Face recognition in schizophrenia: do individual and average ROCs tell the same story?

INTRODUCTION: Many studies have shown that recollection process is impaired in patients with schizophrenia, whereas familiarity is generally spared. However, in these studies, the Receiver Operating Characteristic (ROC) presented is average ROC likely to mask individual differences. METHODS: In the present study using a face-recognition task, we computed the individual ROC of patients with schizophrenia and control participants. Each group was divided into two subgroups on the basis of the type of recognition processes implemented: recognition based on familiarity only and recognition based on familiarity and recollection. RESULTS: The recognition performance of the schizophrenia patients was below that of the control participants only when recognition was based solely on familiarity. For the familiarity-alone patients, the score obtained on the Scale for the Assessment of Positive Symptoms (SAPS) was correlated with the variance of the old-face familiarity. For the familiarity-recollection patients, the score obtained on the Scale for the Assessment of Negative Symptoms (SANS) was correlated with the decision criterion and with the old-face recollection probability. CONCLUSIONS: These results show that one cannot ascribe the impaired recognition observed in patients with schizophrenia to a recollection deficit alone. These results show that individual ROC can be used to distinguish between subtypes of schizophrenia and could serve as a basis for setting up specific cognitive remediation therapy for individuals with schizophrenia.

Impact of intention on the ERP correlates of face recognition.

The present study investigated the impact of study-test similarity on face recognition by manipulating, in the same experiment, the expression change (same vs. different) and the task-processing context (inclusion vs. exclusion instructions) as within-subject variables. Consistent with the dual-process framework, the present results showed that participants performed better on the inclusion task than on the exclusion task, with no response bias. A mid-frontal FN400 old/new effect and a parietal old/new effect were found in both tasks. However, modulations of the ERP old/new effects generated by the expression change on recognized faces differed across tasks. The modulations of the ERP old/new effects were proportional to the degree of matching between the study face and the recognition face in the inclusion task, but not in the exclusion task. The observed modulation of the FN400 old/new effect by the task instructions when familiarity and conceptual priming were kept constant indicates that these early ERP correlates of recognition depend on voluntary task-related control. The present results question the idea that FN400 reflects implicit memory processes such as conceptual priming and show that the extent to which the FN400 discriminates between conditions depends on the retrieval orientation at test. They are discussed in relation to recent controversies about the ERP correlates of familiarity in face recognition. This study suggests that while both conceptual and perceptual information can contribute to the familiarity signal reflected by the FN400 effect, their relative contributions vary with the task demands.

Mismatched expressions decrease face recognition and corresponding ERP old/new effects in schizophrenia.

OBJECTIVE: The objective was to investigate the electrophysiological (ERP) correlates of mismatched expression on face recognition in schizophrenia. METHOD: Expression-change effects and associated ERPs were explored in patients with schizophrenia (n = 20) and paired comparison participants (n = 20) on a long-term face-recognition task. RESULTS: A facial-expression change decreased discriminability for patients with schizophrenia than for healthy participants. The patients' recognition deficit was accompanied by the absence of the midfrontal FN400 and late parietal ERP old/new effects in the mismatched-expression condition. By contrast, preserved midfrontal FN400 and late parietal ERP old/new effects were found in both groups in the unchanged-expression condition. Thus, the preserved parietal old/new effect previously observed in schizophrenia was no longer found here in the situation in which expression changes took place between the study and recognition phases. CONCLUSIONS: These findings suggest that, when they are not supposed to take the change of expression into account, the recognition deficit observed here in patients with schizophrenia resulted from an impairment in the mechanisms underlying the emergence, assessment, or utilization of familiarity--as indexed by the ERP old/new effects. In these natural conditions, the impact of the expression change on the implementation of retrieval processes offers new insight into schizophrenia-linked deficits in face recognition, with substantial phenomenological differences with respect to the emergence of familiarity.

ERP investigation of study-test background mismatch during face recognition in schizophrenia.

Old/new effects on event-related potentials (ERP) were explored in 20 patients with schizophrenia and 20 paired comparison subjects during unfamiliar face recognition. Extrinsic perceptual changes - which influence the overall familiarity of an item while retaining face-intrinsic features for use in structural face encoding - were manipulated between the study phase and the test. The question raised here concerns whether these perceptual incongruities would have a different effect on the sense of familiarity and the corresponding behavioral and ERP measures in the two groups. The results showed that schizophrenia patients were more inclined to consider old faces shown against a new background as distractors. This drop in face familiarity was accompanied by the disappearance of ERP old/new effects in this condition, i.e., FN400 and parietal old/new effects. Indeed, while ERP old/new recognition effects were found in both groups when the picture of the face was physically identical to the one presented for study, the ERP correlates of recognition disappeared among patients when the background behind the face was different. This difficulty in disregarding a background change suggests that recognition among patients with schizophrenia is based on a global perceptual matching strategy rather than on the extraction of configural information from the face. The correlations observed between FN400 amplitude, the rejection of faces with a different background, and the reality-distortion scores support the idea that the recognition deficit found in schizophrenia results from early anomalies that are carried over onto the parietal ERP old/new effect. Face-extrinsic perceptual variations provide an opportune situation for gaining insight into the social difficulties that patients encounter throughout their lives.

Impairment not only in remembering but also in knowing previously seen faces and words in schizophrenia.

Patients with schizophrenia have pronounced deficits in face recognition memory that severely hamper their social skills. The functional mechanisms of these impairments remain unknown. According to the dual-process theory, recognition memory comprises two distinct components: recollection and familiarity. Studies using the Remember/Know procedure in patients with schizophrenia showed impairments in conscious recollection as measured by remember responses, but not in familiarity as measured by know responses. Unfortunately, none of these studies used face material. We investigated both recognition memory components using words and faces and the 'Remember/Know' procedure in 25 patients with schizophrenia and 24 control participants. In the same task, size congruency of stimuli was manipulated between the study and test phases to have a selective impact on know responses for faces. Patients reported fewer remember responses than controls. Size changes between the study and the test affected know responses in controls but not in patients. These results reveal that patients with schizophrenia are impaired in terms of their ability to recollect details about previously seen faces as they are for words.

Use of the process dissociation procedure to study the contextual effects on face recognition in schizophrenia: familiarity, associative recollection and discriminative recollection.

Contextual effects were explored in schizophrenia patients and paired comparison subjects during a long-term face recognition task. The objective was to investigate the contextual effects on face recognition by manipulating, in the same experiment, the perceptual context of the face (intrinsic vs. extrinsic) and the task context (inclusion vs. exclusion instructions). The situation was derived from the Jacoby's [Jacoby, L.L., 1991. A process dissociation framework: separating automatic from intentional uses of memory. Journal of Memory and Language 30, 513-541] process dissociation procedure. The results showed that schizophrenia patients (N=20) presented lower performances than healthy controls (N=20) in the inclusion but not in the exclusion task. This observation emphasizes the heterogeneity of recollection and suggests that the memory impairment in schizophrenia reflects an imbalance between two mechanisms. The first is a deficit in "associative recollection", i.e., the failure to use efficiently associative information. The other is an enhanced "discriminative recollection" that impedes their capacity to process information separately from its perceptual context. In addition, correlation with symptoms suggest that the former is expressed in the loosening of associations characteristic of disorganization symptoms, whereas the latter reflects the lack of flexibility or the contextualization bias related to psychotic symptoms, i.e., delusions and hallucinations.

Processing emotional expression and facial identity in schizophrenia.

Previous studies showed that schizophrenic patients have a deficit in facial information processing. The purpose of the present study was to test the abilities of patients with schizophrenia and normal controls in emotion and identity matching when these two dimensions were varied orthogonally. Subjects (20 schizophrenic patients and 20 controls) had to report if two faces had the same emotion or belonged to the same person. When the task concerned one type of information (i.e. emotion or identity), the other one was either constant (same person or same emotion) or changed (different person or different emotion). Schizophrenic patients performed worse than controls for both kinds of facial information. Their deficit was more important when the secondary factor was changed. In particular, they performed at chance level when they had to match one emotion expressed by two distinct persons. Finally, correlation analysis indicated that performance/deficit in identity and emotion matching co-varied and that in such tasks performance is negatively correlated with the severity of negative symptoms in patients. Schizophrenic patients present a generalised deficit for accessing facial information. A facial emotion and an identity-processing deficit are related to negative symptoms. Implications for face-recognition models are discussed.

Effects of emotion and identity on facial affect processing in schizophrenia.

We used facial affect labeling and matching tasks to study effects of (1) emotion and (2) identity on facial affect processing in patients with remitted schizophrenia (n=30) compared with healthy controls (n=30). The patients (1) had a specific deficit for labeling facial affects of sadness and anger but not happiness, disgust and fear; they (2) performed as well as controls in matching facial affects in one face but were impaired in matching facial affects in two different faces. The patients' impairment in facial affect processing may be emotion-specific. The effects of identity on facial affect processing are discussed in the light of several hypotheses (a deficit of context processing, a global-local processing impairment or a selective attention deficit), and may be related to frontal, prefrontal or amygdala dysfunctions.

Electrophysiological study of contextual variations in a short-term face recognition task.

Event-related potentials (ERPs) were recorded during two short-term recognition tasks using unfamiliar faces. These experiments are based on the process dissociation procedure (PDP), whereby the exclusion criterion was an intrinsic context or extrinsic context, the facial expression (Experiment 1) or background (Experiment 2), respectively. The results indicate that retrieval orientation, in addition to extensive strategic control, affects both the frontal (N250) and temporoparietal (P3b) components. Furthermore, these data indicate that an early frontal modulation interacts between processing that bears on the face (interactive intrinsic context) and processing that bears on two objects at the same time (interactive extrinsic context), in which, in the latter case, that the background change led to an early modulation at the frontal sites in the left hemisphere. These results are consistent with the idea that frontal effects reflect differences in the nature of the information during retrieval and postretrieval processes involved. Furthermore, that the left posterior repetition effect appears to be a manifestation of the retrieval of relevant contextual information that perturbs the recognition decision, whereas the right posterior repetition effect reflects to be the outcome of the retrieval of the face as a whole. Finally, results are in concordance with the hypothesis that the difference during recognition with or without source memory may be in the strength of the relationship between the target and the contextual information to be retrieved. In essence, that automatic and controlled processes in a given context depends on both task-related and target-related constraints.

Symmetry, averageness, and feature size in the facial attractiveness of women.

Female facial attractiveness was investigated by comparing the ratings made by male judges with the metric characteristics of female faces. Three kinds of facial characteristics were considered: facial symmetry, averageness, and size of individual features. The results suggested that female face attractiveness is greater when the face is symmetrical, is close to the average, and has certain features (e.g., large eyes, prominent cheekbones, thick lips, thin eyebrows, and a small nose and chin). Nevertheless, the detrimental effect of asymmetry appears to result solely from the fact that an asymmetrical face is a face that deviates from the norm. In addition, a factor analysis indicated that averageness best accounts for female attractiveness, but certain specific features can also be enhancing.

Context effects in recognition memory: the role of familiarity and recollection.

A variant of the process dissociation procedure was coupled with a manipulation of response signal lag to assess whether manipulations of context affect one or both of the familiarity and search processes described by the dual process model of recognition. Participants studied a list of word pairs (context+target) followed by a recognition test with target words presented in the same or different context, and in the same or different form as study (singular/plural). Participants were asked to recognize any target word regardless of changes to form (inclusion), or to only recognise words that were presented in the same form (exclusion). The standard context reinstatement effect was evident even at the short response lags. Analyses of the estimates of the contributions of familiarity and search processes suggest that the context effect demonstrated here can be attributed in part to the influence of familiarity on recognition, whereas the effect on recollection was less clear.

Gaze direction determination in schizophrenia.

It has been proposed that an impairment in gaze determination is responsible for the paranoid symptoms reported in schizophrenia. To address this, we examined the gaze discrimination system in schizophrenia. Thirty-two patients suffering from schizophrenia (20 patients with persecutory delusions and 12 patients without such delusions) were compared to 32 control subjects on two specific tasks. In the first task, the subjects had to determine whether 130 portraits were looking right or left. In the second task the subjects were asked to determine whether or not 130 portraits were looking at them. The absolute threshold of difference used to investigate the influence of instruction on gaze discrimination did not show any difference between patients with schizophrenia, whatever paranoid or not, and control subjects. Paranoid patients, as well as controls, displayed a significantly finer discrimination threshold in the right vs. left judgment than in the self vs. non-self judgment. Subjects with schizophrenia were able to discriminate gaze direction in the two tasks, but they took significantly more time in the task requiring to determine the presence or the absence of a mutual gaze contact than in the other one, whereas controls took the same duration to elicit both tasks. These data are consistent with those reporting that perceptual abilities are spared in schizophrenia while delusions are related to an impairment of a higher level of analysis.

Gender is a dimension of face recognition.

In an experiment, the authors investigated the impact of gender categorization on face recognition. Participants were familiarized with composite androgynous faces labeled with either a woman's first name (Mary) or a man's first name (John). The results indicated that participants more quickly eliminated faces of the opposite gender than faces of the same gender than the face they were looking for. This gender effect did not result from greater similarity between faces of the same gender. Rather, early gender categorization of a face during face recognition appears to speed up the comparison process between the perceptual input and the facial representation. Implications for face recognition models are discussed.

Selective attention to facial emotion and identity in schizophrenia.

The selective attention to facial emotion and identity was investigated in 12 patients with schizophrenia and 12 healthy participants. Both patients and controls were required to perform two classification tasks (according either to identity or emotion). Two separate values for identity (person A/person B) and for emotion (fear/anger) were used. When the classification task was on one dimension, the other dimension was either correlated, constant, or orthogonal (Garner WR. The Processing of Information and Structure. Potomac, MD: Erlbaum, 1974, Garner WR. Interaction of stimulus dimensions in concept and choice processes. Cognitive Psychology 1976;8:98-123). Results indicated that both patients and healthy participants had an asymmetrical pattern of performance: they were able to selectively attend to the identity of the face presented, regardless of the emotion expressed on the face, but variation in identity interfered with the classification of facial emotion. Moreover, a correlational study indicated that the identity interference on emotion classification for schizophrenic patients covaried with the severity of their negative symptoms. The selective attention competencies in schizophrenia and the independence hypothesis of emotion and face recognition are discussed in the framework of current face recognition models.