The role of configural information in facial emotion recognition in schizophrenia

Abstract

The schizophrenia deficit in facial emotion recognition could be accounted for by a deficit in processing the configural information of the face. The present experiment was designed to further test this hypothesis by studying the face-inversion effect in a facial emotion recognition task. The ability of 26 schizophrenic patients and 26 control participants to recognize facial emotions on upright and upside-down faces was assessed. Participants were told to state whether faces expressed one of six possible emotions (happiness, anger, disgust, fear, sadness, neutrality) in two sessions, one with upright faces and the other with upside-down faces. Discriminability and the decision criterion were computed. The results indicated that the schizophrenic patients were impaired in upright facial emotion discrimination by comparison with the controls. They also exhibited an inversion effect similar to the controls. However, whereas controls tended to adopt a more conservative criterion for all emotions and a liberal criterion for neutrality when the faces were upside-down, schizophrenic patients presented a decision criterion pattern that was similar for the two orientations and similar to controls in upside-down emotion recognition. The lack of a decision criterion shift was associated with positive symptoms such as delusions, hallucinations, and bizarre behavior. Moreover, positive and negative symptoms were associated with inversion effect on discriminability; the more severe the symptoms, the weaker the inversion effect. We conclude that individuals with schizophrenia do process the configural information of the face. However, further investigations are needed to assert whether this information is of good quality in schizophrenia.

What's in a face? Effects of stimulus duration and inversion on face processing in schizophrenia

“…One prior study showed an intact face inversion effect in patients, but a face memory paradigm was used and spacing and part processing were not manipulated (Schwartz et al, 2002). Two studies looking at emotion processing showed that both patients and controls had similar inversion effects (Chambon et al, 2006;Schwartz et al, 2002), consistent with the present findings, and again suggestive of relatively intact intrinsic cortical processing. As stated above, one face processing study failed to find an inversion effect in patients in a spacing task, but upright performance was below chance (Shin et al, in press).…”

Using signal detection theory to test the impact of negative emotion on sub-clinical paranoia

“…This can be considered as a conservative 'attentional bias'. Moreover, Chambon, Baudouin, and Franck (2006) discovered that patients with schizophrenia adopted a more conservative response bias in a facial emotion detection task than control subjects. Similarly, Combs, Michael, and Penn (2006) reported lower anger identification scores the more the subgroup was prone to paranoia.…”

Facial emotion space in schizophrenia

“…The test materials were adapted from Chambon, Baudouin, and Franck (2006; see also Durand, Gallay, Seigneuric, Robichon, & Baudouin, 2007) and consisted of 36 coloured photographs of faces expressing six different emotions (happiness, fear, sadness, anger, disgust, and neutrality), with six photographs for each category. All information about background and body was eliminated.…”