Metacognitive Experience Modeling Using Eye-Tracking

Metacognitive experience (ME) is one of the key facets of metacognition, which serves a critical cuing function in the process of self-regulated learning process. However, the study of ME is hindered by its subjective and implicit nature of and the challenges that are associated with accessing such experiences. In exploring such experiences, eye-tracking offers certain advantages over self-reporting methods. However, to date most studies tend to focus on utilizing eye-tracking to explore metacognitive skills (MS) rather than ME, with those that do explore ME also tending to require participants to self-report rather than relying on observation of possible behavioural indicators of such metacognitive processes. Based on previous works in this field, the research proposed is based on the hypothesis that eye-tracking data can provide a crucial objective measure of learners' implicit ME processes. The research will also investigate the extent to which such data can serve as the basis for automatically predicting the occurrence and intensity of ME during learning using machine learning, in a way that can support the delivery of adaptive domain-independent feedback in a variety of Intelligent Learning Environments (ILEs).