Related Experiment Video
Updated: Sep 2, 2026

Measuring Attention and Visual Processing Speed by Model-based Analysis of Temporal-order Judgments
Published on: January 23, 2017
Estimation method of attentional object using spatial frequency selectivity in pupillary response
Yuhong Lv1, Rumi Hisakata1, Hirohiko Kaneko1
1Department of Information and Communication Engineering, Institute of Science Tokyo, Tokyo, Japan.
Abstract:
This study investigates pupillary dynamics associated with object-based visual attention using two overlaid filtered icon images (hybrid stimulus). Subjects were presented with the hybrid stimulus and instructed to maintain their attention on one of the icon images while the spatial frequency (HSF: High spatial frequency or LSF: Low spatial frequency) components of the two images alternated periodically, where they had different components from each other. Pupil diameter was continuously recorded during the alternations. The combination of motion conditions (static or dynamic) of the icon images in the hybrid stimulus was manipulated. Results revealed that pupil size was relatively small when the spatial frequency component of the attentional icon image was HSF and tended to constrict as the spatial frequency component of the attentional icon image changed from LSF to HSF. In contrast, pupil size was relatively large when the spatial frequency component of the attentional icon image was LSF and tended to dilate when the icon image changed from HSF to LSF. These pupil size changes were particularly prominent under the combination of two dynamic or one dynamic and one static icon images. Based on these findings, a method to infer the attended icon image using pupil size change was proposed. The accuracy of the inference method exceeded 70% in specific conditions, suggesting its feasibility as an attention-based information input system for users with limited motor capabilities.

