Inferring emotions from the eyes in typical development and major depression

Abstract

The present work investigates factors shaping how we read the language of the eyes, focusing on the impact of gender, experience, and the specificity of mental conditions such as MDD. Two paradigms were implemented in this study: The Emotion recognition in Masked Faces (EMF) task and RMET, which provide a comparable amount of information from the eyes. While most previous studies suffered such limitations of online experimenting as a lack of control over settings and female predominance, this work was conducted face-to-face. The outcome of the first study that included a sample of TD individuals balanced in gender, indicates that both males and females demonstrate an uneven profile of emotion recognition, with some emotions affected by masks more severely than others. Contrary to our expectations, however, males were more sensitive to subtle emotional signals transmitted through the eyes. On disgust recognition, the forgotten emotion of psychiatry, a notable effect of masks was found on females exhibiting lower accuracy. These findings are particularly relevant for research on mental disorders such as MDD, which shows gender disparities in prevalence, manifestation, and social cognition. The second study reveals that recognition of subtle emotions, in particular sadness, improves with experience through early to mid-adulthood in TD females. Finally, the third study in female MDD indicates that reading language of the eyes is affected selectively rather than globally: Recognition of subtle negative emotions (such as sadness and disgust) covered by masks as well as interpretation of the eyes’ expressions with positive valence are more heavily compromised. Selective impairments in emotion recognition differ markedly from global deficits in other clinical populations such as SZ. In both MDD and TD individuals, emotion recognition behind a mask and performance on the RMET are linked to each other in processing speed, but not recognition accuracy, suggesting a commonality in the encoding of visual input. Overall, by identifying gender- and experience-related patterns, this work suggests the ability for reading language of the eyes may serve as a marker of mental conditions such as MDD. Moreover, distinct patterns of deficits in reading language of the eyes across mental disorders enhances understanding of nonverbal social cognition, informing the development of diagnostic tools, remediation programs, and tailored interventions within and beyond the context of the COVID-19 pandemic.