Factors that affect social cue recognition in schizophrenia

Ž .Psychiatry Research 78 1998 189]196

Factors that affect social cue recognition in schizophrenia

Patrick W. CorriganU , Denise R. Nelson

Uni ersity of Chicago Center for Psychiatric Rehabilitation, 7230 Arbor Dri e, Tinley Park, IL 60477, USA

Received 20 October 1997; revised 24 December 1997; accepted 23 January 1998

Abstract

Ž Ž . .Earlier research Corrigan and Green, Am. J. Psychiatry, 150 1993 589]594 showed fairly symptomatic personswith schizophrenia give more false-positive responses when answering questions about abstract cues in a social

Ž . Žsituation i.e. affect, rules, and goals inferred about an interpersonal situation than concrete cues i.e. actions and.dialogue observed in a situation . It is unclear, however, whether differential cue recognition is due to schizophrenia

per se, or some aspect of the illness commensurate with significant symptoms and in-patient care. Moreover, theabstract and concrete dimension in the earlier study had not been independently validated. In this study, the 288

Ž .items of the Social Cue Recognition Test SCRT were divided into three sets based on abstraction ratings providedby 38 college students. The SCRT was then completed by 48 participants with DSM-III-R diagnoses of schizophreniaor schizoaffective disorder. Participants with schizophrenia were divided into low and high symptom groups usingscores from the Brief Psychiatric Rating Scale. Results showed both low symptom and high symptom groupsexhibited a differential deficit in cue recognition. False positives were greater for items rated as more abstract.Implications for understanding the social cognitive deficits of persons with schizophrenia are discussed. Q 1998Elsevier Science Ireland Ltd.

Keywords: Cognition; Social behavior; Schizoaffective disorder; Social cue recognition; Test

1. Introduction

Many persons with schizophrenia lack the basicsocial, coping, and problem solving skills neededto manage the demands of independent living.

U Corresponding author. Tel.: q1 708 6144770; fax: q1 7086144780; e-mail: [email protected]

These deficits may occur because persons withschizophrenia never learned appropriate skillsŽ .Liberman et al., 1989 . Alternatively, personswith schizophrenia may have acquired appropri-ate social skills but lack attention and memoryfunctions which lead to recognizing and under-standing the subtleties of interpersonal situations;inability to understand these cues may diminishthe person’s ability to select social skills that meet

0165-1781r98r$19.00 Q 1998 Elsevier Science Ireland Ltd. All rights reserved.Ž .P I I S 0 1 6 5 - 1 7 8 1 9 8 0 0 0 1 3 - 4

( )P.W. Corrigan, D.R. Nelson r Psychiatry Research 78 1998 189]196190

Ž .the demands of the situation Penn et al., 1997 .Research has shown that persons withschizophrenia are significantly less able than com-parison groups to recognize and process facial

Žaffect Morrison et al., 1988; Kerr and Neale,. Ž1993 , non-verbal body cues Monti and Fingeret,. Ž1987 , social knowledge Cutting and Murphy;

. ŽCramer et al., 1992 , situational features Corri-. Žgan et al., 1996 , and social cues Corrigan and

.Green, 1993; Mueser et al., 1993 . Deficits insocial cue recognition are especially interestingbecause research has suggested that social cuerecognition significantly predicts interpersonalproblem solving in persons with schizophreniaŽ .Corrigan and Toomey, 1995 .

Researchers have attempted to identify factorsthat mitigate the perception of social cues. ArgyleŽ . Ž .1986 and Argyle et al. 1981 argued that socialcues are encoded via cognitive schemata whichvary in level of abstraction. Relatively concrete

Žcues include the specific actions e.g. the waiter. Žgave the customer a menu and dialogue the

waiter said, ‘The soup of the day is chicken noo-.dle’ observed in an interpersonal vignette.

Abstract cues are inferred from actionsrdialogueŽand include: affect e.g. the waiter was angry

. Žbecause the cook messed up the order , rules thewaiter gave customers 15 min to peruse the menu

. Žbefore taking their orders , and goals the cus-tomer came to this restaurant because he wanted

.to be served quickly . We hypothesized that per-sons with schizophrenia would perceive concretecues significantly better than abstract cues be-cause they seem to have greater difficulty with themore complex analyses needed to comprehend

Žabstract information Saykin et al., 1991; Gold-.berg et al., 1993 .

This hypothesis was supported on a sample ofpersons with schizophrenia who completed theSocial Cue Recognition Test, 288 true-false itemsthat assess ability to recognize concrete and

Žabstract cues in eight videotaped vignettes Cor-.rigan and Green, 1993 . Results showed persons

with schizophrenia endorsed more false positiveswhen identifying abstract cues in a videotapedvignette. False positives occur when a person says‘true, that cue was observed in the vignette’ when,in fact, the cue was absent or could not be in-

ferred. Additional analyses failed to find signifi-cant differences in correct identification ratesŽ .saying ‘true’ to items that are, in fact, trueacross items representing abstract and concretecues. Hence, the differential deficit in recognizingabstract and concrete social cues seemed to bespecific to false positives, a result that has been

Ž .supported by other research. Penn et al. 1993found false positives on information processingmeasures were associated with interpersonalproblem solving.

Findings of a differential deficit in cue recogni-tion may be confounded by an overall perfor-mance decrement; i.e. persons with schizophreniagive more false-positive responses to abstract cuesnot because of their illness but because of someoverall functioning deficit that corresponds with

Žschizophrenia. Chapman and Chapman 1973,.1978 said this confound can be diminished by

matching items that are included in abstract andconcrete cue recognition tasks for difficulty andconsistency. This goal was accomplished on stan-dardization and cross-validation samples of nor-

Žmal controls in the original study Corrigan and.Green, 1993 .

Despite these strengths in the SCRT, severalflaws were evident in the earlier study and areaddressed in the investigation reported here. In

Ž .the earlier study, Corrigan and Green 1993 as-sumed that the affect, rules, and goals whichdescribe a social situation were more abstractthan actions and dialogue. They did not use anindependent metric of abstraction to substantiatethis claim. In the study described in this article, asample of persons without schizophrenia wereasked to rate all the SCRT items on a seven-pointabstraction scale. SCRT items were then sortedinto three groups } low, intermediate, and highabstraction } based on their mean rating. Weexpect to find that false-positive rates differ sig-nificantly across item groups; false-positive ratesincrease with the level of item abstraction.

The symptom level of participants in the earlierstudy created a second problem that is addressed

Ž .in this follow-up. Corrigan and Green 1993 ex-amined abstract and concrete cue recognition insymptomatic in-patients with schizophrenia. It wasdifficult to conclude from that study whether the

( )P.W. Corrigan, D.R. Nelson r Psychiatry Research 78 1998 189]196 191

differential deficit was due to schizophrenia perse, or some factor that coincides with exacerbatedsymptoms and in-patient care; e.g. significant psy-chosis, institutional care, or relatively high dosesof antipsychotic medication. Perhaps differentialcue recognition would not be observed in a sam-ple of relatively remitted patients whose function-ing was not diminished by these factors. Thus, asecond goal of this study was to determine whetherthe differential deficit in abstract and concretecue recognition was also present in relativelyremitted persons. Relatively remitted persons willcommit more false positives to those items thatwere rated more abstract if differential recogni-tion is not solely due to factors related to signifi-cant symptoms and in-patient care.

2. Methods

2.1. Rating item abstraction

Thirty-eight undergraduates rated the 288 itemsof the SCRT on a seven-point Abstraction ScaleŽ1sdirectly observed, 7s inferred from the situ-

.ation . The Abstraction Scale used the ArgyleŽ .1986 distinction between abstract and concretesocial information; i.e. concrete cues are directlyobservable while abstract cues need to be in-ferred from the situation. The task was adminis-tered twice with one week intervening to de-termine test-retest reliability.

As expected, the frequency distribution of itemratings on abstraction was bimodal and u-shaped;i.e. the undergraduates rated SCRT items witheither high or low extremes of the 7-point scale.This finding is consistent with our earlier asser-tion that the difference between abstract andconcrete items is significant and robust. Test-re-test reliability for the abstraction ratings was sat-isfactory; ICCs0.71. Internal consistency of rat-ings was also good; Cronbach’s alpha was 0.98. Asa result, the 288 items were sorted into thirdsŽ .ns96 based on the mean rating of abstraction.These groups represented low, intermediate, andhigh levels of abstraction.

Abstraction Scale ratings corresponded closelyŽ .to the earlier Corrigan and Green 1993 distinc-

Žtion between abstract cues i.e. affect, rules, and

. Ž .goals and concrete cues i.e. action and dialogue .All the items sorted into the high abstract condi-tion by Abstraction Ratings represented affect,rules, and goals in the 1993 study. All the items inthe low abstract condition represented action anddialogue. The intermediate condition included50% action items and 50% dialogue.

2.2. Participants

Patients for this study were obtained from twosources: Camarillo State Hospital in Californiaand the University of Chicago Psychiatric Clinics.Data from Camarillo in-patients were written up

Žin the earlier study of social cue recognition Cor-.rigan and Green, 1993 . Camarillo patients had

chart diagnoses of schizophrenia or schizoaffec-tive disorder which were validated by completion

Žof the Present State Exam PSE; Wing et al.,.1974 . Out-patients were recruited from the Uni-

versity of Chicago clinics because these patientswere expected to be significantly less symptomaticthan participants in the earlier study. The Univer-sity of Chicago group also had chart diagnoses ofschizophrenia or schizoaffective disorder vali-dated by the Structured Clinical Interview for

Ž .DSM-III-R SCID; Spitzer et al., 1990 . Ratersconducting the PSE and SCID were trained to aminimum Kappa of 0.75 according to criterionratings.

Patients with a history of drug or alcohol de-pendence or abuse during the past year, or toler-ance and withdrawal during their lifetime, wereexcluded from the study. Patients were also ex-cluded if they had a chart history of neurologicaldisorder or developmental disability. To assurethat completion of the social cue recognitionmeasure was not diminished by poor reading levelor eye sight, subjects demonstrated at least afourth-grade reading level on the Wide Range

ŽAchievement Test } Revised Jastak and Wilkin-.son, 1984 and had corrected vision of at least

20r30 on the Snellen Eye Chart.Patients were administered the expanded ver-

Žsion of the Brief Psychiatric Rating Scale BPRS;.24 items to measure psychiatric symptoms

Ž .Lukoff et al., 1986; Ventura et al., 1993 . Raterswho conducted BPRS interviews had been trained


previously to a minimum intraclass correlationŽ Ž . .ICC 1,1 : Shrout and Fleiss, 1979 of 0.80 basedon consensus ratings. An Overall Symptom Scorewas determined by summing the 24 item ratingsfrom the BPRS. The 48 patients were divided intolow and high symptom groups using the medianOverall Symptom Score as the cut point.

Mean and standard deviations of demographicvariables, overall symptoms, and medication lev-els are summarized in Table 1 for the two sam-ples. No significant differences were found acrossgroups in gender, marital status, ethnicity, andeducation. The low BPRS symptom group wasfound to be significantly younger than the highsymptom group. We also expected low and highsymptom groups would differ on several variablesthat corresponded with the large difference insymptoms. This assertion was supported by theratio of in-patients and out-patients in each group.A total of 68.2% of the Camarillo in-patientswere in the high symptom group; 62.5% of theChicago out-patients were in the low symptom

Ž .group. Non-significant trends P - 0.10 sug-gested the high symptom group had been hospi-talized more days in the past 6 months than thelow symptom group. No difference was found,however, across samples in dose of antipsychotic

Žmedication measured as chlorpromazine equiva-.lents . Oral medication dose is a poor marker of

metabolized medication, so this finding needs toŽbe interpreted cautiously Blanchard and Neale,

.1992; Spohn and Strauss, 1989 .

2.3. Procedure

Subjects were administered the Social CueŽ .Recognition Test SCRT in which they were in-

structed to watch eight videotaped vignettes ofŽtwo or three people talking Corrigan and Green,.1993; Corrigan et al., 1990 . Each vignette was

;60 s long. Vignettes included two friends build-ing a puzzle together, three people playing cards,a husband and wife fighting over the television,and a woman trying to help a distraught andsuicidal friend.

Subjects then answered 36 true-false state-ments per vignette about the presence of cuesviewed in the interaction. Statements represented

Žactions in the vignette e.g. Frank threw his cards.at the end of the scene , dialogue among vignette

Žactors e.g. Mary said, ‘Gee, that’s a good TV. Žshow’ , rules guiding the situation e.g. Harry let

Table 1Comparisons among samples

Variable Low BPRS group High BPRS group Significant difference?

N 24 24Gender 26.1% female 41.7% female ns

Ž . Ž .Age 30.9 7.7 35.8 8.0 Fs3.61 P-0.05Marital status

Single 82.3% 91.7% nsMarried 8.7% 8.3%Divorcedrwidowed 8.7%

EthnicityCaucasian 56.5% 54.2% nsAfrican American 29.1% 33.3%Latino 4.0% 8.3%Asian 4.2%

Ž . Ž .Education 12.6 2.1 12.2 2.3Ž . Ž .BPRS total score 41.7 7.2 64.8 9.1 Fs92.46, P-0.0001

2Currently in-patients 31.9% 68.2% x s4.85, P-0.05Ž . Ž .Days hospitalized in past 6 months 17.3 22.8 65.7 87.0 Fs9.34, P-0.10

Ž . Ž . Ž .CPZ equivalents mg 665 656 879 854 ns

Standard deviations are included parenthetically.


Janet enter the room first because that’s con-. Žsidered polite , goals guiding the situation e.g.

Susan studied her cards carefully because she.wanted to win the game , and affect judged from

Žactors’ behavior and facial expression e.g. Samwas angry because his wife called him a bad

.father .Statements were grouped into low, intermedi-

ate, and high abstraction sets according to theearlier ratings. The false-positive rate equalledthe number of items the subject judged to betrue, that were in fact false, divided by totalpossible false positives. The correct identificationrate equalled the number of items the subjectjudged to be true, that were in fact true, dividedby total possible correct identifications. In the

Ž .earlier study Corrigan and Green, 1993 , a signaldetection index, A9, was used to assess subjects’sensitivity to social cues. A9 was not used in thisstudy, however, because the 288 items of theSCRT were divided into thirds. The relatively

Ž .smaller number of items in each third ns96was not sufficient to meet criteria for determining

Žsignal detection variables Davies and Parasura-.man, 1982 .

3. Results

Means and standard deviations of false-positiveand correct-identification rates for the two sam-ples are summarized in Table 2. Two 2=3

Ž .ANOVAs group by abstraction level were com-pleted with false-positive and correct-identifica-tion rates as the dependent measures. Results ofthe 2=3 ANOVA for false-positive rates showed

Ž Ž .significant effects for group F 1,46 s14.36, P-

.0.0005 . Patients in the low symptom group wereless likely to make false positives to social situa-tions than the high symptom group. A significanteffect was also found for abstraction levelŽ Ž . .F 2,92 s17.33, P-0.0001 ; patients createdsignificantly more false positives on test itemsthat had been rated as more abstract. The inter-

Ž Ž . .action was not significant F 2,92 s5.13, ns ,suggesting that both groups demonstrate a dif-ferential deficit in false-positive rates across thethree conditions.

This deficit was further described in subsequentANOVAs. False-positive rates differed signifi-cantly across the three abstraction conditions in

Ž Ž .the high symptom group F 2,46 s13.74, P-.0.0001 . The magnitude of this differential deficit

is evident in post hoc tests. The false-positive ratefor the intermediate abstraction condition wassignificantly greater than that for low abstractionŽ .P-0.05 . Moreover, false positives for interme-diate abstraction were significantly fewer than forthe high abstraction condition. A significant dif-ference was also found across the three abstrac-tion conditions for the low symptom groupŽ Ž . .F 2,46 s4.43, P-0.05 . Post hoc tests, how-ever, showed only the low and high abstraction

Ž .conditions varied significantly P-0.05 .Ž .A 2=3 ANOVA group by abstraction level ,

with correct identification rate as the dependentvariable, failed to find significant effects for groupŽ Ž . .F 1,46 s0.52; P)0.45 . Patients with relativelyhigh symptoms do not differ significantly frompatients with low symptoms in correctly reportingsocial cues. The 2=3 ANOVA also failed to yield

Ž Ž .significant effects for abstraction level F 1,46 s. Ž Ž .1.33; P)0.25 or the interaction F 1,46 s0.03;

Table 2Means and standard deviations of false-positive and correct-identification rates for low, intermediate, and high abstraction items

Abstract condition False-positive rate Correct-identification rate

Low BPRS group High BPRS group Low BPRS group High BPRS group

Ž . Ž . Ž . Ž .Low 0.154 0.09 0.310 0.23 0.809 0.11 0.789 0.13Ž . Ž . Ž . Ž .Intermediate 0.196 0.12 0.380 0.21 0.811 0.13 0.784 0.13Ž . Ž . Ž . Ž .High 0.215 0.12 0.434 0.25 0.831 0.12 0.807 0.13

Standard deviations are included parenthetically.


.P)0.95 . These findings support the results fromthe earlier study; i.e. correct-identification ratesdo not seem to vary with level of cue abstraction.

4. Discussion

Results from this study replicated earlier re-search. Persons with schizophrenia show a sig-nificant differential deficit in cue recognition; theyexhibit more false positives to questions aboutabstract cues rather than concrete cues. Concretecues are the actual sights and sounds of an inter-personal interaction. Abstract cues are inferencesabout the rules, affect, and goals that describe asituation. The number of false positives increasedfor items rated as more abstract. Low, medium,and high levels of abstraction were determined byratings from an independent sample.

Findings from this study also suggest that per-sons with relatively remitted symptoms show thesame differential deficits in social cue recognitionas more symptomatic individuals. Like the highsymptom group, the remitted group made signifi-cantly more false-positive responses to SCRTitems representing relatively abstract cues thanconcrete cues. This finding suggests that the dif-ferential deficit is not due solely to non-specificdisease factors related to significant symptomsand in-patient care; e.g. severity of symptoms,high doses of antipsychotic medication, and insti-tutional care. Even persons with relatively fewmanifest symptoms have difficulty recognizing therules and goals that define a situation rather thanaction and dialogue. One should not conclude,however, that symptom severity, medication dose,and institutional care had no effect on social cueperception. Results from this study showed thatpersons with greater psychiatric symptoms mademore false-positive responses on the SCRT.

Findings from this study suggest that deficits insocial perception related to schizophrenia appearas problems in falsely identifying social informa-tion. When queried, patients make inferencesabout rules, goals, and affect that were not con-sistent with the situation. The rate of correctidentifications was not found to vary with abstrac-tion or symptoms.

How might these findings be understood in

terms of the other cognitive deficits that havebeen shown to describe schizophrenia? More falsepositives in response to questions about abstractcues may reflect a confabulatory process. Thethought disorder of individuals experiencing apsychosis results in recreating their memory of asocial situation. Abstract cues, being relativelynebulous, are likely to lead to false re-creations.Alternatively, the difference in false-positive ratesmay represent a deficit in overinclusive thinking.Persons with schizophrenia are relatively less con-servative in considering rules, goals, and affectwhen asked to recognize cues on the more dif-ficult cognitive task. Future research needs totease out these two explanations.

The pattern of deficits in social cue recognitionfound in this study is consistent with research on

Žtheory of mind and schizophrenia Frith, 1992,.1994 . According to this research, many of the

social deficits and psychotic symptoms experi-enced by persons with schizophrenia are at-tributed to an inability to accurately represent the

Ž .intentions of others. Corcoran et al. 1995 de-vised a hinting task, similar to the social cuerecognition task tested in this report, to examinethe effects of theory of mind deficits. Resultsshowed persons who were less sensitive to inter-personal hints, and therefore less aware of theintentions of others, showed greater positive andnegative symptoms.

We cannot definitively conclude that the dif-ferential deficit in concrete and abstract cuerecognition is specific to schizophrenia alone. Fu-ture research must also determine whether other

Ždiagnostic groups with marked symptoms e.g..persons with bipolar disorder demonstrate dif-

ferential cue recognition. Failure to find this pat-tern in other diagnostic groups will further sup-port the specificity of this deficit for schizophre-

Ž .nia Zubin and Spring, 1977 .These findings have implications for training

persons on skills to improve their interpersonalfunctioning. Firstly, recognition and understand-ing of social situations are expected to improvewith remission of psychotic symptoms. Hence,social perception and functioning should improvewith successful medication management pro-grams. Secondly, persons with schizophrenia find


recognizing the rules, affect, and goals that definea situation to be especially difficult, emphasizingthe need to improve abstract cue recognition.Others have argued that teaching persons to me-chanically react to concrete cues is limited in its

Ž .effectiveness Argyle, 1986; Trower, 1982 . Rather,persons need to learn the rules and goals thatguide interpersonal situations. In this way, theycan ‘generate’ behaviors that meet the demandsof a particular situation. This is a skill area whichhas largely been overlooked in current social skillstraining programs and needs to be added to theregimen.

Acknowledgements

This study was made possible, in part, by agrant from the Illinois Office of Mental Health.The authors wish to thank David Penn, StanleyMcCracken, and Neil Pliskin for helpful reviewsof earlier versions of this manuscript.

References

Argyle, M., 1986. Social skills and the analysis of situationsŽ .and conversations. In: Hollin, C.R., Trower, P. Eds. ,

Handbook of Social Skills Training, vol. 2. Pergamon Press,New York.

Argyle, M., Furnham, A., Graham, J.A., 1981. Social Situa-tions. Cambridge University Press, Cambridge.

Blanchard, J.J., Neale, J.M., 1992. Medication effects: concep-tual and methodological issues in schizophrenia research.Clinical Psychology Review 12, 345]362.

Chapman, L.J., Chapman, J.P., 1973. Problems in the mea-surement of cognitive deficit. Psychological Bulletin 79,380]385.

Chapman, L.J., Chapman, J.P., 1978. The measurement ofdifferential deficit. Journal of Psychiatric Research 14,303]311.

Corcoran, R., Mercer, G., Frith, C.D., 1995. Schizophreniasymptomatology and social inference: investigating ‘theoryof mind’ in people with schizophrenia. Schizophrenia Re-search 17, 5]13.

Corrigan, P.W., Davies-Farmer, R.M., Stolley, M.R., 1990.Social cue recognition in schizophrenia under variable lev-els of arousal. Cognitive Therapy Research 14, 353]361.

Corrigan, P.W., Green, M.F., 1993. Schizophrenic patients’sensitivity to social cues: the role of abstraction. AmericanJournal of Psychiatry 150, 589]594.

Corrigan, P.W., Silverman, R., Stephenson, J., Nugent-Hirsh-beck, J., Buican, B.J., 1996. Situational familiarity and

feature recognition in schizophrenia. Schizophrenia Bul-letin 22, 153]162.

Corrigan, P.W., Toomey, R., 1995. Interpersonal problem solv-ing and the cognitive deficits of schizophrenia. Schizophre-nia Bulletin 21, 395]404.

Cramer, P., Bowen, J., O’Neill, M., 1992. Schizophrenics andsocial judgement. Why do schizophrenics get it wrong?British Journal of Psychiatry 160, 481]487.

Cutting, J., Murphy, G.M., 1990. Impaired ability ofschizophrenics, relative to manics or depressives, to appre-ciate social knowledge about their culture. British Journalof Psychiatry 157, 355]358.

Davies, D.R., Parasuraman, R., 1982. The Psychology of Vigi-lance. Academic Press, New York.

Frith, C.D., 1992. The Cognitive Neuropsychology ofSchizophrenia. Erlbaum, Hillsdale, NJ.

Frith, C.D., 1994. Theory of mind. In: David, A.S., Cutting,Ž .J.C. Eds. , The Neuropsychology of Schizophrenia. Erl-

baum, East Sussex, England.Goldberg, T.E., Gold, J.M., Greenberg, R., Griffin, S., Schulz,

S.C., Pickar, D., Kleinman, J.E., Weinberger, D.R., 1993.Contrasts between patients with affective disorders andpatients with schizophrenia on a neurological test battery.American Journal of Psychiatry 150, 1355]1362.

Jastak, S., Wilkinson, G.S., 1984. Wide Range AchievementTest } Revised, Manual. Jastak Associates, Wilmington,DE.

Kerr, S.L., Neale, J.M., 1993. Emotion perception inschizophrenia: specific deficit or further evidence of gener-alized poor performance? Journal of Abnormal Psychology102, 312]318.

Liberman, R.P., DeRisi, W.J., Mueser, K.T., 1989. Social SkillsTraining for Psychiatric Patients. Pergamon, New York.

Lukoff, D., Liberman, R.P., Nuechterlein, K.H., 1986. Symp-tom monitoring in the rehabilitation of schizophrenicpatients. Schizophrenia Bulletin 12, 578]603.

Monti, P.M., Fingeret, A.L., 1987. Social perception and com-munication skills among schizophrenics and non-schizophrenics. Journal of Clinical Psychology 43, 197]205.

Morrison, R.L., Bellack, A.S., Mueser, K.T., 1988. Deficits infacial-affect recognition and schizophrenia. SchizophreniaBulletin 14, 67]84.

Mueser, K.T., Bellack, A.S., Wade, J.H., Sayers, S.L., Tierney,A., Hass, G., 1993. Expressed emotion, social skill andresponse to negative affect in schizophrenia. Journal ofAbnormal Psychology 102, 339]351.

Penn, D., Corrigan, P.W., Bentall, R., Racenstein, M., New-man, L., 1997. Social cognition in schizophrenia. Psycholog-ical Bulletin 121, 114]132.

Penn, D.L., van der Does, A.J.W., Spaulding, W., Garbin, C.,Linszen, D., Dingemans, P., 1993. Information processingand social-cognitive problem solving in schizophrenia: as-sessment of inter-relationships and changes over time.Journal of Nervous and Mental Disease 181, 13]20.

Saykin, A.J., Gur, R.C., Gur, R.E., Mozley, P.D., Mozley, L.H.,Resnick, S.M., Kester, D.B., Stafniak, P., 1991. Neurologi-


cal function in schizophrenia: selective impairment in me-mory and learning. Archives of General Psychiatry 48,618]624.

Shrout, P.E., Fleiss, J.L., 1979. Intraclass correlations: uses inassessing rater reliability. Psychological Bulletin 86,420]428.

Spitzer, R.L., Williams, J.B.W., Gibbon, M., First, M.B., 1990.Structured Clinical Interview for DSM-III-R. AmericanPsychiatric Association Press, Washington, DC.

Spohn, H.E., Strauss, M., 1989. Relation of neuroleptic andanticholinergic medication to cognitive functions inschizophrenia. Journal of Abnormal Psychology 98,367]380.

Trower, P., 1982. Toward a generative model of social skills: aŽ .critique and synthesis. In: Curran, J.P., Monti, P.M. Eds. ,

Social Skills Training: A Practical Handbook for Assess-ment and Treatment. Guilford Press, New York.

Ventura, J., Lukoff, D., Nuechterlein, K.H., Liberman, R.P.,Green, M.F., Shaner, A., 1993. Training and quality assur-ance with the Brief Psychiatric Rating Scale: ‘the driftbusters’. International Journal of Methods in PsychiatricResearch 3, 221]244.

Wing, J.K., Cooper, J.E., Sartorius, N., 1974. The Descriptionand Classification of Psychiatric Symptoms: An InstructionManual for the PSE and CATEGO Systems. CambridgeUniversity Press, London.

Zubin, J., Spring, B., 1977. Vulnerability: a new view ofschizophrenia. Journal of Abnormal Psychology 86,103]126.

Documents

Factors that affect social cue recognition in schizophrenia