Visualizing mental representations in schizophrenia patients: A reverse correlation approach

• Main Authors: Loek Brinkman, Ron Dotsch, Jelmer Zondergeld, Martijn G.J.C. Koevoets, Henk Aarts, Neeltje E.M. van Haren
• Format: Article
• Language: English
• Published: Elsevier 2019-09-01
• Series: Schizophrenia Research: Cognition
• Online Access: http://www.sciencedirect.com/science/article/pii/S2215001318300593

Schizophrenia patients have difficulties recognizing emotional states from faces, in particular those with negative valence, with severe consequences for daily life. What do these patients see in their minds eye, when they think of a face expressing a particular emotion or trait? The content of such mental representations can shed light into the nature of their deficits, but are usually inaccessible. For the first time, we explored the applicability of reverse correlation, which has been successfully used to visualize mental representations in healthy populations, to visualize mental representations in schizophrenia patients.We investigated mental representations of trustworthy faces, a primary dimension of social face evaluation that is highly correlated with valence. Patients (n = 23) and healthy controls (n = 34) classified images of noise-distorted faces as ‘trustworthy’, ‘untrustworthy’ or ‘neutral’. We visualized their mental representations of these concepts by averaging the noise patterns based on their classifications. These visualizations were then rated on trustworthiness by an independent sample of participants.Patients were able to perform the reverse correlation task, with response times and biases similar to those of healthy controls, and the obtained images vividly reflected the respective constructs of interest. However, there were no significant differences between the ratings of the visualizations of patients and controls.Conclusion: These novel findings provide a proof of principle that the reverse correlation technique can be applied to investigate mental representations in schizophrenia patients. Keywords: Emotion recognition, Schizophrenia, Mental representations, Psychophysics, Data-driven