Gaze-Based Self-Confidence Estimation on Multiple-Choice Questions and Its Feedback

Abstract

Self-confidence - an assurance of one's personal decision and ability - is one of the most important factors in learning. When there is a gap between a learner's confidence and comprehension, the learner loses a chance to review a learning subject correctly. To solve this problem, we propose a system which estimates self-confidence while solving multiple-choice questions by eye tracking and gives feedback about which question should be reviewed carefully. The system was evaluated in our experiment involving 20 participants. We observed that correct answer rates of questions were increased by 14% and 17% by giving feedback about correct answers without confidence and incorrect answers with confidence, respectively.