Saccade selection is driven by physiologically measurable costs
Wed—HZ_7—Talks8—7601
Presented by: Christoph Strauch
Selecting the next saccade target is one of the most frequent decisions humans make, shaping perception and influencing broader cognitive processes. Current models of saccade selection emphasize three key drivers: the observer’s goals, selection history, and the physical salience of stimuli. Recent advances, however, highlight the importance of an additional factor: the intrinsic cost of shifting attention. While theoretically compelling, these costs could not be quantified physiologically.
I will share recent findings demonstrating that intrinsic eye-movement costs can be physiologically measured using pupil size – an established indicator of effort. After mapping saccade costs across directions, we examined saccade preferences using a free-choice paradigm across identical saccade targets. Our results show that participants preferred less costly directions over more effortful ones. Importantly, pupil-inferred costs also predicted saccade preferences when the saliency of potential targets was experimentally manipulated. In these cases, both saliency and cost jointly influenced saccade selection. Finally, we extended this investigation to a visual search task in naturalistic scenes. Here, we observed that cost-driven preferences remained robust, underscoring the significant and generalizable role of cost in saccade selection. These findings suggest that incorporating intrinsic saccade costs into attentional models is critical for a comprehensive understanding of saccade behavior.
I will share recent findings demonstrating that intrinsic eye-movement costs can be physiologically measured using pupil size – an established indicator of effort. After mapping saccade costs across directions, we examined saccade preferences using a free-choice paradigm across identical saccade targets. Our results show that participants preferred less costly directions over more effortful ones. Importantly, pupil-inferred costs also predicted saccade preferences when the saliency of potential targets was experimentally manipulated. In these cases, both saliency and cost jointly influenced saccade selection. Finally, we extended this investigation to a visual search task in naturalistic scenes. Here, we observed that cost-driven preferences remained robust, underscoring the significant and generalizable role of cost in saccade selection. These findings suggest that incorporating intrinsic saccade costs into attentional models is critical for a comprehensive understanding of saccade behavior.
Keywords: Visual attention, eye movements, attentional selection, pupillometry