzodiazepines (%) PANSS Reality disortion Poverty symptoms Disoganization Total
Mean (SD)
t
p
13 F; 19 M x 2 = 1.85 .933 37.79 (12.12) 1.36 .342 13.14 (2.30) 6.31 b.01 31.4 (8.82) 20.50 (3.08) 22.94 (9.45) 72
26
15
41
18
15
7.37 (1.76) 5.74 (2.36) 10.48 (2.90) 68.96 (11.37)
4.27
b.01
144
L.K. Phillips et al. / Schizophrenia Research 77 (2005) 141–149
because of its utility as a measure of potential IQ (Ginsberg, 2003). Per participant report, all but one of the individuals diagnosed with schizophrenia or schizoaffective disorder were taking psychotropic medication (twentyeight individuals on atypical antipsychotics, ten on typical neuroleptics, six on benzodiazepines, sixteen on antidepressants, six on antiParkinsonian agents, and seven were taking a mood stabilizer, Table 1). The schizophrenia/schizoaffective group experienced a mean age of onset of 20.50 years (SD = 3.08) and an average duration of illness of 22.9 years (SD = 9.45). All interviews were audiotaped and administered with 2 interviewers present who came to a consensus on all clinical ratings. A median split was used to divide those individuals diagnosed with schizophrenia and schizoaffective disorder, who scored higher on the disorganization cluster of the Positive and Negative Symptom Scale (PANSS; Kay et al., 1987), from those who scored relatively lower. These two groups (disorganized, n = 12; non-disorganized, n = 15) differed significantly on all four items of disorganization ( F = 10.7, p b .001, eta = .81), conceptual disorganization ( p b .01) mannerisms and posturing ( p b .01), difficulty in abstract thinking ( p b .01), and poor attention ( p b .01). The two groups did not differ on any of the demographic variables [gender ( p = .93), age ( p = .34), education ( p = .80), WTAR ( p = .61), diagnosis ( p = .62), duration of illness ( p = .16), age of onset ( p = .94), medication status ( p = .39), or type of antipsychotic medication (typical versus atypical) ( p = .21)]. The control group consisted of twenty-two participants (11 males and 11 females), who were recruited through newspaper advertisements and flyers in the Boston area. They were interviewed, with the use of the SCID, by the same clinician and graduate students. There were no significant differences between the groups on gender and age, and all subjects were right handed as determined by the Annett Handedness Inventory (Annett, 1970). However, individuals in the schizophrenia/schizoaffective group, on average, experienced fewer years of education and had lower WTAR scores compared to those in the control group. Exclusionary criteria, for both groups, included: learning disabilities, head injuries resulting in loss of consciousness for more than 10 min, anorexia
nervosa, seizure disorders, and other primary neurological disorders. Additional criteria for the control group included no current or past history of Axis I psychiatric disorder. The group with schizophrenia disorders had no other history of Axis I mood disorder. All study procedures were approved by the Harvard Institutional Review Board. The details of the study were explained to all participants, and after all questions were addressed, written informed consent was obtained. Participants were compensated US $10 for each hour of their participation. 2.2. Procedures 2.2.1. Questionnaires After establishment of each participant’s eligibility, we administered the Positive and Negative Symptom Scale (PANSS; Kay et al., 1987), the Annett Handedness Inventory (Annett, 1970), and the Wechsler Test of Adult Reading (WTAR: The Psychological Corporation, 2001). The PANSS variables were divided into three symptom clusters: reality distortion, poverty symptoms, and disorganization (similar to Liddle, 1987; Burbridge and Barch, 2002). Reality distortion included delusions and hallucinations. Poverty symptoms included blunted affect, motor retardation, and lack of spontaneity. Disorganization encompassed conceptual disorganization, mannerisms and posturing, difficulty in abstract thinking, and poor attention. 2.2.2. Emotional Stroop color-naming paradigm, modified Participants were presented with five blocks of 15 words from the Affective Norms for English Words (ANEW) list (Bradley and Lang, 1999). Words were displayed in four different colors: blue, red, green, and brown. The specific colors were selected because each contains only one syllable. Participants were instructed to name the color of the ink of the words as quickly as possible. The conditions differed by valence and arousal (see Appendix for word list), consisting of words that were: neutral (e.g., constant); negative, low arousal (e.g., weak); positive, low arousal (e.g., sincere); negative, high arousal (e.g., furious); and positive, high arousal (e.g., ecstatic). The conditions appropriately differed according to ratings
L.K. Phillips et al. / Schizophrenia Research 77 (2005) 141–149
of valence and arousal and were matched on frequency and length. Valence and arousal ratings, taken from the ANEW word list, were on a scale of 1– 5, with 1 being the most positive and of the highest arousal (neutral: valence = 2.94, arousal = 3.41; negative low: valence =4.06, arousal = 3.30; positive low: valence = 1.84, arousal = 3.13; negative high: valence = 4.22, arousal = 2.01; and positive high: valence = 1.73, arousal = 1.91). Each block consisted of 2 practice stimuli followed by 60 scored stimuli presented on an Apple Macintosh. A word was never repeated more than once. A custom-made program, through the use of SuperLab, controlled the interstimulus interval. Each word remained on the screen until the participant responded, with a 1000 ms interval between the participant’s response and the next stimulus presentation. Automatic collection of response times was performed through a computer microphone with a voice-activated trigger. We used Adobe Photoshop to create stimuli in the font Times, 150 pixels. Participants pressed the space bar to initiate the presentation of stimuli. The vocal responses were manually recorded by the experimenter with the use of a list of correct responses, and the participants’ vocal responses were tape-recorded to ensure accuracy of coding. Blocks, word order, and color of words were pseudo-randomly counterbalanced with the following restrictions: the neutral condition occurred in an equal number of times in each of the five possible order positions and the color of the words was never repeated more than 3 times in succession. 2.3. Data analyses, scoring of the emotional Stroop First, difference scores were obtained by subtracting the response time to the neutral Stroop condition from those of each of the emotional Stroop conditions A repeated measures multivariate analysis of variance (MANOVA) was performed, with diagnosis (schizophrenia, schizoaffective, control) as the between subjects factor and valence (positive, negative) and arousal (low, high) as within subjects factors, in order to determine if there were any differences among the three diagnostic groups. This was followed by pairwise comparisons to assess differences between the two psychiatric groups to ensure that diagnosis was not a confound of any group differences between
145
those individuals who are more versus less Disorganized. Correlations among each of the four emotional Stroops and each of the symptom clusters (reality distortion, poverty symptoms, and disorganization) were examined. A repeated measure multivariate analysis of variance (MANOVA), with group (disorganized, non-disorganized) as the between subjects factor and response time to valence (positive, negative) and arousal (low, high) as within subjects factors, was performed to examine valence versus arousal affects between the groups. This was followed by planned comparisons, in order to examine differences between the groups (disorganized, non-disorganized) in response time to each of the emotional Stroop conditions (negative valence, low arousal; negative valence, high arousal; positive valence, low arousal; positive valence, high arousal). The above analyses were repeated for accuracy data, where an error equaled reading the actual word (rather than naming its color) or naming the color incorrectly. Lastly, a binary stepwise logistic regression analysis was performed to determine if response times on the emotional Stroop conditions could correctly classify those with the highest scores on the disorganization dimension from those with the lowest.
3. Results Although there were no significant differences among the three diagnostic groups (schizophrenia, schizoaffective, control) ( p = .09) or between the two psychiatric groups on any of the emotional Stroop conditions, significant differences emerged in response time to emotional stimuli between those individuals who scored higher on levels of disorganization and those who scored lower (Fig. 1). Response time to the negative, high arousal condition correlated with the dimension of disorganization (r = .47, p = .02); neither reality distortion nor poverty symptoms correlated with any of the emotional Stroop conditions. The repeated measures analysis revealed that group differences existed across both valence and arousal. When the conditions were collapsed across levels of arousal, those who scored higher on disorganization showed increased response times for negative and decreased response times for positive blocks, relative to those who showed lower
146
L.K. Phillips et al. / Schizophrenia Research 77 (2005) 141–149
Response Time in ms. (difference scores: emotional conditions - neutral)
80
Neg Low Neg High Pos Low Pos High
40
0
-40
-80 Non-Disorganized
Disorganized
Fig. 1. Emotional Stroop conditions by group.
levels of disorganization, valence disorganization ( F [1, 23] = 8.38, p = .01, eta = .52), though, overall, all participants took longer to respond to negative valence, valence ( F [1, 23] = 5.90, p = .02, eta = .45). The two groups responded significantly different to high and low arousal, arousal disorganization ( F [1, 23] = 4.38, p = .048, eta = .40). More specifically, pairwise comparisons revealed a significant difference between the two groups on the negative high arousal block of the Stroop ( p b .01, eta = .57). The analyses pertaining to the accuracy data, though nonsignificant between those were more disorganized and those who were less disorganized, were in the same direction as the response time data. The disorganized group showed a trend toward a greater number of errors in the negatively valenced conditions and fewer errors in the positive conditions, F [1, 23] = .58, p = .45. In addition, the disorganized group showed the greatest number of errors in the negative, high arousal condition, relative to the other emotional Stroop conditions (mean negative, high arousal = 1.00 [SD = 1.15]; mean negative, low arousal = .89 [SD = 1.27]; mean positive, high arousal = .40 [SD = .52]; mean positive, low arousal = .60 [SD = 1.08]). Further support for differences in responsivity between the two groups to emotional conditions comes from the rate of discrimination of those who scored
higher on disorganization from those who were less disorganized by the four conditions of the emotional Stroop. The binary logistic regression analysis revealed that the emotional Stroop conditions correctly predicted 90% of the group higher in disorganization and 93.3% of those lower in disorganization. Overall, correct classification was 92% (x 2 [4, N = 25] = 24.42, p b .0001). A Hosmer–Lemeshow statistic (Tabachnick and Fidell, 2001, p. 539) was employed to ensure the use of a good model (x 2 [6, N = 25] = .39, p = .99).
4. Discussion The purpose of the current study was to clarify the disorganization symptom dimension of schizophrenia, through an examination of emotional information processing. Further, we aimed to disentangle responsivity to valence from that of arousal, to determine which components of emotion—valence, arousal, or both—are differentially processed in those with higher levels of disorganization. To do this, we employed an objective, quantifiable measure of emotional processing: the emotional Stroop. This measure allowed us to examine response time and accuracy associated with words differing in positive, negative, high arousal, and low arousal emotional content.
L.K. Phillips et al. / Schizophrenia Research 77 (2005) 141–149
Individuals diagnosed with schizophrenia or schizoaffective disorder, as a whole, did not respond differently from the control group in response time or accuracy to words with valenced and arousing content. Once the individuals diagnosed with schizophrenia or schizoaffective disorder were separated according to level of disorganization symptomatology, group differences within the schizophrenia/ schizoaffective group emerged. Specifically, the disorganized group displayed increased response time to negative and decreased response time to positive content. In addition, the disorganized group responded with increased response times particularly to the negative high arousal condition, relative to the neutral condition. Similarly, though not significant, individuals who were disorganized showed the highest number of errors in the negative conditions and the fewest in the positive conditions. Further, the high rate of accuracy in discrimination of disorganization by the emotional Stroops suggests that there may be an overlap in the mechanisms associated with disorganization with those that contribute to aberrant emotional processing, particularly to that of negative affective content. Similar to that mentioned in the above analyses, it was the words with negative content with higher arousal ratings that most distinguished those individuals who were more disorganized from those who were less. Emotion and cognition are highly integrated processes. Together, they contribute to the monitoring and control of thought and behavior (Gray, 2004). Differences in the processing of negative, more highly arousing stimuli, by individuals who are more disorganized, may be associated with the disinhibition of cognitive control that is found in this population. The impact of arousal levels on cognition is evidenced by the disruptive influence increased arousal can have on certain cognitive function, such as selective attention (Burbridge and Barch, 2002). Indeed, thought disorder and disorganization in schizophrenia have been associated with diminished inhibitory function, coupled with heightened arousal and increased emotional intensity. In addition, rather than increasing attention to task relevant information, abnormal emotional processing may interfere and result in the loss of perspective that has been observed in individuals who are disorganized (Harrow et al., 1989).
147
Aberrant emotional processing, evident in this disorganized sample, may be associated with an increased sensitivity to emotional stress. Indeed, cortisol change, indexing novelty or stress, has been associated with disorganization symptom severity (Walder et al., 2000). Further, since stress sensitivity has been a predecessor of psychotic episodes, understanding which symptoms accompany abnormalities in emotional processing might be useful in understanding the mechanisms involved in the exacerbation of symptomatology. Such information has the potential to lead to improved treatment and prevention strategies. There were significant differences between the schizophrenia/schizoaffective group and the healthy control group on demographic variables, such as education, WTAR, and medication status. However, given there were no significant differences among the three groups on emotional Stroop performance, and given other confounds raised by matching on such variables (Meehl, 1970), we believe that the demographic differences do not pose significant challenge to our conclusions. Further, because there were no significant differences between those higher in Disorganization and those lower on any of the demographic variables (gender, age, education, WTAR, handedness, age of onset, length of illness, medication status, and diagnosis) differential performance on the Stroop is probably unrelated to demographic differences. One limitation of this study is the multiple interpretations associated with performance on the Stroop test. We are unable to determine whether increased response time indicates increased attention toward a stimulus or difficulty disengaging from a stimulus. While it is especially compelling that stimuli in the form of words, considered less ecologically valid than faces, pictures, or film clips, elicited differential responses among the groups, follow up work with other forms of emotional stimuli would be helpful to understanding the relationship between emotional responsivity and symptoms of disorganization. Additional work aimed at examining aberrant processing in those who are disorganized might include independent measures of different aspects of emotional responsivity and affective arousal, such as skin conductance and facial electromyography (EMG). Lastly, it would be interesting to examine emotional processing in the
148
L.K. Phillips et al. / Schizophrenia Research 77 (2005) 141–149
relatives of individuals who are disorganized in order to determine whether aberrant emotional processing and/ or levels of disorganization are heritable components of the disease process.
Acknowledgements This work was supported by Harvard University Restricted Funds. The authors would like to thank Richard J. McNally for his help with revising initial drafts of this paper.
Appendix A. Words selected from affective norms for English word list
Negative, low arousal
Positive, low arousal
Pessimistic Dreary Alienated Inadequate Weak Unwanted Glum Dishonest Hopeless Somber Sluggish Gloomy Unremarkable Desolate Plain
Reliable Dependable Kind Understanding Loved Considerate Carefree Pleasant Secure Sincere Peaceful Harmonious Valued Generous Sociable
Negative, high arousal
Positive, high arousal
Furious Devastated Tormented Annoyed Reckless Frightened Irritable Ruthless Desperate Wicked Jealous Tense Horrified Greedy Suffering
Sparkling Cheerful Animated Optimistic Passionate Delighted Spirited Witty Excited Splendid Jovial Ambitious Euphoric Creative Exuberant
Neutral Impartial Placid Passable Medium Customary Everyday Neutral Temperate Indifferent Centered Tolerable Pacifist Passive Moderate Constant
References Andreasen, N.C., Arndt, S., Alliger, R., Miller, D., Flaum, M., 1995. Symptoms of schizophrenia: methods, meanings, and mechanisms. Arch. Gen. Psychiatry 52, 341 – 351. Annett, M., 1970. A classification of hand preference by association analysis. Br. J. Psychol. 61, 303 – 321. Basso, M.R., Nasrallah, H.A., Olson, S., Bornstein, R., 1998. Neuropsychological correlates of negative, disorganized, and psychotic symptoms in schizophrenia. Schizophr. Res. 31, 99 – 111. Bentall, R.P., Kaney, S., 1989. Content specific information processing and persecutory delusions: an investigation using the emotional Stroop test. Br. J. Med. Psychol. 62, 355 – 364. Bradley, M.M., Lang, P.J., 1999. Affective norms for English words (ANEW): instruction manual and affective ratings. Technical Report C-1, The Center for Research in Psychophysiology. University of Florida. Burbridge, J.A., Barch, D.M., 2002. Emotional valence and reference disturbance in schizophrenia. J. Abnorm. Psychology 111, 186 – 191. Cohen, A.S., Docherty, N.M., 2003. Affective reactivity of speech and emotional experience in patients with schizophrenia. Schizophr. Res. 69, 7 – 14. Daban, C., Amado, I., Bayle, F., 2003. Disorganization syndrome is correlated to working memory deficits in medicated schizophrenic patients with recent onset schizophrenia. Schizophr. Res. 61, 323 – 324. Docherty, N.M., Herbert, A.S., 1997. Comparative affective reactivity of different types of communication disturbances in schizophrenia. J. Abnorm. Psychology 106, 325 – 330. Docherty, N.M., Evans, I.M., Sledge, W.H., 1994. Affective reactivity of language in schizophrenia. J. of Nerv. Ment. Dis. 182, 98 – 102. Epstein, J., Stern, E., Silbersweig, D., 1999. Mesolimbic activity associated with psychosis in schizophrenia: symptom-specific PET studies. Annals of the New York Academy of Sciences 877, 562 – 574. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 1995. Structured Clinical Interview for the DSM-IV Axis I Disor-
L.K. Phillips et al. / Schizophrenia Research 77 (2005) 141–149 ders—patient Edition (SCID I-P, Version 2.0). American Psychiatric Press, Washington, DC. Ginsberg, J.P., 2003. Wechsler test of adult reading. Appl. Neuropsychol. 10, 182 – 184. Gray, J.R., 2004. Integration of emotion and cognitive control. Curr. Dir. Psychol. Sci. 13, 46 – 48. Hardy-Bayle, M.-C., Sarfati, Y., Passerieux, C., 2003. The cognitive basis of disorganization symptomatology in schizophrenia and its clinical correlates: toward a pathogenetic approach to disorganization. Schizophr. Bull. 29, 459 – 471. Harrow, M., Lanin-Kettering, I., Miller, J.G., 1989. Impaired perspective and thought pathology in schizophrenic and psychotic disorders. Schizophr. Bull. 15, 605 – 623. Kay, S.R., Fiszbein, P.S., Opler, L.A., 1987. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr. Bull. 13, 261 – 276. Lapidus, L.B., Schmolling, P., 1975. Anxiety, arousal, and schizophrenia: a theoretical integration. Psychol. Bull. 82, 689 – 710. Liddle, P.F., 1987. Syndromes of chronic schizophrenia: a reexamination of the positive–negative dichotomy. Br. J. Psychiatry 151, 145 – 151. Meehl, P.E., 1970. Nuisance variables and the ex post facto design. In: Radner, M., Winokur, S. (Eds.), Minnesota Studies in the
149
Philosophy of Science. University of Minnesota Press, Minneapolis, MN, pp. 373 – 402. Ngan, E.T.C., Liddle, P.F., 2000. Reaction time, symptom profiles and course of illness in schizophrenia. Schizophr. Res. 46, 195 – 201. Quirk, S.W., Strauss, M.E., Sloan, D.M., 1998. Emotional response as a function of symptoms in schizophrenia. Schizophr. Res. 32, 31 – 39. Rosenthal, D., Lawlor, W.G., Zahn, T.P., 1960. The relationship of some aspects of mental set to degree of schizophrenic disorganization. J. Person. 28, 26 – 38. Tabachnick, B.G., Fidell, L.S., 2001. Logistic regression. In: Tabachnick, B.G., Fidell, L.S. (Eds.), Using Multivariate Statistics, 4th ed. A Pearson Education Company, Needham Heights, MA, p. 539. The Psychological Corporation, 2001. Wechsler Test of Adult Reading (WTAR) Manual. Author, San Antonio, TX. Walder, D.J., Walker, E.F., Lewine, R.J., 2000. Cognitive functioning, cortisol release, and symptom severity in patients with schizophrenia. Biol. Psychiatry 48, 1121 – 1132. Williams, J.M.G., Mathews, A., MacLeod, C., 1996. The emotional Stroop task and psychopathology. Psychol. Bull. 120, 3 – 24.
Emotional Stroop performance predicts disorganization in schizophrenia Laura K. Phillipsa,T, Patricia J. Deldinb, Martina M. Voglmaierc, Sarah Rabbittd a Department of Psychology, Harvard University, 33 Kirkland Street Rm. 1205, Cambridge, MA, 02138, USA Department of Psychology, University of Michigan, Ann Arbor, 525 E. University, 2252 East Hall, Ann Arbor, MI, 48109, USA c Department of Psychiatry, Harvard Medical School, Cambridge Health Alliance, 1493 Cambridge St., Cambridge, MA 02139, USA d Clinical and Research Program in Pediatric Psychopharmacology, Massachusetts General Hospital, 185 Alewifebrook Parkway, Cambridge, MA 02138 USA b
Received 29 October 2004; received in revised form 6 February 2005; accepted 16 February 2005 Available online 1 April 2005
Abstract An examination of emotional processing in individuals with schizophrenia may aid in understanding the heterogeneous disease states of schizophrenia. An emotional Stroop test, comprising conditions separated by both emotional valence (positive, negative) and arousal (low, high), was administered to disorganized (N = 12) and non-disorganized (N = 15) schizophrenic and schizoaffective participants, and 22 non-patient controls. Results indicate that the performance of disorganized versus nondisorganized participants differed particularly on the negative, high arousal condition. Understanding which symptom dimensions accompany aberrant emotion processing might be useful in understanding the mechanisms involved in the exacerbation of symptoms, particularly disorganization. Such information may lead to improved treatment and prevention strategies for schizophrenia spectrum diagnoses. D 2005 Elsevier B.V. All rights reserved. Keywords: Disorganization; Schizophrenia; Stroop; Emotion
1. Introduction Rather than comprising a single disease entity, schizophrenia may be a cluster of heterogeneous syndromes that differ in underlying structure (e.g.,
T Corresponding author. Tel.: +1 617 496 8563; fax: +1 617 495 3728. E-mail address: [email protected] (L.K. Phillips). 0920-9964/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2005.02.009
Andreasen et al., 1995). Efforts have been made toward the clarification of subtypes, encompassing more homogenous manifestations of the illness and perhaps common pathophysiology. At the descriptive level, the positive and negative symptom dimensions of schizophrenia can be broken down further into three syndromes. Negative symptoms, such as affective flattening, withdrawal, and avolition are included in a bpsychomotor povertyQ dimension. Positive symptoms can be divided into breality distortion,Q comprising
142
L.K. Phillips et al. / Schizophrenia Research 77 (2005) 141–149
hallucinations and delusions, and bdisorganization,Q describing attentional, conceptual, and behavioral disorganization (see Liddle, 1987; Andreasen et al., 1995, for review). A number of theorists have attempted to refine the construct of disorganization. Based on the results of semantic priming studies, disorganization has been explained as a deficit in the ability to integrate contextual information (Hardy-Bayle et al., 2003) and in the capacity to maintain a mental set (Rosenthal et al., 1960). In addition, the phenomena associated with disorganization have been explained by deficits in working memory (Daban et al., 2003), sensory motor ability, IQ (Basso et al., 1998), selective attention (Ngan and Liddle, 2000), and theory of mind (Hardy-Bayle et al., 2003). Lastly, individuals diagnosed with higher levels of disorganization have displayed higher levels of arousal (Lapidus and Schmolling, 1975) and emotional reactivity (Burbridge and Barch, 2002). Evidence for the last relationship, between the disorganization syndrome and abnormalities of arousal and emotional reactivity, is becoming increasingly evident. Docherty et al. (1994) categorize individuals as affectively reactive if they produce more speech errors, in the form of referential communication failure, when discussing affectively negative topics than when discussing affectively positive topics. This disruption in language production is speculated to reflect a differential responsivity to negative valence (Docherty et al., 1994). Specifically, language reactivity, during the discussion of stressful events, is greater in participants who scored higher on positive symptoms (includes both the positive and disorganization dimensions) (Docherty and Herbert, 1997), while unrelated to deficit symptoms (Cohen and Docherty, 2003). Similarly, Burbridge and Barch (2002) found that symptoms under the disorganization dimension predicted affective reactivity in language. Finally, in a study where individuals with schizophrenia rated slides on pleasantness and arousal, higher levels of disorganization was associated with lower ratings on the pleasant slides and with a more negative mood post study participation (Quirk et al., 1998). Thus, both aberrant emotional processing and emotional reactivity may be characteristic of individuals meeting criteria for the disorganized dimension of schizophrenia.
1.1. The emotional Stroop In the current study, we examined whether aberrant emotional processing is evident in individuals with schizophrenia or schizoaffective disorder who score higher on levels of disorganization. Specifically, we compared the emotional Stroop performance of those within the schizophrenia spectrum group who scored higher in disorganization with those who scored lower. The emotional Stroop is a variant of the standard color Stroop (see Williams et al., 1996, for review). In this task, words with positive or negative emotional valence are printed in colored ink, and the participant is asked to name the color of the word while attempting to ignore the meaning of the word. Interference is measured by the delay in response time and errors in naming the color of the word. Increased response time in color naming reflects increased attention allocated toward the meaning of the word, as well as difficulty disengaging attention from it. Few studies have examined emotional Stroop effects in schizophrenia, and none has specifically addressed the Stroop in the disorganized subtype. This is a surprising oversight given that a possible dysfunctional affective system, as suggested by the associated differential responsivity to negative valence (as cited above), may be more likely to be associated with aberrant processing on this task. Epstein et al. (1999) employed an emotional Stroop test to measure responsivity to threatening verses neutral information. Although controls showed increased mesolimbic activity to threat but not neutral stimuli, individuals with schizophrenia showed increased mesolimbic activity in both threatening and neutral contexts, reflecting heightened responsivity to neutral information. Bentall and Kaney (1989) found that individuals with persecutory delusions showed an increased response time to words with hostile content versus neutral words. Thus, it remains unclear whether a Stroop effect in schizophrenia is an effect of emotionality, negativity, arousal, or specific to hostile information. 1.2. Purpose of the current study The purpose of the current study was to determine whether the disorganization symptom dimension is associated with abnormalities in emotional Stroop
L.K. Phillips et al. / Schizophrenia Research 77 (2005) 141–149
performance. Further, we aimed to disentangle responsivity to valence from that of arousal to determine which specific components of emotion are most related to disorganization. It is unknown whether individuals with schizophrenia process arousal differently, regardless of valence; or if instead, they process valence differently, regardless of the level of the arousal of the stimulus. Although past work (cited above) has shown that individuals with schizophrenia as a group show increased interference with threat related words, relative to neutral words, no work to date involving the Stroop and schizophrenia has examined the effect of more generally positively and negatively valenced emotional stimuli, separated by level of arousal. Further, no study has addressed group differences, defined by level of disorganization on an emotional Stroop task.
diagnoses of schizophrenia or schizoaffective disorder. A phone screen served as a preliminary test for study eligibility. A doctoral-level clinical psychologist (PJD) or a trained clinical psychology graduate student used the Structured Clinical Interview for the DSM-IV Axis I Disorders Patient Edition (SCID-I/ P: First et al., 1995) to interview all participants. All subjects were right handed as determined by the Annett Handedness Inventory (Annett, 1970). The demographic variables collected include: gender, age, and education (Table 1). Reading ability, relative to other cognitive domains, is presumed to be less affected by the pathophysiology associated with schizophrenia; thus, the WTAR was also administered Table 1 Demographic and clinical characteristics Characteristic
1.3. Hypotheses Quirk et al. (1998) showed that disorganized schizophrenics interpreted negative stimuli as more salient than positive stimuli compared to non-disorganized schizophrenics. Based on these findings, we hypothesized that individuals with a schizophrenic disorder who were more disorganized would display increased attention allocated to the content of the negative words relative to words that were positive in content. Specifically, we predicted that individuals diagnosed with a schizophrenic disorder who show higher levels of disorganization, relative to those with lower levels of disorganization, would show longer response time and decreased accuracy in the negatively valenced, high arousal condition, relative to the neutral condition.
2. Methods 2.1. Participants The group consisting of individuals diagnosed with schizophrenia (SZ) or schizoaffective disorder (SA) included 32 (18 SZ [13 males and 5 females] and 14 SA [6 males and 8 females]) individuals who met DSM-IV criteria. These participants were recruited through newspaper advertisements and flyers, announcing a study looking for individuals with
143
Control group Patient group Mean (SD)
Gender 11 F; 11 M Age 42.16 (9.06) Education 16.86 (1.83) (in years) Wechsler test 41.9 (8.96) of adult reading Age of onset (in years) Length of illness (in years) Taking atypical antipsychotics (%) Taking typical antipsychotics (%) Taking antiparkinsonians (%) Taking antidepressants (%) Taking moodstabilizers (%) Taking benzodiazepines (%) PANSS Reality disortion Poverty symptoms Disoganization Total
Mean (SD)
t
p
13 F; 19 M x 2 = 1.85 .933 37.79 (12.12) 1.36 .342 13.14 (2.30) 6.31 b.01 31.4 (8.82) 20.50 (3.08) 22.94 (9.45) 72
26
15
41
18
15
7.37 (1.76) 5.74 (2.36) 10.48 (2.90) 68.96 (11.37)
4.27
b.01
144
L.K. Phillips et al. / Schizophrenia Research 77 (2005) 141–149
because of its utility as a measure of potential IQ (Ginsberg, 2003). Per participant report, all but one of the individuals diagnosed with schizophrenia or schizoaffective disorder were taking psychotropic medication (twentyeight individuals on atypical antipsychotics, ten on typical neuroleptics, six on benzodiazepines, sixteen on antidepressants, six on antiParkinsonian agents, and seven were taking a mood stabilizer, Table 1). The schizophrenia/schizoaffective group experienced a mean age of onset of 20.50 years (SD = 3.08) and an average duration of illness of 22.9 years (SD = 9.45). All interviews were audiotaped and administered with 2 interviewers present who came to a consensus on all clinical ratings. A median split was used to divide those individuals diagnosed with schizophrenia and schizoaffective disorder, who scored higher on the disorganization cluster of the Positive and Negative Symptom Scale (PANSS; Kay et al., 1987), from those who scored relatively lower. These two groups (disorganized, n = 12; non-disorganized, n = 15) differed significantly on all four items of disorganization ( F = 10.7, p b .001, eta = .81), conceptual disorganization ( p b .01) mannerisms and posturing ( p b .01), difficulty in abstract thinking ( p b .01), and poor attention ( p b .01). The two groups did not differ on any of the demographic variables [gender ( p = .93), age ( p = .34), education ( p = .80), WTAR ( p = .61), diagnosis ( p = .62), duration of illness ( p = .16), age of onset ( p = .94), medication status ( p = .39), or type of antipsychotic medication (typical versus atypical) ( p = .21)]. The control group consisted of twenty-two participants (11 males and 11 females), who were recruited through newspaper advertisements and flyers in the Boston area. They were interviewed, with the use of the SCID, by the same clinician and graduate students. There were no significant differences between the groups on gender and age, and all subjects were right handed as determined by the Annett Handedness Inventory (Annett, 1970). However, individuals in the schizophrenia/schizoaffective group, on average, experienced fewer years of education and had lower WTAR scores compared to those in the control group. Exclusionary criteria, for both groups, included: learning disabilities, head injuries resulting in loss of consciousness for more than 10 min, anorexia
nervosa, seizure disorders, and other primary neurological disorders. Additional criteria for the control group included no current or past history of Axis I psychiatric disorder. The group with schizophrenia disorders had no other history of Axis I mood disorder. All study procedures were approved by the Harvard Institutional Review Board. The details of the study were explained to all participants, and after all questions were addressed, written informed consent was obtained. Participants were compensated US $10 for each hour of their participation. 2.2. Procedures 2.2.1. Questionnaires After establishment of each participant’s eligibility, we administered the Positive and Negative Symptom Scale (PANSS; Kay et al., 1987), the Annett Handedness Inventory (Annett, 1970), and the Wechsler Test of Adult Reading (WTAR: The Psychological Corporation, 2001). The PANSS variables were divided into three symptom clusters: reality distortion, poverty symptoms, and disorganization (similar to Liddle, 1987; Burbridge and Barch, 2002). Reality distortion included delusions and hallucinations. Poverty symptoms included blunted affect, motor retardation, and lack of spontaneity. Disorganization encompassed conceptual disorganization, mannerisms and posturing, difficulty in abstract thinking, and poor attention. 2.2.2. Emotional Stroop color-naming paradigm, modified Participants were presented with five blocks of 15 words from the Affective Norms for English Words (ANEW) list (Bradley and Lang, 1999). Words were displayed in four different colors: blue, red, green, and brown. The specific colors were selected because each contains only one syllable. Participants were instructed to name the color of the ink of the words as quickly as possible. The conditions differed by valence and arousal (see Appendix for word list), consisting of words that were: neutral (e.g., constant); negative, low arousal (e.g., weak); positive, low arousal (e.g., sincere); negative, high arousal (e.g., furious); and positive, high arousal (e.g., ecstatic). The conditions appropriately differed according to ratings
L.K. Phillips et al. / Schizophrenia Research 77 (2005) 141–149
of valence and arousal and were matched on frequency and length. Valence and arousal ratings, taken from the ANEW word list, were on a scale of 1– 5, with 1 being the most positive and of the highest arousal (neutral: valence = 2.94, arousal = 3.41; negative low: valence =4.06, arousal = 3.30; positive low: valence = 1.84, arousal = 3.13; negative high: valence = 4.22, arousal = 2.01; and positive high: valence = 1.73, arousal = 1.91). Each block consisted of 2 practice stimuli followed by 60 scored stimuli presented on an Apple Macintosh. A word was never repeated more than once. A custom-made program, through the use of SuperLab, controlled the interstimulus interval. Each word remained on the screen until the participant responded, with a 1000 ms interval between the participant’s response and the next stimulus presentation. Automatic collection of response times was performed through a computer microphone with a voice-activated trigger. We used Adobe Photoshop to create stimuli in the font Times, 150 pixels. Participants pressed the space bar to initiate the presentation of stimuli. The vocal responses were manually recorded by the experimenter with the use of a list of correct responses, and the participants’ vocal responses were tape-recorded to ensure accuracy of coding. Blocks, word order, and color of words were pseudo-randomly counterbalanced with the following restrictions: the neutral condition occurred in an equal number of times in each of the five possible order positions and the color of the words was never repeated more than 3 times in succession. 2.3. Data analyses, scoring of the emotional Stroop First, difference scores were obtained by subtracting the response time to the neutral Stroop condition from those of each of the emotional Stroop conditions A repeated measures multivariate analysis of variance (MANOVA) was performed, with diagnosis (schizophrenia, schizoaffective, control) as the between subjects factor and valence (positive, negative) and arousal (low, high) as within subjects factors, in order to determine if there were any differences among the three diagnostic groups. This was followed by pairwise comparisons to assess differences between the two psychiatric groups to ensure that diagnosis was not a confound of any group differences between
145
those individuals who are more versus less Disorganized. Correlations among each of the four emotional Stroops and each of the symptom clusters (reality distortion, poverty symptoms, and disorganization) were examined. A repeated measure multivariate analysis of variance (MANOVA), with group (disorganized, non-disorganized) as the between subjects factor and response time to valence (positive, negative) and arousal (low, high) as within subjects factors, was performed to examine valence versus arousal affects between the groups. This was followed by planned comparisons, in order to examine differences between the groups (disorganized, non-disorganized) in response time to each of the emotional Stroop conditions (negative valence, low arousal; negative valence, high arousal; positive valence, low arousal; positive valence, high arousal). The above analyses were repeated for accuracy data, where an error equaled reading the actual word (rather than naming its color) or naming the color incorrectly. Lastly, a binary stepwise logistic regression analysis was performed to determine if response times on the emotional Stroop conditions could correctly classify those with the highest scores on the disorganization dimension from those with the lowest.
3. Results Although there were no significant differences among the three diagnostic groups (schizophrenia, schizoaffective, control) ( p = .09) or between the two psychiatric groups on any of the emotional Stroop conditions, significant differences emerged in response time to emotional stimuli between those individuals who scored higher on levels of disorganization and those who scored lower (Fig. 1). Response time to the negative, high arousal condition correlated with the dimension of disorganization (r = .47, p = .02); neither reality distortion nor poverty symptoms correlated with any of the emotional Stroop conditions. The repeated measures analysis revealed that group differences existed across both valence and arousal. When the conditions were collapsed across levels of arousal, those who scored higher on disorganization showed increased response times for negative and decreased response times for positive blocks, relative to those who showed lower
146
L.K. Phillips et al. / Schizophrenia Research 77 (2005) 141–149
Response Time in ms. (difference scores: emotional conditions - neutral)
80
Neg Low Neg High Pos Low Pos High
40
0
-40
-80 Non-Disorganized
Disorganized
Fig. 1. Emotional Stroop conditions by group.
levels of disorganization, valence disorganization ( F [1, 23] = 8.38, p = .01, eta = .52), though, overall, all participants took longer to respond to negative valence, valence ( F [1, 23] = 5.90, p = .02, eta = .45). The two groups responded significantly different to high and low arousal, arousal disorganization ( F [1, 23] = 4.38, p = .048, eta = .40). More specifically, pairwise comparisons revealed a significant difference between the two groups on the negative high arousal block of the Stroop ( p b .01, eta = .57). The analyses pertaining to the accuracy data, though nonsignificant between those were more disorganized and those who were less disorganized, were in the same direction as the response time data. The disorganized group showed a trend toward a greater number of errors in the negatively valenced conditions and fewer errors in the positive conditions, F [1, 23] = .58, p = .45. In addition, the disorganized group showed the greatest number of errors in the negative, high arousal condition, relative to the other emotional Stroop conditions (mean negative, high arousal = 1.00 [SD = 1.15]; mean negative, low arousal = .89 [SD = 1.27]; mean positive, high arousal = .40 [SD = .52]; mean positive, low arousal = .60 [SD = 1.08]). Further support for differences in responsivity between the two groups to emotional conditions comes from the rate of discrimination of those who scored
higher on disorganization from those who were less disorganized by the four conditions of the emotional Stroop. The binary logistic regression analysis revealed that the emotional Stroop conditions correctly predicted 90% of the group higher in disorganization and 93.3% of those lower in disorganization. Overall, correct classification was 92% (x 2 [4, N = 25] = 24.42, p b .0001). A Hosmer–Lemeshow statistic (Tabachnick and Fidell, 2001, p. 539) was employed to ensure the use of a good model (x 2 [6, N = 25] = .39, p = .99).
4. Discussion The purpose of the current study was to clarify the disorganization symptom dimension of schizophrenia, through an examination of emotional information processing. Further, we aimed to disentangle responsivity to valence from that of arousal, to determine which components of emotion—valence, arousal, or both—are differentially processed in those with higher levels of disorganization. To do this, we employed an objective, quantifiable measure of emotional processing: the emotional Stroop. This measure allowed us to examine response time and accuracy associated with words differing in positive, negative, high arousal, and low arousal emotional content.
L.K. Phillips et al. / Schizophrenia Research 77 (2005) 141–149
Individuals diagnosed with schizophrenia or schizoaffective disorder, as a whole, did not respond differently from the control group in response time or accuracy to words with valenced and arousing content. Once the individuals diagnosed with schizophrenia or schizoaffective disorder were separated according to level of disorganization symptomatology, group differences within the schizophrenia/ schizoaffective group emerged. Specifically, the disorganized group displayed increased response time to negative and decreased response time to positive content. In addition, the disorganized group responded with increased response times particularly to the negative high arousal condition, relative to the neutral condition. Similarly, though not significant, individuals who were disorganized showed the highest number of errors in the negative conditions and the fewest in the positive conditions. Further, the high rate of accuracy in discrimination of disorganization by the emotional Stroops suggests that there may be an overlap in the mechanisms associated with disorganization with those that contribute to aberrant emotional processing, particularly to that of negative affective content. Similar to that mentioned in the above analyses, it was the words with negative content with higher arousal ratings that most distinguished those individuals who were more disorganized from those who were less. Emotion and cognition are highly integrated processes. Together, they contribute to the monitoring and control of thought and behavior (Gray, 2004). Differences in the processing of negative, more highly arousing stimuli, by individuals who are more disorganized, may be associated with the disinhibition of cognitive control that is found in this population. The impact of arousal levels on cognition is evidenced by the disruptive influence increased arousal can have on certain cognitive function, such as selective attention (Burbridge and Barch, 2002). Indeed, thought disorder and disorganization in schizophrenia have been associated with diminished inhibitory function, coupled with heightened arousal and increased emotional intensity. In addition, rather than increasing attention to task relevant information, abnormal emotional processing may interfere and result in the loss of perspective that has been observed in individuals who are disorganized (Harrow et al., 1989).
147
Aberrant emotional processing, evident in this disorganized sample, may be associated with an increased sensitivity to emotional stress. Indeed, cortisol change, indexing novelty or stress, has been associated with disorganization symptom severity (Walder et al., 2000). Further, since stress sensitivity has been a predecessor of psychotic episodes, understanding which symptoms accompany abnormalities in emotional processing might be useful in understanding the mechanisms involved in the exacerbation of symptomatology. Such information has the potential to lead to improved treatment and prevention strategies. There were significant differences between the schizophrenia/schizoaffective group and the healthy control group on demographic variables, such as education, WTAR, and medication status. However, given there were no significant differences among the three groups on emotional Stroop performance, and given other confounds raised by matching on such variables (Meehl, 1970), we believe that the demographic differences do not pose significant challenge to our conclusions. Further, because there were no significant differences between those higher in Disorganization and those lower on any of the demographic variables (gender, age, education, WTAR, handedness, age of onset, length of illness, medication status, and diagnosis) differential performance on the Stroop is probably unrelated to demographic differences. One limitation of this study is the multiple interpretations associated with performance on the Stroop test. We are unable to determine whether increased response time indicates increased attention toward a stimulus or difficulty disengaging from a stimulus. While it is especially compelling that stimuli in the form of words, considered less ecologically valid than faces, pictures, or film clips, elicited differential responses among the groups, follow up work with other forms of emotional stimuli would be helpful to understanding the relationship between emotional responsivity and symptoms of disorganization. Additional work aimed at examining aberrant processing in those who are disorganized might include independent measures of different aspects of emotional responsivity and affective arousal, such as skin conductance and facial electromyography (EMG). Lastly, it would be interesting to examine emotional processing in the
148
L.K. Phillips et al. / Schizophrenia Research 77 (2005) 141–149
relatives of individuals who are disorganized in order to determine whether aberrant emotional processing and/ or levels of disorganization are heritable components of the disease process.
Acknowledgements This work was supported by Harvard University Restricted Funds. The authors would like to thank Richard J. McNally for his help with revising initial drafts of this paper.
Appendix A. Words selected from affective norms for English word list
Negative, low arousal
Positive, low arousal
Pessimistic Dreary Alienated Inadequate Weak Unwanted Glum Dishonest Hopeless Somber Sluggish Gloomy Unremarkable Desolate Plain
Reliable Dependable Kind Understanding Loved Considerate Carefree Pleasant Secure Sincere Peaceful Harmonious Valued Generous Sociable
Negative, high arousal
Positive, high arousal
Furious Devastated Tormented Annoyed Reckless Frightened Irritable Ruthless Desperate Wicked Jealous Tense Horrified Greedy Suffering
Sparkling Cheerful Animated Optimistic Passionate Delighted Spirited Witty Excited Splendid Jovial Ambitious Euphoric Creative Exuberant
Neutral Impartial Placid Passable Medium Customary Everyday Neutral Temperate Indifferent Centered Tolerable Pacifist Passive Moderate Constant
References Andreasen, N.C., Arndt, S., Alliger, R., Miller, D., Flaum, M., 1995. Symptoms of schizophrenia: methods, meanings, and mechanisms. Arch. Gen. Psychiatry 52, 341 – 351. Annett, M., 1970. A classification of hand preference by association analysis. Br. J. Psychol. 61, 303 – 321. Basso, M.R., Nasrallah, H.A., Olson, S., Bornstein, R., 1998. Neuropsychological correlates of negative, disorganized, and psychotic symptoms in schizophrenia. Schizophr. Res. 31, 99 – 111. Bentall, R.P., Kaney, S., 1989. Content specific information processing and persecutory delusions: an investigation using the emotional Stroop test. Br. J. Med. Psychol. 62, 355 – 364. Bradley, M.M., Lang, P.J., 1999. Affective norms for English words (ANEW): instruction manual and affective ratings. Technical Report C-1, The Center for Research in Psychophysiology. University of Florida. Burbridge, J.A., Barch, D.M., 2002. Emotional valence and reference disturbance in schizophrenia. J. Abnorm. Psychology 111, 186 – 191. Cohen, A.S., Docherty, N.M., 2003. Affective reactivity of speech and emotional experience in patients with schizophrenia. Schizophr. Res. 69, 7 – 14. Daban, C., Amado, I., Bayle, F., 2003. Disorganization syndrome is correlated to working memory deficits in medicated schizophrenic patients with recent onset schizophrenia. Schizophr. Res. 61, 323 – 324. Docherty, N.M., Herbert, A.S., 1997. Comparative affective reactivity of different types of communication disturbances in schizophrenia. J. Abnorm. Psychology 106, 325 – 330. Docherty, N.M., Evans, I.M., Sledge, W.H., 1994. Affective reactivity of language in schizophrenia. J. of Nerv. Ment. Dis. 182, 98 – 102. Epstein, J., Stern, E., Silbersweig, D., 1999. Mesolimbic activity associated with psychosis in schizophrenia: symptom-specific PET studies. Annals of the New York Academy of Sciences 877, 562 – 574. First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 1995. Structured Clinical Interview for the DSM-IV Axis I Disor-
L.K. Phillips et al. / Schizophrenia Research 77 (2005) 141–149 ders—patient Edition (SCID I-P, Version 2.0). American Psychiatric Press, Washington, DC. Ginsberg, J.P., 2003. Wechsler test of adult reading. Appl. Neuropsychol. 10, 182 – 184. Gray, J.R., 2004. Integration of emotion and cognitive control. Curr. Dir. Psychol. Sci. 13, 46 – 48. Hardy-Bayle, M.-C., Sarfati, Y., Passerieux, C., 2003. The cognitive basis of disorganization symptomatology in schizophrenia and its clinical correlates: toward a pathogenetic approach to disorganization. Schizophr. Bull. 29, 459 – 471. Harrow, M., Lanin-Kettering, I., Miller, J.G., 1989. Impaired perspective and thought pathology in schizophrenic and psychotic disorders. Schizophr. Bull. 15, 605 – 623. Kay, S.R., Fiszbein, P.S., Opler, L.A., 1987. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr. Bull. 13, 261 – 276. Lapidus, L.B., Schmolling, P., 1975. Anxiety, arousal, and schizophrenia: a theoretical integration. Psychol. Bull. 82, 689 – 710. Liddle, P.F., 1987. Syndromes of chronic schizophrenia: a reexamination of the positive–negative dichotomy. Br. J. Psychiatry 151, 145 – 151. Meehl, P.E., 1970. Nuisance variables and the ex post facto design. In: Radner, M., Winokur, S. (Eds.), Minnesota Studies in the
149
Philosophy of Science. University of Minnesota Press, Minneapolis, MN, pp. 373 – 402. Ngan, E.T.C., Liddle, P.F., 2000. Reaction time, symptom profiles and course of illness in schizophrenia. Schizophr. Res. 46, 195 – 201. Quirk, S.W., Strauss, M.E., Sloan, D.M., 1998. Emotional response as a function of symptoms in schizophrenia. Schizophr. Res. 32, 31 – 39. Rosenthal, D., Lawlor, W.G., Zahn, T.P., 1960. The relationship of some aspects of mental set to degree of schizophrenic disorganization. J. Person. 28, 26 – 38. Tabachnick, B.G., Fidell, L.S., 2001. Logistic regression. In: Tabachnick, B.G., Fidell, L.S. (Eds.), Using Multivariate Statistics, 4th ed. A Pearson Education Company, Needham Heights, MA, p. 539. The Psychological Corporation, 2001. Wechsler Test of Adult Reading (WTAR) Manual. Author, San Antonio, TX. Walder, D.J., Walker, E.F., Lewine, R.J., 2000. Cognitive functioning, cortisol release, and symptom severity in patients with schizophrenia. Biol. Psychiatry 48, 1121 – 1132. Williams, J.M.G., Mathews, A., MacLeod, C., 1996. The emotional Stroop task and psychopathology. Psychol. Bull. 120, 3 – 24.