Psychopathology and neuropsychological impairments in deficit and nondeficit schizophrenia of Chinese origin

Available online at www.sciencedirect.com

8 (2008) 195–205www.elsevier.com/locate/psychres

Psychiatry Research 15

Psychopathology and neuropsychological impairments in deficit andnondeficit schizophrenia of Chinese origin

Xiang Wang a, Shuqiao Yao a,⁎, Brian Kirkpatrick b, Chuan Shi c, Jinyao Yi a

a The Medical Psychological Research Center, Second Xiangya Hospital, Central South University, Changsha, Hunan, 410011, Chinab Department of Psychiatry and Health Behavior, Medical College of Georgia, 1515 Pope Avenue, Augusta, GA 30912, USA

c Mental Health Institute of Peking University, Beijing, 100083, China

Received 5 March 2006; received in revised form 14 June 2006; accepted 8 September 2006

Abstract

Deficit schizophrenia is a relatively homogeneous subtype of patients which is considered helpful to explore the pathogenesis ofschizophrenia. The aim of the present study was to reexamine the clinical characteristics of deficit (n=30) and nondeficitschizophrenia (n=93) in a Chinese sample and investigate the differences of neurocognitive function among the two subtypes ofschizophrenia and the normal controls (n=103). Schizophrenia patients completed the Brief Psychiatric Rating Scale (BPRS),Scale for the Assessment of Negative Symptoms (SANS) and Scale for the Assessment of Positive Symptoms (SAPS).Additionally, all participants completed an abbreviated version of the Wechsler Adult Intelligence Scale (WAIS-RC) and aneuropsychological test battery examining the executive functions, visuospatial abilities and explicit memory related to the frontal,parietal, and temporal lobe functions. The deficit group received higher scores than the nondeficit group on the BPRS anergiafactor and SANS affective flattening, alogia, avolition-apathy, anhedonia-asociality subscales, but not on the SAPS. Both twoschizophrenia subgroups performed more poorly on the WAIS-RC and neuropsychological tests than the normal controls.Moreover, deficit patients performed worse than nondeficit patients on the prorated IQ, the Trail Making Test, Wisconsin CardSorting test and Block Design test. The present study replicated symptom profiles in deficit vs. nondeficit schizophrenia in theChinese sample. Furthermore, this study suggested that deficit schizophrenia is associated with frontal and parietal lobeimpairment, and that temporal lobe dysfunction may be a common basis for cognitive impairment in schizophrenia as a whole.© 2006 Elsevier Ireland Ltd. All rights reserved.

Keywords: Deficit schizophrenia; Nondeficit schizophrenia; Psychopathology; Neurocognitive functioning

1. Introduction

Schizophrenia is characterized by marked clinicalvariance among patients in their symptoms, biologicalcorrelates, disease impact on function, and response totreatment. Present research has been hindered by extremeheterogeneity among patients (Maj, 1998; Tsuang et al.,

⁎ Corresponding author. Tel.: +86 731 5292126; fax: +86 731 5361328.E-mail address: [email protected] (S. Yao).

0165-1781/$ - see front matter © 2006 Elsevier Ireland Ltd. All rights resedoi:10.1016/j.psychres.2006.09.007

2000). As a result, researchers have systematically parsedthe symptomatology of schizophrenia into more homo-geneous diagnostic categories, such as paranoid and non-paranoid subtypes or positive and negative subtypes.Deficit schizophrenia, proposed by Carpenter andcolleagues (1988), is defined as a putative schizophreniasubtype, in which the primary negative symptomsare enduring, trait-like features throughout periods ofclinical stability. Further, the primary negative symptomsare not attributable to depression, anxiety, paranoia, or

rved.

196 X. Wang et al. / Psychiatry Research 158 (2008) 195–205

medication side effects. The validity of the deficit/nondeficit schizophrenia classification has been sup-ported by many studies indicating marked differencesregarding (1) clinical signs and symptoms, (2) course ofillness, (3) etiological vulnerability factors, (4) biologicalcorrelates, and (5) treatment response (Kirkpatrick et al.,2001). Longitudinal studies have demonstrated that thedeficit subtype is a highly stable form of schizophrenia(Amador et al., 1999; Tek et al., 2001). Moreover,comparative studies have shown that the deficit subtypediffers from the traditionally defined negative subtype inconcept, treatment response and underlying neuropsy-chological correlates (Kirkpatrick et al., 1989, 1993;Kopelowicz et al., 1997, 2000; Buchanan et al., 1998;Cohen and Docherty, 2004). Therefore, with respect tothe exploration of the etiology and pathogenesis ofschizophrenia, the implications of the deficit/nondeficitcategorization are profound.

With regard to the clinical symptomatology, paststudies from a number of countries consistently indicatethat deficit patients have more severe anhedonia, lesssevere depression and anxiety, less suicidal ideation, lesssevere delusions and suspiciousness, and less substanceabuse than nondeficit patients (Kirkpatrick and Bucha-nan, 1990; Kirkpatrick et al., 1996a, 1996b; Earnst andKring, 1999; Subotnik et al., 2000; Tiryaki et al., 2003;Galderisi et al., 2002; Arango et al., 2004). Furthermore,deficit schizophrenia has been shown to be characterizedby poor premorbid functioning coupled with poor socialand occupational functioning and quality of life after theonset (Fenton and McGlashan, 1994; Kirkpatrick et al.,1996a,b; Tek et al., 2001). Studies in the literaturesupport the validity of the deficit/nondeficit categoriza-tion and, moreover, facilitate the identification of asubgroup of patients who appear to exhibit consistentclinical characteristics across independent studies.

In addition to the differences in symptomatology,individuals with deficit schizophrenia exhibit moresevere neurological impairments and subsequentlygreater neurocognitive impairment, which likely contri-bute to their poorer social functioning (Kirkpatrick et al.,2001). Additionally, neuropsychological studies indicatethat deficit schizophrenia patients show a significantlydifferent profile from nondeficit patients when examin-ing the dysfunction of separate brain regions. Morespecifically, several studies have shown that deficitpatients performed significantly worse than nondeficitpatients on executive function measures related to frontallobe functioning (Putnam and Harvey, 2000; Bryson etal., 2001; Galderisi et al., 2002). Further, studies havefound abnormalities among deficit schizophreniapatients in the sequencing of complex motor acts, visual

attention/information processing tasks, and antisaccadicand eye tracking tasks, all of which are thought to derivefrom dysfunction in parietal and prefrontal areas (Thakeret al., 1989; Malaspina et al., 1994, 2002; Ross, 2000;Ross et al., 1996, 1997; Buchanan et al., 1997; Nkam etal., 2001). In contrast, when comparing the functioningof the temporal lobe between the two subgroups, most ofabove studies failed to find a significant difference.

To further verify the validity of the deficit/nondeficitcategorization and better understand the differences be-tween the two subtypes of schizophrenia, studies exam-ining different geographical and racial samples should beconducted. In our previous work, we developed theChinese version of the Schedule for the Deficit Syndrome(SDS, Kirkpatrick et al., 1989), the instrument used toclassify deficit and nondeficit schizophrenia, and dem-onstrated that it had good reliability and validity (Wanget al., 2005). Therefore, it is valuable to compare Easternand Western patients in their symptom profiles and brainfunctioning impairments on the basis of our reliability andvalidity research on the SDS-C. To date, there is a paucityof research examining the pathogenesis of deficit andnondeficit schizophrenia in mainland China. Thus, weattempt to use a convergence-of-evidence approach toinvestigate the symptoms and the underlying biologicalcorrelates of the deficit syndrome, including multiplemeasures such as psychopathological evaluation, neuro-psychological assessments, neuroelectrophysiologicalassessment, and functional neuroanatomical imaging.

In the present study, we first examined the clinicalcharacteristics and psychopathology of deficit andnondeficit schizophrenia in a Chinese sample. Wehypothesized that the psychopathological characteristicsof Chinese patients with the deficit syndrome would beconsistent with the pattern that has emerged from paststudies conducted in Western countries. Second, wecompared the performances on the neuropsychologicalmeasures among the deficit, nondeficit and normalgroups. We hypothesized that deficit patients wouldperform worse on the general ability measure and showa different cognitive function profile from the nondeficitsample, although both subgroups would show a poorerperformance than the normal group. In particular, wehypothesized that neuropsychological indices of exec-utive function and visuospatial abilities, which arerelated to frontal and parietal lobe functioning, would bemore impaired in deficit schizophrenia than in non-deficit schizophrenia. However, the indices sensitive totemporal lobe dysfunction, such as the verbal and visualexplicit memory measures, would be expected to showno significant differences between the two subtypes ofschizophrenia.

197X. Wang et al. / Psychiatry Research 158 (2008) 195–205

2. Method

2.1. Participants

Participants in the current study included 123individuals diagnosed with schizophrenia who wererecruited through the inpatient rehabilitation units of theChangsha Mental Rehabilitation Hospital and 103healthy controls who were recruited from the localcommunity. Although 129 schizophrenia patients wererecruited, only those who completed all the studyprocedures were included in the data analysis of thecurrent study. The final group consisted of 123 patients.All schizophrenia patients were reported to be clinicallystable and were required to satisfy the followinginclusion criteria: (1) a DSM-IV (American PsychiatricAssociation, 1994) diagnosis of schizophrenia utilizingthe Chinese version of the Structured Clinical Interviewfor DSM-IV (SCID-I, First et al., 1996a); (2) duration ofillness longer than 2 years; (3) between the ages of 16and 55; (4) no history of organic brain disorder, severemental retardation, severe head trauma, alcoholism, ordrug abuse or dependence in the last 12 months; (5) nohistory of electroconvulsive therapy (ECT); (6) nosignificant changes in clinical state or in drug treatmentduring the preceding 3 months; and (7) willingness toparticipate in the study procedures. Patients meeting allof the criteria were then further classified as eitherdeficit or nondeficit schizophrenia after completing theSchedule for the Deficit Syndrome (SDS) interview (30deficit and 93 nondeficit). All of the patients were takingconventional antipsychotic medications, such as chlor-promazine, trifluoperazine, perphenazine, and haloper-idol. The exclusion criteria for the controls were asfollows (1) a lifetime history of psychotic, mood, andsubstance use disorders, ascertained by the StructuredClinical Interview for DSM-IV Non-Patient version(SCID-NP) (First et al., 1996b); (2) a history of organicbrain disorder, mental retardation, or severe headtrauma; and (3) a family history of major psychiatricdisorders. All the subjects reported were of Hanethnicity and right-handed.

2.2. Clinical assessments

2.2.1. The Schedule for the Deficit Syndrome (SDS)The SDS (Kirkpatrick et al., 1989) was used to

categorize patients as deficit or nondeficit. The sixenduring negative symptoms assessed by the SDSinclude: restricted affect; diminished emotional range;poverty of speech; curbing of interest; diminishedsense of purpose; and diminished social drive. To

meet criteria for the deficit syndrome, two of the sixdeficit symptoms must have been at least moderatelysevere, persistent over 12 months prior to the time ofinterview, and not attributable to secondary sources(e.g., medication side effects, depression, paranoia, andanxiety).

Further, deficit symptoms must always have beenpresent during periods of clinical stability (includingchronic psychotic states). Ratings of the SDS, whichwere conducted by two clinical psychologists who hadreceived special training, is based on a set of standardprobe questions assessed during a semi-structuredinterview, a comprehensive medical chart review, andconsultation with treating clinicians who are familiarwith the patients' clinical histories.

In Wang's study (2005), the original version of theSDS was translated into Mandarin and back-translatedinto English, and discussed with one of the authorsof the schedule (Brian Kirkpatrick). Moreover, thatstudy demonstrated that the Chinese version of the SDS(SDS-C) exhibited strong validity and reliability, whichinvolved the same sample of patients as the currentstudy. The inter-rater reliabilities for the unweightedkappa coefficient was 0.88, the weighted kappacoeffients ranged from 0.61 to 0.78 for each of the sixitems, Cronbach's alpha coefficient was 0.84, the meaninter-item correlation coefficient was 0.48, the rank-order correlations ranged from 0.44 to 0.79 among thesix negative symptoms, and the correlations betweenglobal severity and the six negative symptoms were0.69–0.88.

2.2.2. Symptomatology evaluationDifferent investigators carried out psychopathological

and neuropsychological evaluations to avoid ‘halo’effects. All of them and the patients were blind to thedeficit and nondeficit categorization and had not beeninformed of the specific study aims and hypotheses. Thepsychopathological state was evaluated by the BriefPsychiatric Rating Scale (BPRS, Overall and Gorham,1962), the Scale for the Assessment of NegativeSymptoms (SANS, Andreasen, 1983), and the Scale forthe Assessment of Positive Symptoms (SAPS, Andrea-sen, 1984). The BPRS, SANS, and SAPS have been wellvalidated to measure the psychopathology of schizophre-nia and have been widely used in the schizophreniaresearch. Investigators were trained in the use of theseinstruments, and at the end of the training, adequate inter-rater agreement was established among the four raterswho assessed three patients (for the BPRS, ICC=0.75–0.93; for the SANS, ICC=0.75–0.96; for the SAPS,ICC=0.81–1.00).

198 X. Wang et al. / Psychiatry Research 158 (2008) 195–205

2.3. Neuropsychological assessment

2.3.1. General intellectual abilityGeneral intellectual ability was assessed with an

abbreviated version of the Wechsler Adult IntelligenceScale-Chinese Revision (WAIS-RC, Wechsler, 1981;Gong, 1982), the standardized measurement of adultgeneral intelligence. Silverstein's study (1982) showedthat the Vocabulary and Block Design subtests havehigh reliabilities and have a higher correlation with fullscale IQ scores than any other dyad (r=0.90).Therefore, we chose these subtests for this study as ashort form of the WAIS-R.

2.3.2. Visuospatial abilitiesThe Benton Judgment of Line Orientation Test

(Benton et al., 1975, 1994; Muriel, 1983; Tang andWang, 2001) was used to assess the ability to representand coordinate visuospatial relationships. The subjectswere shown two shorter lines of different orientations anda semicircular array of 11 lines that vary in orientation byevenly spaced angular degrees, from 0° to 180°, and wereasked to determine to which lines in the array two linescorresponded. The numbers of items correctly identifiedas the comparison variables were used in this report.

The WAIS-RC Block Design subtest is a commonlyused single neuropsychological test to measure visuo-constructive ability and nonverbal reasoning whichrequires subjects to correctly reproduce two-dimension-al colored designs using colored blocks. Therefore,we used the Block Design as another index for thevisuospatial ability impairment in schizophrenia.

2.3.3. Explicit memorySelected subtests of the Wechsler Memory Scale-

Revised (WMS-RC; Wechsler, 1987; Gong, 1989) wereused to assess verbal andvisual explicitmemory.We chosetheLogicalMemory and theVisualReproduction subtests,which assess paragraph and basic geometric design recallability. Variables represented the total number of items intwo paragraph-length passages that was correctly repeatedand the accuracy criteria met in drawing designs.

2.3.4. Executive functionsThe Wisconsin Card Sorting Test (WCST) is a com-

monly used neuropsychological test sensitive to assessconcept formation, flexibility of abstract thought andfrontal lobe impairment in schizophrenia. In this study,we used the manual method (Heaton, 1981) in whichparticipants were required to sort cards according toseveral different dimensions (color, form, number). Thesorting principle must be deduced from verbal feedback

to the participant and would be changed without warningafter a particular response mode had been established(i.e., 10 consecutive correct responses). Measures ofperformance included the number of categories com-pleted, the number of perseverative errors and thenonperseverative errors, which have been consistentlylinked to cognitive deficits in schizophrenia.

The Trail Making Test, which consists of two parts,A and B (Trail A, Trail B), is also a commonly usedneurocognitive test and a good predictor of brainimpairment. Trail A consists of encircled numbersfrom 1 to 25 randomly spread across a sheet of paper.Trail B is more complex than A because it requires thesubject to connect numbers and letters in an alternatingpattern (1-A-2-B-3-C, etc.) in as little time as possible,usually being chosen as a measure of executive functionand set alternation (Reitan, 1978; Gong, 1986, 1993). Inaddition, we used Trail A as a covariation for Trail Bsince it makes all of the same cognitive demands exceptset alternation. Thus, it is possible to assess frontalfunctions by eliminating factors such as psychomotorspeed and ability (Jarvis and Barth, 1986). The outcomevariable was the time required to complete Trail MakingForms A and B, and the flexibility index was the time ofTrail B − Trail A.

2.4. Procedure

Approval to carry out the research was obtained fromthe local research ethics committee. After individualscompleted the informed consent and demographics form,diagnostic and categorizing information was collectedusing the SCID and SDS interviews. Following theinterview, all the neuropsychological tests were adminis-tered individually in a small roomwithminimal extraneousstimulations. Standard instructions and scoring methodsfor each test were used. Total time for the neuropsycho-logical test battery was approximately 60 min. All testswere administered by the same authors, whowere unawareof the subjects' status at the time of administration andscoring. Psychiatric assessments such as theBPRS, SANS,and SAPS were completed on the same day.

2.5. Data analysis

To determine the appropriateness of parametric stat-istics for each variable, the distributions of all variableswere examined for normality and homogeneity of vari-ancewithin each group. The non-normally distributed andnonhomogeneous variables included the following neu-ropsychological indices: (1) the score of the BentonJudgment of Line Orientation Test, (2) the indices of Trail

199X. Wang et al. / Psychiatry Research 158 (2008) 195–205

Making Test, and (3) the indices of the WCST. Toexamine differences among the groups (i.e., deficit,nondeficit, and control) on those non-normally distributedvariables and the categorical variables of demographicdata, a nonparametric statistic, the Kruskal–WallisH-test,was utilized. Additionally, theMann–WhitneyU-test wasused to establish which comparisons accounted for thesignificance of a given effect. For normally distributedand homogeneous variables, comparisons among thethree groups were conducted by a series of one-wayanalyses of variance (ANOVAs). Post hoc analyses (LSDtests) were carried out only when a significant main effecthad been found in the ANOVAs. For all analyses, thesignificance level was set at Pb0.05.

Measures from the BPRS were grouped into fivefactors: (1) anxiety/depression, (2) anergia; (3) thoughtdisturbance; (4) activation and (5) hostility/suspicious-ness. Indices used in analyses of the SANS included thetotal score and five subscale scores (affective flattening,alogia, avolition-apathy, anhedonia-asociality, and atten-tion). Similarly, the indices used in analyses of the SAPSincluded the total score and four subscale scores(hallucinations, delusions, bizarre behavior, and positivethought disorder). The general assessment score of eachsubscale was used as the subscale score for the SANS andSAPS in this study. Comparisons of clinical symptomsbetween the deficit and nondeficit groups were conductedusing t-tests; the level of significance was set at Pb0.05.

3. Results

3.1. Demographic characteristics

In the present study, 30 participants (24.4%) met thecriteria for deficit syndrome schizophrenia, while 93participants (75.6%) were classified as nondeficit.Demographic information of the sample and clinicalcharacteristics for the deficit and nondeficit groups arepresented in Table 1. There are no significant differences

Table 1Demographic characteristics for deficit schizophrenia, nondeficit schizophre

DS (n=30) NDS

Age 42.63±9.26 42.66Education (years) 10.30±2.88 10.94Male/female 21/9 64/29Marital status (married/not married) 15/15 59/34Mean age of onset 25.47±6.94 26.04Duration of illness (years) 17.17±8.50 17.02CPZ-equivalent daily dose (mg) 509.29±245.66 511.96

DS=deficit schizophrenia, NDS=nondeficit schizophrenia. Comparative anaand t-tests, and comparative analysis of categorical variables were performe⁎⁎Pb0.01.

among the three groups with respect to age, education,or gender distribution. However, the three groupsdiffered significantly in marital status (χ2 =22.82,Pb0.001), with deficit and nondeficit patients lesslikely to have ever been married than controls (Deficitvs. Normal: Z=4.40, Pb0.001; Nondeficit vs. Normal:Z=3.91, Pb0.001), but the two patient groups do notdiffer significantly (Z=1.30, P=0.193). ANOVA resultsindicated that deficit and nondeficit groups have nosignificant difference in the mean age of onset, durationof illness and antipsychotic medicine dosage (chlor-promazine equivalents).

3.2. Psychopathological evaluation

The three repeated measures ANOVAs on the BPRSfactors and the SANS/SAPS subscales (Table 2) showedsignificant group-by-factor/subscales interactions forthe five factors of the BPRS (F=8.95, df=4, 484,Pb0.001) and for the five subscales of the SANS(F=3.81, df=4, 484, Pb0.01). At the same time, nosignificant interaction was found for the four subscalesof the SAPS (F=1.68, df=3, 363, PN0.05). Follow-upunivariate tests showed that deficit schizophrenicpatients had higher scores on the BPRS anergia factorand the SANS affective flattening, alogia, avolition-apathy, and anhedonia-asociality subscales, and hadlower scores on the SAPS delusions subscale (seeTable 2). For the SANS attention subscale, the diffe-rence between deficit and nondeficit schizophreniaapproached significance (F=3.10, P=0.08).

3.3. Neuropsychological evaluation

The mean±S.D. scores or the median (quartile range)scores for the deficit and nondeficit patients and thehealthy comparison subjects are presented in Table 3.One-way ANOVAs or Kruskal–Wallis tests were per-formed to assess the group differences for each test.

nia and normal control subjects

(n=93) Normal (n=103) F/χ2 P

±10.02 40.91±10.45 1.046 0.353±2.76 10.36±2.94 0.937 0.393

71/32 0.016 0.99290/13 22.815⁎⁎ 0.000

±8.19 – 0.120 0.729±9.21 – 0.006 0.940±259.04 – 0.002 0.962

lyses of continuous variables were carried out with one-way ANOVAsd using the chi-square test.

Table 2Severity of psychopathology for deficit and nondeficit schizophrenia

DS (n=30) NDS (n=93) F (df=1, 121) P

BPRS Total 32.87±5.14 31.10±7.45 1.462 0.229Anxiety/depression 6.67±1.63 6.81±2.29 0.096 0.757Anergia 10.87±2.70 8.38±2.55 20.981⁎⁎ 0.000Thought disturbance 6.63±1.99 7.25±3.00 1.139 0.288Activation 3.63±1.22 3.57±1.35 0.052 0.820Hostility/suspiciousness 4.80±1.69 5.00±2.09 0.227 0.634

SANS Total 53.70±15.82 35.81±21.41 17.766⁎⁎ 0.000Affective flattening 2.80±0.92 1.81±1.13 19.172⁎⁎ 0.000Alogia 2.03±0.89 1.54±1.08 5.184⁎ 0.025Avolition-apathy 2.63±0.67 1.84±1.12 13.587⁎⁎ 0.000Anhedonia-asociality 2.83±0.91 2.03±1.13 12.485⁎⁎ 0.001Attention 1.30±0.95 0.96±0.92 3.101 0.081

SAPS Total 16.40±12.15 15.62±14.28 0.072 0.789Hallucinations 0.50±0.82 0.73±1.15 1.035 0.311Delusions 1.17±0.87 1.69±1.27 4.388⁎ 0.038Bizarre behavior 1.03±0.96 1.18±1.03 0.491 0.485Positive thought disorder 1.20±0.81 1.13±1.01 0.089 0.766

Follow-up univariate tests conducted after repeated measures ANOVAs showed a significant group-by-factor interaction (F=8.950, df=4, 484,Pb0.0001) for the five factors of the BPRS and a significant group-by-subscale interaction (F=3.805, df=4, 484, Pb0.01) for the five subscalesof the SANS, while no significant group-by-subscale interaction emerged (F=1.680, df=3, 363, PN0.05) for the four subscales of the SAPS.⁎Pb0.05, ⁎⁎Pb0.01.

200 X. Wang et al. / Psychiatry Research 158 (2008) 195–205

Overall, significant group effects were found for allneurocognitive indices with the exception of the numberof nonperseverative errors on the WCST. The post-hocanalysis showed that both deficit and nondeficit groups

Table 3Neuropsychological test scores of deficit schizophrenia, nondeficit schizoph

Tests DS (n=30) NDS (n=93)

Mean±S.D. Mean±S.D.

WAIS (Block Design) 7.67±3.19c 8.92±2.59WAIS (Vocabulary) 7.90±3.25c 9.90±2.62IQ 86.96±15.81c 96.55±13.17WMS (Visual Reproduction) 6.94±3.32 7.34±3.19WMS (Logical Memory) 8.53±5.43 9.19±4.13

Median (quartile) Median (quarti

Benton Judgment of Lines 24.5 (4.63) 24.5 (3.5)Trail Making TestsTrail A (time) 89.5 (61.75) 68.5 (51.5)Trail B (time) 295 (390.0)c 193.5 (157.0)Flexibility index (Trail B minus A) 212.5 (365.5)c 124.55 (153.75

WCSTNumber of perseverative errors 39.50 (57.25)c 33.00 (26.5)Number of nonperseverative errors 19.5 (27.75) 21.00 (18.5)Number of categories completed 1.00 (3.0)c 2.00 (3.5)

One-way ANOVAs and follow-up post hoc analyses (LSD tests) were pernonnormally distributed and nonhomogeneous variables were evaluated usiMann–Whitney U-test).⁎Pb0.05, ⁎⁎Pb0.01.aDeficit patients vs. normal controls. bNondeficit patients vs. normal control

performed worse than controls (Pb0.05) on thoseindices on which the ANOVA was significant.

Univariate tests showed significantly lower proratedIQ in patients with deficit schizophrenia than that in

renia and normal control subjects

Normal (n=103) Analysis

Mean±S.D. F (one-way ANOVA) Cohen's d(DS vs. NDS)

10.24±2.99 a,b 11.305⁎⁎ 0.4312.05±2.42a,b 35.649⁎⁎ 0.68106.75±13.27a,b 30.167⁎⁎ 0.669.87±2.50a,b 23.366⁎⁎ 0.1213.47±4.22a,b 28.670⁎⁎ 0.14

le) Median (quartile) χ2 (Kruskal–Wallis H-test)

26 (4.0)a,b 8.124⁎ 0.15

49 (34.0)a,b 34.931⁎⁎ 0.13119 (72.25)a,b 41.862⁎⁎ 0.65

) 67 (81.0)a,b 25.181⁎⁎ 0.64

26.00 (22.0)a,b 31.841⁎⁎ 0.5717.00 (18.0) 1.399 0.124.00 (3.0)a,b 36.063⁎⁎ 0.56

formed on the normally distributed and homogeneous variables. Theng nonparametric statistics (Kruskal–Wallis H-test and the follow-up

s. cDeficit patients vs. nondeficit patients.

201X. Wang et al. / Psychiatry Research 158 (2008) 195–205

nondeficit schizophrenia (F=10.978, df=1, 121,Pb0.001). Additionally, the deficit patients performedsignificantly worse on one of the two parietal measures,the Block Design test (F=4.668; df=1, 121; Pb0.05),while the difference between the two groups on thesecond parietal measure, the Benton Judgment of Linestest, indicated non-significant findings (Z=0.428,P=0.668). The deficit patients, compared with thenondeficit patients, had significantly longer Trail B timeand Flexibility index (Z=2.462, Pb0.05; Z=2.306,Pb0.05), whereas there was no significant differencebetween the two subgroups on Trail A time. On theWCST, the deficit group made more perseverative errorsand achieved fewer categories than the nondeficit group(Z=2.507, Pb0.05; Z=2.850, Pb0.01). The effect sizefor deficit vs. nondeficit in Block design was 0.43, andthose for the other indices on which the deficit–non-deficit differences were significant all achieved mediumlevel effect sizes (Cohen's d N0.50, Cohen, 1988).1

There were no significant differences between the twoschizophrenic subgroups on any of the explicit memorymeasures (i.e. visual reproduction and logical memory).Overall, the deficit group performed worse than thenondeficit group for the two frontal lobe function mea-sures, one of the two parietal lobe function measures andthe general ability measure, prorated IQ, whereas therewere no significant differences between the two groupson the two temporal lobe function measures.

4. Discussion

To date, the present study is the first to utilize Car-penter's categorization (Carpenter et al., 1988) of deficitschizophrenia in a Chinese sample to examine thedifferences in clinical symptoms and neuropsychologicalfunctions between deficit and nondeficit schizophrenia.Our results are consistent with findings in past studiesregarding the psychopathological characteristics of deficitand nondeficit patients (Buchanan et al., 1990; Fenton andMcGlashan, 1994; Kirkpatrick et al., 1996c; Roy et al.,2001; Tiryaki et al., 2003). In line with our hypotheses,results from the neuropsychological test battery suggestedthat deficit and nondeficit schizophrenia exhibited differentpatterns of impaired cognitive functioning. Further,significant differences were found in executive andvisuospatial functioning between the deficit and nondeficitsubgroups, but not explicit memory functioning.

1 After excluding the scores of six questionable cases in thenondeficit group, the effect sizes for those neuropsychological tests onwhich the deficit–nondeficit differences achieved significance wereall greater and at the medium level (Cohen's d N0.50).

In the present study, 30 out of 123 chronic schizo-phrenia patients (24.4%) met criteria for the deficitsubtype, which was consistent with the rate reported inprevious studies. According to a recent review of studiesin Europe and North America, prevalences of the deficitsubtype are roughly 15% for the first-episode patientsand 25% to 30% for those with chronic schizophrenia(Kirkpatrick et al., 2001). The similarity in theprevalence of deficit schizophrenia of Chinese originsupported the validity of our deficit and nondeficitcategorization.

In addition to similarity in the prevalence, results fromthe Chinese sample are also consistent with the clinicalsyndrome profiles of previous studies (Kirkpatrick et al.,1994, 1996a,b; Subotnik et al., 2000; Roy et al., 2001;Tiryaki et al., 2003). On the BPRS, deficit patientsreceived higher ratings than nondeficit patients only onthe anergia factor, but not on the other factors and totalscore. Further, deficit patients received higher ratingsthan nondeficit patients on the global severity and eachsubscale of the SANS with the exception of the attentionsubscale (P=0.081), but not on the global severity or anysubscale of the SAPS. These findings suggested thatprimary and enduring negative symptoms of deficitschizophrenia were not secondary to factors other thanthe disease process, i.e. depression or anxiety disorder,severe suspiciousness, delusions, hallucinations, orformal thought disorder. Additionally, it is important tonote that demographic variables including age, educa-tion, gender composition, or duration of illness did notaccount for differences in the clinical characteristicbetween the two schizophrenic subgroups. Thus, theresults above provided evidence that the deficit andnondeficit categorization was both replicable and reliablein the Chinese sample.

The examination of symptomatological characteris-tics in a vastly different geographical and racial settingmakes some valuable contributions. The first is that asyndrome with clinical and demographic features andcorrelates very similar to those of Western patients canbe delineated in a Chinese sample using the SDS. Thissuggests that the underlying biological abnormalitieshave similar manifestations across cultures. Second, thecomparison of demographic and psychopathologicalcharacteristics will also help researchers to betterunderstand the differences between deficit and non-deficit schizophrenia regarding the impairment ofneuropsychological functioning.

Similar to past studies examining Western samples(Buchanan et al., 1994; Putnam and Harvey, 2000;Bryson et al., 2001; Galderisi et al., 2002), the presentstudy found that deficit patients showed greater

202 X. Wang et al. / Psychiatry Research 158 (2008) 195–205

impairment in two executive function measures, theWCST and the Trail Making Test, which are sensitive tofrontal lobe dysfunction. In addition, deficit patientsperformed worse than the nondeficit patients on one oftwo visuospatial ability tests, the Block Design test.There have been relatively inconsistent studies examin-ing the parietal lobe dysfunction of deficit schizophre-nia. Although many studies reported that deficit patientsdemonstrated more impairment than nondeficit patientson the tasks mediated by the parietal circuitry (Thakeret al., 1989; Buchanan et al., 1994, 1997; Malaspinaet al., 1994, 2002; Ross, 2000; Ross et al., 1996, 1997;Nkam et al., 2001; Goudsmit et al., 2003; Seckingeret al., 2004), some studies argued against the specialimpairment of parietal lobe function in deficit schizo-phrenia (Quarantelli et al., 2002; Tiryaki et al., 2003).

Compared with the neuropsychological studies, theanatomy-related studies showed more consistent results.Many structural and functional imaging studies reportedthe prefrontal and parietal cortex as the possiblesubstrates involved in the neural process of deficitschizophrenia (Tamminga et al., 1992; Carpenter et al.,1996; Buchanan et al., 1997; Heckers et al., 1999;Turetsky et al., 1995, 1998; Kirkpatrick et al., 1999,2003; Lahti et al., 2001; Yurekli et al., 2003; Gonul etal., 2003; Delamillieure et al., 2000, 2004). Addition-ally, our neuroelectrophysiological and functionalneuroanatomical imaging studies revealed that prefron-tal lobe dysfunction was associated with deficits inworking memory tasks for deficit patients (unpublisheddata).

Based on the consistent results of analyses of multipleareas, the prefrontal cortex dysfunction would appear tobe an important component that is particularly associatedwith deficit symptoms. Further, the relative inconsistencyin the neuropsychological studies and the more consistentanatomy-related findings of a specific associationbetween the parietal cortex and deficit syndromes suggestthat only part of the parietal cortex, likely the inferiorparietal lobe which is connected to the prefrontal cortex,may be involved in the neural circuitry underlying thecognitive impairment of deficit schizophrenia (Kirkpa-trick et al., 2001). Additional investigations with morespecific cognitive probes should be conducted.

Contrary to the frontal and parietal functional mea-sures, deficit and nondeficit patients did not showsignificant differences on any explicit memory mea-sures, with both schizophrenic groups performing lesswell than the normal controls, which was also consistentwith previous studies (Buchanan et al., 1994; Putnamet al., 2000; Bryson et al., 2001; Galderisi et al., 2002).Thus, the lack of deficit vs. nondeficit differences for

explicit memory may indicate that the temporal lobedysfunction is a shared impairment of both schizo-phrenic subgroups.

Interestingly, in the present study we found that deficitpatients performed significantly worse than nondeficitpatients on the WAIS vocabulary subtest, which isthought to be relatively more strongly associated withtemporal functions. In addition, in this study the largesteffect size was on the vocabulary measures (0.68). Theremay be two reasons to explain the markedly poorervocabulary functioning of deficit schizophrenia. First,whereas alphabetical systems are based on the associa-tion of phonemes with graphemic symbols, Chinese isbased inherently on the association of meaningfulmorphemes with graphic units. Thus, the Vocabularysubtest of the WAIS-RC, which measures the compre-hension of vocabulary and expression of language, mayrelate to the functions of more brain areas beyond thetemporal lobe, especially the frontal lobe (Tan et al.,2000; Chee et al., 2000; Yoon et al., 2006). And thefrontal functioning impairment of deficit patients mayinfluence their performances on the vocabulary test.Second, the poorer vocabulary functioning of deficitpatients may be related to their severe symptom of“poverty of speech”.

As prevalence rates, symptom profile, and neuro-psychological patterns provided evidence for the deficitand nondeficit categorization of schizophrenic patients,results from the intelligence testing yielded similarfindings. Similar to the results from Diao et al. (1999)and Galderisi et al. (2002), the present study found thatdeficit patients demonstrated significantly more severeimpairment than nondeficit patients in general intellec-tual function. Further, both subgroups performedsignificantly worse than the normal control group.Some studies failed to find a significant relationshipbetween lower IQ and deficit schizophrenia (Buchananet al., 1994; Horan and Blanchard, 2003; Seckingeret al., 2004). One possible explanation may be the sizeof the sample that was examined. In those studies,sample sizes ranged from 13 to 18, and as a result oflimited statistical power, it likely limited the ability todetect the group differences.

Understanding the deficit–nondeficit differences inintelligence is an important issue in interpreting thepresent study results. According to the neurodevelop-mental hypothesis, low IQ is an indicator of the earlyneurobiological impairment of schizophrenia from child-hood (Addington et al., 1991; Jones and Done, 1997;David et al., 1997). Recently, intelligence impairment hasbeen found in significant association with deficit schizo-phrenia. Further, intelligence impairment in schizophrenia

203X. Wang et al. / Psychiatry Research 158 (2008) 195–205

has also been related to other features that differentiatedeficit and nondeficit patients, specifically neurologicalsigns, poor premorbid adjustment and lower socioeco-nomic status (Buchanan et al., 1990; Kirkpatrick et al.,2001; Galderisi et al., 2002).

In the present study, the greater general intellectualimpairment in deficit schizophrenia provided furthersupport for the special relationship between low IQ andthe deficit syndrome. Therefore, it is not clear that IQshould be treated as a “nuisance” variable to becontrolled in comparisons of the neuropsychologicalperformance of deficit and nondeficit schizophrenia.With respect to the complex interaction betweenintelligence and deficit schizophrenia, it may beimpossible to disentangle the effects of intelligencefrom more specific impairments on neuropsychologicaltests using statistical means.

Several limitations of the current study should beconsidered. First, as the prevalence of the deficit subtypein the present study was 24.4% (30 out of 123), the size ofthe deficit group was relatively small and the number ofcomparisons, including the nondeficit and control groups,was relatively high, which may likely increase the chanceof significant findings. Second, according to the instruc-tions of the SDS manual, patients with a questionabledeficit or nondeficit classification were placed in thenondeficit group. However, it is possible that heteroge-neity within the nondeficit group may have influenced theresults of our study. Therefore, it might considered infuture studies whether other criteria should be used todefine the nondeficit schizophrenia subgroup. Finally, dueto the complexity of brain-behavior relations, the resultsof neuropsychological studies can only suggest thepresence of frontal and parietal lobe impairments indeficit schizophrenia. Further studies in multiple areas areneeded to explore the underlying neural mechanisms ofthe deficit syndrome. Our future reports will focus on thedifferences between deficit and nondeficit schizophreniain working memory tasks related to frontal and parietallobe function when utilizing neuroelectrophysiologicaland functional neuroimaging methods.

In summary, the current study replicated previouslyreported differences in clinical symptom patterns be-tween deficit and nondeficit schizophrenia in a Chinesesample. The results provide further support for thevalidity of the deficit–nondeficit categorization of schi-zophrenia. At the same time, the present study expandsupon previous studies by providing evidence that deficitschizophrenia is associated with frontal and parietal lobeimpairment, and that temporal dysfunction may be acommon cognitive impairment of the two subtypes ofschizophrenia.

Acknowledgment

The authors thank Dr. Robert W. Buchanan forhelping with the design of the study. Additionally, wethank Professor Lin Xu and Randy Patrick Auerbach fortheir suggestions and proofreading of an earlier versionof the manuscript.

References

Addington, J., Addington, D., Maticka-Tyndale, E., 1991. Cognitivefunctioning and positive and negative symptoms in schizophrenia.Schizophrenia Research 5, 123–134.

Amador, X.F., Kirkpatrick, B., Buchanan, R.W., Carpenter, W.T.,Marcinko, L., Yale, S.A., 1999. Stability of the diagnosis of deficitsyndrome in schizophrenia. American Journal of Psychiatry 156,637–639.

American Psychiatric Association, 1994. Diagnostic and StatisticalManual of Mental Disorders, 4th ed. Author, Washington, DC.

Andreasen, N.C., 1983. Scale for the Assessment of NegativeSymptoms. (SANS). University of Iowa, Iowa City.

Andreasen, N.C., 1984. Scale for the Assessment of PositiveSymptoms. (SAPS). University of Iowa, Iowa City.

Arango, C., Buchanan, R.W., Kirkpatrick, B., Carpenter, W.T., 2004.The deficit syndrome in schizophrenia: implications for thetreatment of negative symptoms. European Psychiatry 19, 21–26.

Benton, A.L., Hannay, H.J., Varney, N.R., 1975. Visual perception ofline direction in patients with unilateral disease. Neurology 25,907–910.

Benton, A.L., Sivan, A.B., Hamsher, K.D., Varney, N.R., Spreen, O.,1994. Judgment of line orientation. Contributions to Neuropsy-chological Assessment: A Clinical Manual. Oxford UniversityPress, New York, pp. 44–54.

Bryson, G., Whelahan, H.A., Bell, M., 2001. Memory and executivefunction impairments in deficit syndrome schizophrenia. Psychi-atry Research 102, 29–37.

Buchanan, R.W., Kirkpatrick, B., Heinrichs, D.W., Carpenter Jr., W.T.,1990. Clinical correlates of the deficit syndrome of schizophrenia.American Journal of Psychiatry 147, 290–294.

Buchanan, R.W., Strauss, M.E., Kirkpatrick, B., Holstein, C., Breier,A., Carpenter Jr., W.T., 1994. Neuropsychological impairments indeficit vs. nondeficit forms of schizophrenia. Archives of GeneralPsychiatry 51, 804–811.

Buchanan, R.W., Strauss,M.E., Breier, A., Kirkpatrick, B., Carpenter Jr.,W.T., 1997. Attentional impairments in deficit and nondeficit formsof schizophrenia. American Journal of Psychiatry 154, 363–370.

Buchanan, R.W., Breier, A., Kirkpatrick, B., Ball, P., Carpenter Jr.,W.T., 1998. Positive and negative symptom response to clozapinein schizophrenic patients with and without the deficit syndrome.American Journal of Psychiatry 155, 751–760.

Carpenter Jr., W.T., Heinrichs, D.W., Wagman, A.M., 1988. Deficitand nondeficit forms of schizophrenia: the concept. AmericanJournal of Psychiatry 145, 578–583.

Carpenter Jr., W.T., Lahti, A.C., Holcomb, H.H., Zhao, M., Buchanan,R.W., Tamminga, C.A., 1996. Frontal and parietal blood flowactivation during an auditory task differentiate schizophrenicpatients with and without primary negative symptoms. Abstracts -Society for Neuroscience 22 (676) (Abstract 268.4).

Chee, M.W.L., Weekes, B., Lee, K.M., Soon, C.S., Schreiber, A.,Hoon, J.J., Chee, M., 2000. Overlap and dissociation of semantic

204 X. Wang et al. / Psychiatry Research 158 (2008) 195–205

processing of Chinese characters, English words, and pictures:evidence from fMRI. NeuroImage 12, 392–403.

Cohen, J., 1988. Statistical PowerAnalysis for the Behavioral Sciences,2nd ed. Lawrence Erlbaum Associates, Hillsdale, NJ.

Cohen, A.S., Docherty, N.M., 2004. Deficit versus negative syndromein schizophrenia: prediction of attentional impairment. Schizo-phrenia Bulletin 30, 827–835.

David, A.S., Malmberg, A., Brandt, L., Allebeck, P., Lewis, G., 1997.IQ and risk for schizophrenia: a population-based cohort study.Psychological Medicine 27, 1311–1323.

Delamillieure, P., Fernandez, J., Constans, J.M., Brazo, P., Benali, K.,Abadie, P., Vasse, T., Thibaut, F., Courtheoux, P., Petit, M., Dollfus,S., 2000. Proton magnetic resonance spectroscopy of the medialprefrontal cortex in patients with deficit schizophrenia: preliminaryreport. American Journal of Psychiatry 157, 641–643.

Delamillieure, P., Constans, J.M., Fernandez, J., Brazo, P., Dollfus, S.,2004. Relationship between performance on the Stroop test andN-acetylaspartate in the medial prefrontal cortex in deficit andnondeficit schizophrenia: preliminary results. Psychiatry Re-search: Neuroimaging 132, 87–89.

Diao, J.R., Li, C., Lv, F., Yu, X.J., 1999. TheWAIS-RC performance indeficit and nondeficit schizophrenia. Shandong Psychiatry 12,30–32.

Earnst, K.S., Kring, A.M., 1999. Emotional responding in deficit andnon-deficit schizophrenia. Psychiatry Research 88, 191–207.

Fenton, W.S., McGlashan, T.H., 1994. Antecedents, symptomprogression, and long-term outcome of the deficit syndrome inschizophrenia. American Journal of Psychiatry 151, 351–356.

First, M.B., Gibbons, M., Spitzer, R.L., Williams, J.B.W., 1996a.Users Guide for the Structured Clinical Interview for DSM-IVAxisI Disorders-Research Version (SCID-I, Version 2.0, February 1996Final Version). Biometrics Research Department, New York.

First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 1996b.Structured Clinical Interview for DSM-IVAxis I Disorders: Non-patient Edition (SCID-NP). Biometrics Research Department, NewYork.

Galderisi, S., Maj, M., Mucci, A., Cassano, G.B., Invernizzi, G., Rossi,A., Vita, A., Dell'Osso, L., Daneluzzo, E., Pini, S., 2002. Histor-ical, psychopathological, neurological, and neuropsychologicalaspects of deficit schizophrenia: a multicenter study. AmericanJournal of Psychiatry 159, 983–990.

Gong, Y.X., 1982. Manual for the Wechsler Adult Intelligence Scale-Chinese Revision. Hunan Medical College, Changsha, Hunan,China.

Gong, Y.X., 1986. The Revision of Adult Halstead-Reitan Test Batteryin China. Acta Psychologica Sinica 18, 433–442.

Gong, Y.X., 1989. Manual for the Wechsler Memory Scale-ChineseRevision. Hunan Medical College, Changsha, Hunan, China.

Gong, Y.X., 1993. Manual for the Halstead Reitan Battery Revised inChina. Hunan Medical College, Changsha, Hunan, China.

Gonul, A.S., Kula, M., Esel, E., Tutus, A., Sofuoglu, S., 2003. A Tc-99m HMPAO SPECT study of regional cerebral blood flow indrug-free schizophrenic patients with deficit and non-deficitsyndrome. Psychiatry Research: Neuroimaging 123, 199–205.

Goudsmit, N., Coleman, E., Seckinger, R.A., Wolitzky, R., Stanford,A.D., Corcoran, C., Goetz, R.R., Malaspina, D., 2003. A briefsmell identification test discriminates between deficit and non-deficit schizophrenia. Psychiatry Research 120, 155–164.

Heaton, R.K., 1981. Wisconsin Card Sorting Test Manual. Psycho-logical Assessment Resources, Inc., Odessa., FL.

Heckers, S., Goff, D., Schacter, D.L., Savage, C.R., Fischman, A.J.,Alpert, N.M., Rauch, S.L., 1999. Functional imaging of memory

retrieval in deficit vs. nondeficit schizophrenia. Archives of GeneralPsychiatry 56, 1117–1123.

Horan, W.P., Blanchard, J.J., 2003. Neurocognitive, social, andemotional dysfunction in deficit syndrome schizophrenia. Schizo-phrenia Research 65, 125–137.

Jarvis, P.E., Barth, J.T., 1986. Halstead-Reitan Test Battery: anInterpretative Guide. Psychological Assessment Resources, Inc.,Odessa, FL.

Jones, P., Done, D.J., 1997. From birth to onset: a developmentalperspective of schizophrenia in two national birth cohorts. In:Keshavan, M.S., Murray, R.M. (Eds.), Neurodevelopment andAdult Psychopathology. Cambridge University Press, Cambridge,UK, pp. 119–136.

Kirkpatrick, B., Buchanan, R.W., 1990. Anhedonia and the deficitsyndrome of schizophrenia. Psychiatry Research 31, 25–30.

Kirkpatrick, B., Buchanan, R.W., McKenney, P.D., Alphs, L.D.,Carpenter Jr., W.T., 1989. The Schedule for the Deficit syndrome:an instrument for research in schizophrenia. Psychiatry Research30, 119–123.

Kirkpatrick, B., Buchanan, R.W., Breier, A., Carpenter Jr., W.T., 1993.Case identification and stability of the deficit syndrome ofschizophrenia. Psychiatry Research 47, 47–56.

Kirkpatrick, B., Buchanan, R.W., Breier, A., Carpenter Jr., W.T., 1994.Depressive symptoms and the deficit syndrome of schizophrenia.Journal of Nervous and Mental Disease 182, 452–455.

Kirkpatrick, B., Amador, X.F., Flaum, M., Yale, S.A., Gorman, J.M.,Carpenter Jr., W.T., Tohen, M., McGlashan, T., 1996a. The deficitsyndrome in the DSM-IV Field Trial: I. Alcohol and other drugabuse. Schizophrenia Research 20, 69–77.

Kirkpatrick, B., Amador, X.F., Yale, S.A., Bustillo, J.R., Buchanan,R.W., Tohen, M., 1996b. The deficit syndrome in the DSM-IVField Trial. Part II. Depressive episodes and persecutory beliefs.Schizophrenia Research 20, 79–90.

Kirkpatrick, B., Ram, R., Bromet, E., 1996c. The deficit syndrome inthe Suffolk County Mental Health Project. Schizophrenia Research22, 119–126.

Kirkpatrick, B., Conley, R.C., Kakoyannis, A., Reep, R.L., Roberts,R.C., 1999. Interstitial cells of the white matter in the inferiorparietal cortex in schizophrenia: an unbiased cell-counting study.Synapse 34, 95–102.

Kirkpatrick, B., Buchanan, R.W., Ross, D.E., Carpenter Jr., W.T.,2001. A separate disease within the syndrome of schizophrenia.Archives of General Psychiatry 58, 165–171.

Kirkpatrick, B., Messias, N.C., Conley, R.R., Roberts, R.C., 2003.Interstitial cells of the white matter in the dorsolateral prefrontalcortex in deficit and nondeficit schizophrenia. Journal of Nervousand Mental Disease 191, 563–567.

Kopelowicz, A., Liberman, R.P., Mintz, J., Zarate, R., 1997.Comparison of efficacy of social skills training for deficit andnondeficit negative symptoms in schizophrenia. American Journalof Psychiatry 154, 424–425.

Kopelowicz, A., Zarate, R., Tripodis, K., Gonzalez, V., Mintz, J., 2000.Differential efficacy of olanzapine for deficit and nondeficitnegative symptoms in schizophrenia. American Journal ofPsychiatry 157, 987–993.

Lahti, A.C., Holcomb, H.H., Medoff, D.R., Weiler, M.A., Tamminga,C.A., Carpenter Jr., W.T., 2001. Abnormal patterns of regionalcerebral blood flow in schizophrenia with primary negativesymptoms during an effortful auditory recognition task. AmericanJournal of Psychiatry 158, 1797–1808.

Maj, M., 1998. Critique of the DSM-IVoperational diagnostic criteriafor schizophrenia. British Journal of Psychiatry 172, 458–460.

205X. Wang et al. / Psychiatry Research 158 (2008) 195–205

Malaspina, D., Wray, A.D., Friedman, J.H., Amador, X., Yale, S.,Hasan, A., Gorman, J.M., Kaufmann, C.A., 1994. Odor discrim-ination deficits in schizophrenia: association with eye movementdysfunction. Journal of Neuropsychiatry and Clinical Neuros-ciences 6, 273–278.

Malaspina, D., Coleman, E., Goetz, R.R., Harkavy-Friedman, J.,Corcoran, C., Amador, X., Yale, S., Gorman, J.M., 2002. Odoridentification, eye tracking and deficit syndrome schizophrenia.Biological Psychiatry 51, 809–815.

Muriel, D.L., 1983. Neuropsychological Assessment. Oxford Univer-sity Press, New York, pp. 349–350.

Nkam, I., Thibaut, F., Denise, P., Van Der. Elst., A., Segard, L., Brazo,P., Menard, J., Thery, S., Halbeck, I., Delamilleure, P., Vasse, T.,Etard, O., Dollfus, S., Champion, D., Levillain, D., Petit, M., 2001.Saccadic and smooth-pursuit eye movements in deficit and non-deficit schizophrenia. Schizophrenia Research 48, 145–153.

Overall, J.E., Gorham, D.R., 1962. The brief psychiatric rating scale.Psychological Reports 18, 799–812.

Putnam, K.M., Harvey, P.D., 2000. Cognitive impairment and enduringnegative symptoms: a comparative study of geriatric and nonger-iatric schizophrenia patients. Schizophrenia Bulletin 26, 867–878.

Quarantelli, M., Larobina, M., Volpe, U., Amati, G., Tedeschi, E.,Ciarmiello, A., Brunetti, A., Galderisi, S., Alfano, B., 2002.Stereotaxy-based regional brain volumetry applied to segmentedMRI: validation and results in deficit and nondeficit schizophrenia.NeuroImage 17, 373–384.

Reitan, R.M., 1978. Manual for Administration of NeuropsychologicalTest Battery: For Adults and Children. Reitan NeuropsychologyLaboratories, Tucson, Arizona.

Ross, D.E., Thaker, G.K., Buchanan, R.W., Lahti, A.C., Medoff, D.,Bartko, J.J., Moran, M., Hartley, J., 1996. Association of abnormalsmooth pursuit eye movements with the deficit syndromein schizophrenic patients. American Journal of Psychiatry 153,1158–1165.

Ross, D.E., 2000. The deficit syndrome and eye tracking disorder mayreflect a distinct subtype within the syndrome of schizophrenia.Schizophrenia Bulletin 26, 855–866.

Ross, D.E., Thaker, G.K., Buchanan, R.W., Kirkpatrick, B., Lahti, A.C.,Medoff, D., Bartko, J.J., Goodman, J., Tien, A., 1997. Eye trackingdisorder in schizophrenia is characterized by specific ocular motordefects and is associated with the deficit syndrome. BiologicalPsychiatry 42, 781–796.

Roy, M.A., Merette, C., Maziade, M., 2001. Subtyping schizophreniaaccording to outcome or severity: a search for homogeneoussubgroups. Schizophrenia Bulletin 27, 115–138.

Seckinger, R.A., Goudsmit, N., Coleman, E., Harkavy-Friedman, J.,Yale, S., Rosenfield, P.J., Malaspina, D., 2004. Olfactoryidentification and WAIS-R performance in deficit and nondeficitschizophrenia. Schizophrenia Research 69, 55–65.

Silverstein, A.B., 1982. Two- and four-subtest short forms of theWechsler Adult Intelligence Scale-Revised. Journal of Consultingand Clinical Psychology 50, 415–418.

Subotnik, K.L., Nuechterlein, K.H., Ventura, J., 2000. MMPI discrim-inators of deficit vs. non-deficit recent-onset schizophreniapatients. Psychiatry Research 93, 111–123.

Tamminga, C.A., Thaker, G.K., Buchanan, R., Kirkpatrick, B., Alphs,L.D., Chase, T.N., Carpenter Jr., W.T., 1992. Limbic systemabnormalities identified in schizophrenia using positron emissiontomography with fluorodeoxyglucose and neocortical alterationswith deficit syndrome. Archives of General Psychiatry 49, 522–530.

Tang, C.M., Wang, X.D., 2001. Single Neuropsychological Test. In:Tang, C.M. (Ed.), Neuropsychology. People's Military SurgeonPress, China, pp. 315–332.

Tan, L.H., Spinks, J.A., Gao, J.H., Liu, H.L., Perfetti, C.A., Xiong, J.,Stofer, K.A., Pu, Y., Liu, Y., Fox, P.T., 2000. Brain activation in theprocessing of Chinese characters and words: a functional MRIstudy. Human Brain Mapping 10, 16–27.

Tek, C., Kirkpatrick, B., Buchanan, R.W., 2001. A five-year follow-upstudy of deficit and nondeficit schizophrenia. SchizophreniaResearch 49, 253–260.

Thaker, G., Kirkpatrick, B., Buchanan, R.W., Ellsberry, R., Lahti, A.,Tamminga, C., 1989. Oculomotor abnormalities and their clinicalcorrelates in schizophrenia. Psychopharmacology Bulletin 25,491–497.

Tiryaki, A., Yazici, M.K., Anil, A.E., Kabakci, E., Karaagaoglu, E.,Gogus, A., 2003. Reexamination of the characteristics of thedeficit schizophrenia patients. European Archives of Psychiatryand Clinical Neuroscience 253, 221–227.

Tsuang, M.T., Stone, W.S., Faraone, S.V., 2000. Toward reformulatingthe diagnosis of schizophrenia. American Journal of Psychiatry157, 1041–1050.

Turetsky, B.I., Cowell, P.E., Gur, R.C., Grossman, R.I., Shtasel, D.L.,Gur, R.E., 1995. Frontal and temporal lobe brain volumes inschizophrenia. Relationship to symptoms and clinical subtype.Archives of General Psychiatry 52, 1061–1070.

Turetsky, B.I., Colbath, E.A., Gur, R.E., 1998. P300 subcomponentabnormalities in schizophrenia: I. Physiological evidence forgender and subtype specific differences in regional pathology.Biological Psychiatry 43, 84–96.

Wang, X., Yao, S.Q., Fan, X.H., Yi, Y.Q., Zhu, W.A., Yi, J.Y., 2005.The Chinese Version of the Schedule for the Deficit Syndrome:reliability and validity. Chinese Journal of Clinical Psychology 13,392–395.

Wechsler, D., 1981. Wechsler Adult Intelligence Scale-Revised,Manual. The Psychological Corp, New York, Cleveland, Ohio.

Wechsler, D., 1987. Wechsler Memory Scale-Revised Manual. ThePsychological Corporation. Harcourt Brace Jovanovich, Inc, NewYork.

Yoon, H.W., Chung, J.Y., Kim, K.H., Song, M.S., Park, H.W., 2006.An fMRI study of Chinese character reading and picture namingby native Korean speakers. Neuroscience Letters 392, 90–95.

Yurekli, Y., Bodur, Z., Gulseren, L., Mete, L., 2003. Comparison ofregional cerebral blood flow in deficit and nondeficit schizophrenicpatients. Turkish Journal of Psychiatry 14, 255–262.