Dysfunction of configural face processing in individuals at ultra-high risk for schizophrenia

Abstract

Introduction: Discrepancy in the ability to recognize faces for patients with schizophrenia is an important aspect of the impairment in the social cognition abilities of these individuals [1]. Many studies have revealed that adults who are 'experts' in face processing depend on configural processing, that is, the specific spatial relationships among face features [2]. Previous reports have suggested that specific problems with the configural processing of faces affect the face-recognition deficit in patients with schizophrenia [3]. However, little is known as to whether these deficits in face recognition are potential risk markers for schizophrenia. The aim of this study was to examine the discriminating ability in configural and featural processing of facial information involved in face recognition in individuals at ultra-high risk for schizophrenia. We hypothesized that configural processing of face recognition in subjects at ultra-high risk for schizophrenia as well as schizophrenia patients would demonstrate lower performance compared to normal controls. Methods: This study compared the performance of a face processing task in three groups: individuals at ultra-high risk for schizophrenia (n=20), patients with schizophrenia (n= 18) and normal controls (n = 20). Participants were required to decide whether the two pictures were the same or different in a facediscrimination task, using upright and inverted pairs of face photographs that differed in featural or configural information. They were shown faces consisted of 50 configural and 50 featural trials in each set of upright and inverted blocks and chairs comprised of 50 trials in each set ofupright and inverted blocks. The configural face task had the eyes modified by moving them five pixels in and out horizontally and mouths modified by moving them five pixels up and down vertically. The featural faces were changed by pasting in three different sets of eyes from a person of the same gender while maintaining the overall configuration. For a control task, the chair-discrimination task ofthe featural information were comprised of a pair of photographs of chairs, which were similar to each other in their general contour but differed in their armrests, back, and legs. Results: the individuals at ultra-high risk for schizophrenia showed poor performance in configural processing of faces but not featural processing compared with the normal controls and showed lower performance in discriminating configural tasks compared to featural tasks related to the face. The individuals at ultra-high risk for schizophrenia as well as the patients with schizophrenia were significantly lower inversion effect than the normal controls in the configural faces, whereas there were no differences among the three groups in the featural faces and chairs. Conclusions: This study suggests that the specific dysfunction in configural face processing, which has an impact on face recognition, may be at increased risk of developing psychosis. These results suggest that the deficits of social perception such as face recognition in the ultra-high risk for schizophrenia may be presented before the pre-psychotic phase and worsened during the prodromal stages. It may be a vulnerable marker that can be to impact deficits of face recognition in patients with schizophrenia.