Prediction of recall accuracy in a contextual understanding task using eye movement features

Contextual understanding, which consists of memory, reasoning and recall, is a key process of human-computer interactions and interfaces. To determine the possibility of predicting the recall accuracy of reading and memorizing tasks using features of eye movements for a targeted text, a contextual understanding task experiment was conducted. The relationship between eye movement during memorization and recall performance was hypothesized. Viewer's eye-movements during the reading of definition statements were observed, and twelve sector features of these eye-movements across a two-dimensional space were measured. The prediction procedure was developed using Support Vector Regressions with Gaussian kernel. The correlational relationship between predicted and experimental recall accuracies was significant; therefore the hypothesis was supported. Also, it was found that prediction performance depended on a combination of features of eye movements.