Monitoring Accuracy Across Domains of General Knowledge and Emotional Face Recognition

Confidence and its accuracy have been most commonly examined in domains such as general knowledge and learning, and have been rarely studied in social domains. The current study extends the study of metacognitive monitoring accuracy in an academic context, to metacognitive monitoring accuracy in a social context. Monitoring accuracy indices, both calibration and discrimination, were examined for emotional face recognition in 136 university students through an experimental task paradigm developed by Kelly and Metcalfe (2011). To examine whether monitoring accuracy is stable within individuals and across tasks that represent diverse areas, metacognitive monitoring for emotional face recognition was compared to metacognitive monitoring for general knowledge. In addition, correlations between monitoring accuracy and cognitive abilities (intellectual ability and working memory), several aggregated judgments regarding each task as a whole (ratings of predicted and postdictive performance, difficulty, and effort required), as well as self-perceptions relating to social anxiety, empathic ability, differentiation of emotions, and emotion regulation skills were explored. Calibration, but not resolution, was positively correlated across tasks, reflecting a person-centered trait. Cognitive abilities were predictive of monitoring accuracy across tasks, while other task-specific judgments and self-perception variables demonstrated specific associations with monitoring accuracy for emotional face recognition. Monitoring accuracy for emotional face recognition was not predictive of self-reported social-emotional challenges. Overall, study findings support that calibration and resolution are separate indicators of monitoring accuracy, both relevant for understanding metacognitive monitoring in a social domain.