確率モデルおよび空間統計学を用いた注視点の分析

抄録

This study examined the availability of probabilistic modeling and spatial statistics for analyzing eye movement data. The data from Kamiya, Sakagami, Masuda, & Omori (2006), which showed that eye movement speed (EMS) can be controlled by operant conditioning, were analyzed. In Examination 1, the best fitting model to the EMS distributions was selected based on Akaike's information criterion, the Bayesian information criterion, and residual analysis; next fixation criterion speeds in each participant and condition were calculated from the model. In Examination 2, distributions of the fixation point defined by the above criterion were analyzed by spatial statistics. The results evinced the following: (1) the model consists of one exponential distribution as fixation and two normal distributions as saccade, which best describes the EMS distributions; (2) the fixation criterion speeds were approximately 16 deg/s; (3) there was no adequate duration to define the fixation; and (4) fixation points were more clustered under the condition with high EMS. These results demonstrated the merit of using the above two methods.