Proceedings of the Annual Meeting of the Cognitive Science Society

Emotional experiences have a significant impact on learningabout complex topics. Yet, challenges exist becauseemotions are typically operationalized as end products,excluding if, how, and when emotions change duringlearning and their relation to metacognition and performancewith advanced learning technologies such as intelligenttutoring systems (ITSs). In this paper, we addressed thesechallenges by capturing and analyzing 117 college students’concurrent and self-reported emotions at 3 time points duringlearning with MetaTutor, an ITS. Analyses revealed negativerelationships between increases in boredom, metacognitivemonitoring accuracy, and performance. We also foundthat if confusion persisted over time during learning, itwas detrimental to performance. These findings provideimplications for designing affect-sensitive ITSs which fosteremotion-regulation and metacognitive monitoring based onchanges in emotions during learning to optimize performance.