Error Observation in Schizophrenia

Abstract

Despite the pervasive and impairing nature of social difficulties in schizophrenia, the causes of these problems are not fully understood. It has been suggested that problems with cognitive functioning contribute to the social deficits of schizophrenia. However, little is known about the neural mechanisms that underlie cognitive processes directly linked to social dysfunction in schizophrenia. Recent studies of the mirror neuron system have focused on the error-related negativity (ERN), a negatively-deflected event-related brain potential that is elicited following the commission of an erroneous response. This study examined ERN activity in schizophrenia patients and psychiatrically healthy controls during performance and observation of a confederate performing a computerized flanker task. The lateralized readiness potential (LRP) allowed for a direct comparison of brain activation reflecting response readiness verses error signaling. Correlations between ERN activity during flanker observation, social cognition (i.e., theory of mind), and community social functioning were explored. Finally, correlations between verbal memory, executive functioning, and social functioning were examined and social cognition was explored as a mediator between neurocognition and social functioning. Results indicated that controls produced a robust ERN during execution of the flanker task, whereas ERN activity among patients was comparatively attenuated in amplitude. During observation, there were no significant group differences and no identifiable observation ERN; however, there was greater negative activity following error than correct trials in this condition for all participants. LRP activity did not parallel that of the ERN, supporting the differentiation of motor activity and error-related processing during observation. The only significant correlation to emerge between ERN activity and social cognition and social functioning was between occupational status and execution ERN activity among controls only. Unexpectedly, neurocognition and social functioning were negatively correlated in the patient group. Expectedly, these variables were positively correlated among controls. Therefore, regression analyses were conducted separately by group; however, neither neurocognition nor social cognition predicted a significant proportion of the variance in social functioning. Despite limitations, this research is discussed as a starting point for integrating the study of psychophysiological activity with social behavior and functioning, particularly in a clinical population with pronounced social deficits.