Impairments of social cues recognition and social functioning in Chinese people with schizophrenia


Kai Wang MD, PhD, Neuropsychology Laboratory, Department of Neurology, First Hospital of Anhui Medical University, Hefei, Anhui Province, China. Email:


Abstract  Previous studies have suggested that social cognition deficits, and impaired social functioning, were associated with schizophrenia. However, specificity of the relationships between social cues recognition and social functioning remain largely undefined. The authors speculated that the two were related, and further that recognition of verbal and non-verbal social cues were impaired in people with schizophrenia. A total of 71 people (40 schizophrenia patients and 31 matched normal controls) voluntarily participated in this study. Social cues recognition abilities were measured by the eye gaze discrimination task and the faux pas recognition task. Social functioning was assessed using the Social Functioning Scale. Correlation analysis revealed a significant association between faux pas and, in particular, the social functioning subscales of the Social Functioning Scale (independence and employment) in patients with schizophrenia. Furthermore, the authors also observed that clinical participants performed significantly worse in both the eye gaze discrimination and faux pas recognition tasks than their healthy counterparts. These findings suggested that impaired social cues recognition in people with schizophrenia may be a possible explanation for their impaired social functioning.


Schizophrenia is a persistent psychotic illness associated with deficits in social functioning. These include difficulties with working, engaging in social relationships, looking after oneself, and participating in recreational and community activities.1,2 Poor social competence is thought to be an important contributor to both vulnerability to relapse and a poorer social prognosis, and hence, adversely affecting the quality of living.3 Social dysfunction has been recognized as a core feature of schizophrenia and indeed (in most systems) constitutes one of the diagnostic criteria.

‘Social cognition is a prerequisite for social functioning’.4‘Impaired social perception and social cognition probably contributes to the understanding of social behavioral problems in schizophrenia’.5 Social cognitive impairments are gaining attention for their possible role in the social and occupational deficits that accompany schizophrenia. Individuals with schizophrenia have been clearly shown to suffer on impairment of social cognition, even during periods of remission from psychotic episodes. Such a deficit is a relatively stable characteristic in most patients, with little progress usually being made through the course of the illness.

Social cognition includes social cue recognition and social problem solving.6 Several studies have found that people with schizophrenia have deficits in perceiving wsocial cues (verbal and non-verbal), such as perception of basic and complex affective facial expression, recognition of familiar social situations,visual scanning, and recognition of others' intentions (i.e. theory of mind [ToM]).7–11

Ability to detect others' mental states may be based on decoding non-verbal social cues such as body movement, facial expression, and gaze direction.12–13 Most studies have indicated that the ability of people with schizophrenia to tell whether other people are looking directly at them, or to the left or right, is impaired.14–15 Deficit of eye gaze discrimination could be the main contributor to impairments in ToM and communication.15

Verbal information could be another kind of social cue through which one can detect what is on someone else's mind. Baron-Cohen et al.'s ‘Faux Pas Recognition’ test is a ToM task which uses verbal story material to measure the ability to recognize mental intention.13 The definition of faux pas can be set out as the situation where a speaker says something without considering whether or not the listener might want to hear it, and which typically has negative consequences which the speaker did not intend. Faux pas is a more advanced ToM test, because detecting it successfully requires both an appreciation that there may be a difference between the speaker and the listener's knowledge states, and also an understanding of the emotional impact of a statement on a listener. Individuals with normal social cognition abilities can usually recognize when someone has committed a faux pas, whereas people with social cognition deficits, such as autism, usually fail to do so. Although a number of studies have suggested that the ability of people with schizophrenia to detect the emotions, minds, or intentions of others is impaired,7–9 as yet there have been no reports of whether or not people with schizophrenia show a similar deficit in perceiving faux pas.

Investigations of the relationship between functional outcome and social cues recognition ability in patients with schizophrenia have yielded mixed findings. Most studies have suggested that social skills and social cognition are related.16 However, the specificity of this relationship remains unclear. Is the ability to recognize verbal and non-verbal social cues essential to social functioning? Is social dysfunction associated with particular deficits in recognition of such cues? Are these relationships independent of other relevant variables, such as general cognitive function, use of antipsychotic medication, and hospitalization history? This study set out to test whether: (i) the ability to perceive verbal and non-verbal social cues is damaged in people with schizophrenia; and (ii) any significant correlation exists between impairments in eye gaze discrimination and faux pas recognition and patients' social function, or a subcomponent thereof.



The study included 40 chronic patients with schizophrenia, all of whom met the 4th edition of the Diagnostic and Statistical Manual diagnostic criteria.17 They were recruited from the Hefei Psychiatry Hospital (affiliated to Anhui Medial University, Anhui Province, China). There were 18 male and 22 female patients, 36 were right-handed, three ambidextrous, and one left-handed. They were aged from 16 to 46 years, and had received between 4 and 16 years of education. The following exclusion criteria were also fulfilled by all patients: (i) no demonstrable brain disease (i.e. no loss of consciousness nor any history of neurological conditions such as epilepsy, Parkinson's disease or brain injury, or head trauma); (ii) no history of mental retardation; and (iii) no evidence of current substance (including alcohol) abuse. The mean duration of schizophrenic illness was 9.2 years (from 2 to 23 years). All patients were on maintenance levels of antipsychotic medicine using clozapine, with a mean dosage of 170.63 mg (standard deviation [SD] 123.08), but no other antipsychotic medications. The mean duration of hospitalization was 149.31 days (SD 272.06), and the mean number of times admitted to hospital was 2.26 (SD 1.91).

In total, 31 healthy adults were recruited from the general population in the same region. They comprised nine males and 22 females, 29 right-handed, and two ambidextrous, were aged between 20 and 59 years, and had received between 6 and 16 years of education. They also met the same exclusion criteria as the schizophrenic group.

All participants had normal vision and hearing and were able to understand the experimental procedures. Informed consent was obtained in writing from all subjects.

Psychopathology rating

All patients underwent a clinical assessment using the Positive and Negative Syndrome Scale for Schizophrenia (PANSS).18 The PANSS includes a structured interview to enable patients to be assessed on 30 items covering positive symptoms (e.g. hallucinations, delusions), negative symptoms (e.g. blunted affect, emotional withdrawal), and general symptoms (e.g. tension, depression). The administering psychiatrist rated the symptom severity on a scale from 1 to 7 for each item.

Social functioning rating

The Social Functioning Scale (SFS),19 a 79-item questionnaire, has been shown to be a reliable, valid, and sensitive measure of social functioning (Birchwood et al. 1990). The SFS measures abilities or performance in seven areas including withdrawal, interpersonal behavior, social activities, recreation, independence-performance, independence-competence, and employment. The authors used the Chinese version of the SFS. The test–retest reliability for an interval of 3 months varying between 0.72 and 0.88 in all the scales, the interrater reliability was 0.87. Concurrent validity was analyzed based on the correlation of the scores from the SFS with Global Assessment of Functioning scale (GAF)17 ranging between 0.62 and 0.81.

The subject answers each question by selecting one of four possible responses from 0 to 3. The sum of the numbers chosen gives the subcomponent and total scores. The SFS was given to all subjects, taking approximately 20–30 min each.

Neuropsychological background tests

The Mini-Mental State Examination (MMSE)20 is a 21-item instrument, on which scores can range from 0 to 30. The items assess orientation, immediate recall, short-term memory, serial sevens, constructional abilities, and the ability to follow verbal and written instructions.

The Digit Span test is taken from the Wechsler Adult Intelligence Scale (Revised Chinese Version)21 and estimates short-term memory and attention span. The test administrator recites a series of groups of numbers. After that, the subject is requested to repeat the numbers forward or backward to the experimenter in the correct sequence. The maximum number of digits repeated correctly in one of two trials is used as the person's score.

The Verbal Fluency test22 assesses the executive and language abilities. It is a controlled word association test in which the subject is instructed to name as many different animals, fruits, and vegetables in 1 min. The score is the number of unique items listed.

Recognition of social cues

Two social cues recognition tasks were given: one non-verbal task (eye gaze discrimination) and one verbal task, faux pas recognition.13,23

Eye gaze discrimination

The stimulus material for this experiment consisted of 90 color photos shot using six models (two children, two younger adult, and two elder adult Chinese). Each model was positioned so that they were looking in five different directions (15° to the left of the camera position, 5° left, 0°, 5° right, and 15° right) with three head orientations (10° to the left of the camera position, straight ahead, and 10° to the right). All the pictures were standardized for eye position and interocular distance, and the background of each shot was excluded with an oval mask. The order of presentation of the 90 pictures was randomized for each subject. The participants were instructed as follows: ‘In each trial, you are going to see one portrait, whose look is either direct or averted. You will have to ascertain where the gaze of the portrait is directed. If the model is looking at you, you say ‘direct’, otherwise you say ‘left’ or ‘right’ accordingly'. The subjects were given 1 point for each photo they identified correctly.

Recognition of faux pas

A total of 10 faux pas stories were used.13 Each involved two or three characters and at least two separate statements, and contained a social faux pas. The experimenter read the story aloud while the participant followed using his or her own print copy. After each story the subject was asked a series of questions. The text remained in front of the participant during the questions, to reduce memory demands. The questions were as follows:

Question 1 (faux pas Recognition question): ‘In the story, did someone say something that they should not have said?’

Questions 2–4 (faux pas Identification questions): ‘What did they say that they should not have said?’, ‘Why shouldn’t the individual in the story have said what they did?' (This question tests whether or not the participant understood that the listener would be hurt or insulted. In other words, can the participant draw an inference about affective mental states?), ‘Why do you think they did say it?’ (This question tests whether the participant understood that the faux pas was unintentional. In other words, can the participant draw an inference about epistemic mental states and intentionality?).

Therefore, there were four faux pas-related questions if the participant said that a faux pas had been committed. Finally, as a control for story comprehension, participants were asked a question about some important detail of the story, participants who answered ‘no’ to the first question answered the control question immediately. Participants' answers were written down by the experimenter, and the session was also tape-recorded. The subjects were given 1 point for each question they answered correctly. The scores for the questions 1–4 across the 10 stories were added up to make the faux pas-related questions.

The authors used the Chinese version of the Faux pas task. The test–retest reliability for an interval of 3 months was 0.83 in faux pas-related questions, the interrater reliability was 0.76.

The PANSS, SFS were completed by an experienced psychiatrist, who routinely uses all of these measures. Interrater reliability was determined by 100% agreement on the diagnosis and at least 80% agreement for symptom presence. The neurocognitive and social cognition data came from laboratory-based assessments in a facility designed for this study, they were given by the first author, who was blinded to the results of the SFS.

Data analysis

Differences between groups were examined using the independent samples t-test. Pearson correlation analysis was used to test for correlation between two continuous variables. The level of significance was set at P = 0.05.


Background data

There was no significant difference between the two groups in terms of age, gender, or years of education. The participants in the schizophrenic group demonstrated the mean level of intelligence, memory, and semantic ability. There was no significant difference between the patients with schizophrenia and the normal controls on these dimensions, as shown in Table 1.

Table 1.  Demographic, clinical data, and neuropsychological background tests scores of patients with schizophrenia and normal controls
 Schizophrenic groupControl groupP-value
  1. PANSS, Positive and Negative Syndrome Scale; MMSE, Mini-Mental State Examination.

Age (years)30.2 (8.02)29.97 (11.03)0.922
Gender (male/female)18m/22f9m/22f0.129
Education level (years)10.65 (2.68)11.74 (2.61)0.089
Handiness36R/3RL/1 L29R/2RL0.662
Disease duration (months)110.44 (82.11)  
Times in hospital2.26 (1.91)  
Duration in hospital (days)149.31 (272.06)  
Clozapine dosage (mg)170.63 (123.08)  
 Positive symptoms9.83 (4.17)  
 Negative symptoms12.3 (4.54)  
 Global psychopathology43.32 (9.91)  
MMSE29.18 (1.10)29.52 (1.00)0.189
Digit Span
 Forwards7.28 (0.91)7.54 (0.71)0.214
 Backwards4.58 (0.77)4.82 (0.73)0.314
Verbal fluency test15.07 (3.85)16.28 (3.02)0.147

Social cues recognition

As summarized in Table 2, the participants with schizophrenia demonstrated impairment in discrimination of eye gaze direction. The difference between the two groups on this measure was significant for 0° (t = −2.286, P < 0.05), but not significant for 15° left, 5°left, 5° right or 15° right (t = −1.838, −0.490, −0.830, −1.515, respectively, and P > 0.05). There were no interactions among six models, five different gaze directions, and three head directions. There was also a significant difference in recognition of faux pas questions and faux pas-related questions (t = −5.533, −5.636, respectively, and P < 0.05) between the schizophrenic and normal control groups, but not significant for control question. The results, therefore, show that the participants with schizophrenia were impaired in their recognition of both verbal and non-verbal social cues (Table 3).

Table 2.  Eye gaze discrimination task of patients with schizophrenia and normal controls
 Schizophrenic group
(n = 40)
Control group
(n = 30)
Significance level
(df = 70)
  • *

    P < 0.05.

  • SE, standard error of mean.

Table 3.  Faux pas recognition of patients with schizophrenia and normal controls
 Schizophrenic group
(n = 37)
Control group
(n = 31)
Significance level
(df = 68)
  • ***

    P < 0.001.

  • SE, standard error of the mean.

Faux pas recognition question6.682.449.161.13−5.5330.000***
Faux pas related questions25.7610.3236.454.72−5.6360.000***
Control questions9.751.0510.000.001.4260.163

Social functioning rating

Table 4 shows that the differences between each of the subscales and the total SFS score were significant between the schizophrenic and normal control groups.

Table 4.  Social functioning assessment of patients with schizophrenia and normal controls
SFS SubscaleSchizophrenic group
(n = 40)
Control group
(n = 31)
Significance level
(df = 71)
  • **

    P < 0.01.

  • *** 

    P < 0.001.

  • SFS, Social Functioning Scale.

Social withdrawal13.782.4515.681.94−3.5410.001**
Social activities5.583.9220.978.66−9.1930.000***
Recreational activities12.405.3419.975.24−5.9700.000***
Independence (competence)33.584.2937.581.34−5.5670.000***
Independence (performance)28.657.6635.651.87−5.5670.000***
SFS total115.0822.98163.5513.29−11.1500.000***

Correlation between Social functioning rating and Positive and Negative Syndrome Scale for Schizophrenia subscales

Table 5 shows the correlation between the SFS and the PANSS subscale scores. The SFS subscale scores were correlated with the PANSS negative symptoms, global psychopathology symptoms and total scores (P < 0.01; P < 0.001), but were not correlated with PANSS positive symptoms scores (P > 0.05).

Table 5.  Pearson correlations between Positive and Negative Syndrome Scale subscales and Social Functioning Scale subscales
 Positive symptomsNegative symptomsGlobal psychopathologyTotal
  • **

    Correlation is significant at the 0.01 level (two-tailed);

  • *** 

    *** Correlation is significant at the 0.001 level (two-tailed).

Social withdrawal−0.247−0.565***−0.494***−0.571***
Social activities−0.129−0.391**−0.359**−0.389**
Recreational activities−0.042−0.441***−0.379**−0.373**
Independence (competence)−0.079−0.447***−0.369**−0.394**
Independence (performance)0.053−0.610***−0.516***−0.481***
SFS total−0.097−0.749***−0.639***−0.654***

Correlation between patient characteristics and social cues recognition

Table 6 shows that there was no significant correlation between discrimination of eye gaze at 0° or faux pas scores and any demographic or clinical characteristics. These include disease duration, times in hospital, duration of hospitalization, clozapine dosage (P > 0.05). There was no significant correlation between faux pas scores and any PANSS subscales (P > 0.05), but the correlation between the discrimination of eye gaze 0° and PANSS positive scale was significant (r = 0.579, P < 0.001).

Table 6.  Pearson correlations between characteristics of patient and social cues recognition
 Eye gaze 0°Faux pas recognition questionFaux pas related questions
  • ***

    Correlation is significant at the 0.001 level (two-tailed).

  • PANSS, Positive and Negative Syndrome Scale.

Disease duration0.099−0.052−0.044
Times in hospital0.220−0.212−0.203
Duration in hospital0.0100.0700.066
Clozapine dosage0.2280.0290.039
PANSS subscales
 Positive symptoms0.579***0.1600.172
 Negative symptoms−0.135−0.081−0.137
 Global psychopathology0.229−0.060−0.091

Correlation between social functioning and social cues recognition

Table 6 sets out the relationships between measures of social functioning and social cues recognition. The correlation between the faux pas tests and two of the SFS subscale scores (independence-performance and employment) was significant (faux pas recognition question r = 0.349, 0.365, P < 0.05; faux pas related questions r = 0.376, 0.377, P < 0.05), but the correlation between eye gaze discrimination at 0° and the SFS subscale scores was not (P > 0.05; Table 7).

Table 7.  Correlations between social functioning and social cues recognition
 Eye gaze 0°Faux pas recognition questionFaux pas related questions
  • *

    Correlation is significant at the 0.05 level (two-tailed).

  • SFS, Social Functioning Scale.

Social withdrawal−0.1360.1570.168
Social activities0.0450.0710.056
Recreational activities0.2510.0170.064
Independence (competence)0.1280.1050.133
Independence (performance)0.1360.349*0.376*
SFS total0.0980.2310.258


This study investigated the ability of patients with schizophrenia to recognize social cues. Special emphasis was placed on the relation between social cognition and social functioning. Both initial predictions were confirmed. As expected, patients with schizophrenia were impaired, relative to healthy control subjects, on both the tasks involving social cognition. A significant association between faux pas detection and the social functioning measures of SFS (independence and employment) was identified.

Eye gaze discrimination

The direction of one's gaze conveys a wealth of information in social interactions. This study has confirmed that the performance of Chinese patients with schizophrenia for gaze discrimination task was impaired. For all gaze directions, patients had lower scores than controls, but the only differences which were statistically significant were when the subject in the photograph was looking directly at the participant. The hypothesis one was confirmed by the current data.

The results of this study also seem to be consistent with data from Rosse et al.14 In their experiment, subjects were asked to report whether or not portraits presented on a screen were looking at them. They found that patients with schizophrenia presented significantly different performance profiles relative to control subjects, and produced different types of errors. In contrast, the current finding seems to be at odds with the work of Franck et al.,24 who found that subjects with schizophrenia were able to distinguish gazes in all directions successfully, but took significantly more time over the right versus left task than with the self versus non-self judgment.

The authors' contradictory finding, that patients with schizophrenia were impaired when the gaze was at 0°, may be due to the difference between the task that was used and that employed by Franck et al. In the current study, the selection of participants with schizophrenia excluded those suffering from significant psychiatric symptoms and taking high doses of antipsychotics at the time of the experiment. Alternatively, since the current subjects were required to determine the direction of gaze in terms of left or right, some of the instructions in this test may have placed greater demands on the semantic system than in other tests. The different head directions also upgraded the sensitivity.

There is convergent evidence that performance in determining gaze direction involves the superior temporal sulcus (STS) region of the brain, which has been implicated in the pathogenesis of schizophrenia.25,26 The authors' results imply that patients with schizophrenia are probably impaired in terms of the function not only of the prefrontal cortex but also the STS. The eye gaze discrimination task measures a person's ability to judge the mental state of another person solely from a photograph of the eyes. The authors, therefore, suggest that the deficits demonstrated by the lower scores achieved by patients with schizophrenia in this test have contributed to impairments in the subjects' ToM.

Faux pas recognition

A number of studies have shown that patients with autism fail the faux pas test,13 but the performance of patients with schizophrenia has not previously been reported. The authors' results clearly demonstrate that patients with schizophrenia are also impaired in the performance of faux pas task, including the failure to detect when something inappropriate has been said, and to accurately infer the story characters' belief states.

The same task has been used in a number of studies to measure ToM. Although normal 9 to 11-year-old children were skilled at detecting faux pas, children with Asperger's syndrome or high-functioning autism showed severe impairment in performance.13 Furthermore, the performance of individuals with bilateral amygdala damage was also significantly impaired.27 Gregory et al. found that patients with a frontal variant of frontotemporal dementia were also impaired on the faux pas test,28 which may explain some abnormalities in interpersonal behavior. As with Asperger's syndrome, patients with bilateral orbito-frontal lesion also demonstrated deficits on the faux pas task, despite their normal performance on other cognitive tests. It has also been reported that unilateral dorsolateral frontal lesion patients had no specific ToM deficits.23 The convergent evidence of neuropsychological studies strongly suggests that the orbito-frontal lesion may be implicated in the neural basis of the faux pas test.

Many pieces of evidence have indicated that patients with schizophrenia, even those of normal intelligence, were impaired in different tests of ToM.5,29 subjects with schizophrenia were impaired in understanding irony,30 they performed more poorly on hinting tasks compared to controls. A wide variety of theory of mind tasks, have been used to measure one or two mental state attribution abilities: intentionality judgments, epistemic mental state inferences, and affective mental state inferences in schizophrenia.5,8,11,31 The faux pas test, as a more advanced or comprehensive ToM test, requires an understanding of both inappropriate utterances and the emotional impact of a statement on the listener, measures all these types of mental states, can help to clarify the impairments of ToM in schizophrenia. The authors' data are, maybe, the first report showing impaired performance on faux pas detection in individuals with schizophrenia, the current results which are consistent with the results of related ToM tasks in another study,5,29,30 indicates that patients with schizophrenia are indeed impaired in these domains.

In addition to the hypothesis of dysfunction of the dorsolateral prefrontal cortex,32 several lines of evidence have recently suggested that the orbitofrontal cortex (OFC) may be dysfunctional in people with schizophrenia.33 The OFC dysfunction hypothesis is further supported by the impairment in faux pas detection shown in the current study.

Relation of social functioning and psychiatric characteristics in people with schizophrenia

The results of the current study suggested that patients with schizophrenia had certain deficits in social functioning. They had particular difficulties in each category from the SFS compared with normal control subjects. Concerning the correlations between the SFS and the PANSS scores, the significant association between each category of SFS and the PANSS negative symptoms, global psychopathology symptoms and total scores suggested that, similar to the negative symptoms, impaired social functioning was persistent in patients with schizophrenia, even in the periods of remission from psychotic episode.

Relation of social cognition and psychiatric characteristics in people with schizophrenia

There was no correlation between the patients' psychiatric characteristics (disease duration, number of hospitalizations, duration in hospital, and administration of antipsychotic medication) and their ability to recognize social cues, this is consistent with previous research,9,34 suggesting that these social cognition abilities are relatively independent of these particular therapeutic factors. Compared with the eye gaze discrimination, there was no correlation between the faux pas scores and each category from the PANSS, but the correlation between the eye gaze discrimination and PANSS positive symptoms was significant. The results suggested that the faux pas task is a stable index of social function but the eye gaze discrimination is susceptible to the positive symptoms.

Relation of social cues perception to social functioning in people with schizophrenia

Several studies have found that people with schizophrenia demonstrate significant deficits in social cues recognition and reduced quality of life. Performance in the two social cognition tasks outlined above revealed that the subjects with schizophrenia had difficulties with several aspects of inferring others' mental state. However, the relationship between social functioning and social cues recognition has hitherto been somewhat unclear. This study is, to the best of the authors' knowledge, the first to examine eye gaze discrimination and faux pas perception in relation to social functioning in patients with schizophrenia. It is noteworthy insofar as it demonstrates a significant correlation between impairments in performance of the faux pas test and the particular social aspects measured on the SFS (independence and employment) for a group of patients with schizophrenia. Performance on the eye gaze discrimination tasks was, however, independent of the social functioning scores on the SFS.

Correlation analysis of the patients' performance in the two social cues recognition tasks and the social function measure revealed a number of interesting findings. First, the faux pas task was strongly correlated with the independence and employment subscales of the SFS, implying that the task may be a useful adjunct to the clinical diagnostic process for schizophrenia. The ability to draw an inference about other people's affective mental state during a conversation is an important aspect of social communication and competence. It may, therefore, be that a deficit in the perception of faux pas in people with schizophrenia may result in lowered abilities in terms of social independence and employment.

Second, no correlation was found between social function and eye gaze discrimination, or inferring others' intention and attention focus by observing eye gaze, which in turn indicated that patients with schizophrenia had general social cognitive processing deficits, but faux pas recognition deficits led to more problems in social functioning. The two social cues recognition tasks may emerge from independent cognitive processes.

Furthermore, the findings of Vauth et al.'s study offer another useful approach.16 Their results supported the proposition that the direct impact of non-social cognition on vocational functioning is smaller than the impact of social cognition on work-related social skills. Penn et al. describe the potential implications of a social-cognitive model, differentiating social from non-social cognition. In schizophrenia, performance of social cognition may contribute greater variance to social functioning than does non-social cognition, perhaps enabling the mechanism of schizophrenia to be better understood within a social-cognitive framework.35

Finally, the current study has a number of potential limitations. First, the response time in the eye gaze discrimination task was not analyzed. Franck et al. found patients with schizophrenia took significantly more time to complete the task. A second limitation is that the social cues recognition task used here is not comprehensive. Patients with schizophrenia have deficits in other such tasks, such as the perception of facial affective expression, recognition of familiar social situations, visual scanning, and recognition of the intent of others. They also showed deficits in the complex judgments underlying performance in a gambling test and on a reversal learning test.33 The relationship between these tasks and social functions will be clarified in the authors' future research.


This work was supported by the Natural Science Foundation of China (030670706, 30370479), the National Basic Research Program of China (2005CB522800), and the Distinguished Youth Science Grants of Anhui Province (04043071, 2004kj192zd). This project was also sponsored by the University Development Fund administered by The University of Hong Kong.