Mental health stigma and emotional face processing: An ERP study

This study takes the first step toward developing an event-related potential (ERP) measure of bias toward people with mental illnesses. We recorded participants' EEGs while they viewed happy and angry faces; they were told half of the people in the photos were mentally healthy while the other half had a diagnosis of schizophrenia, which was indicated by the color of the frame around each face. Afterward, participants rated how dangerous each face looked to them. Participants rated faces with the schizophrenia designation as more dangerous than those designated as mentally healthy. There were significant interactions between facial expression and mental health label for both the early N170 and later P3 amplitudes. Angry expressions elicited larger N170s than happy ones, but only for faces with the schizophrenia label. Thus, angry faces with the schizophrenia designation appeared to be particularly salient and automatically grabbed our participants' attention, at least initially. In contrast, happy faces with the schizophrenia label elicited larger P3s than those with the mentally healthy designation. This finding suggests that later attentional processes were recruited more when someone with a diagnosis of schizophrenia looked happy, perhaps indicating that participants found this surprising.