Dynamic integration of information about salience and value for smooth pursuit eye movements

Abstract

Eye movement behavior can be determined by bottom-up factors like visual salience and by top-down factors like expected value. These different types of signals have to be combined for the control of eye movements. In this study we investigated how smooth pursuit eye movements integrate salience and value information. Observers were asked to track a random-dot kinematogram containing two coherent motion directions. To manipulate salience, the coherence or the density of one of the motion signals was varied. To manipulate value, observers won or lost money in a separate experiment if they were tracking one or the other motion direction. Our results show that pursuit direction was initially determined only by salience. 300-400 ms after target motion onset, pursuit steered towards the rewarded direction and the salience effects disappeared. The time course of this effect depended crucially on the difficulty to segment the two signal directions. These results indicate that salience determines early pursuit responses in the same way as saccades with short latencies. Value information is processed slower and dominates pursuit after several 100 ms.

Reinforcement effects in anticipatory smooth eye movements

“…Nonetheless, our results demonstrate overall that aSPEMs are sensitive to the rewarding value of a stimulus even when the reward is related to the nonvisually guided phase of pursuit rather than the visually guided phase (Brielmann & Spering, 2015;Joshua & Lisberger, 2012;Schütz et al, 2015) and that reinforcement contingencies have to be taken into account in addition to attentional factors (Khurana & Kowler, 1987) and prediction (G. R. Barnes, Barnes, & Chakraborti, 2000; G. R. Barnes & Donelan, 1999;Santos & Kowler, 2017;Wells & Barnes, 1998) for the buildup of prior expectation about a target's direction.…”

“…In newborn infants (1-7 days), Darcheville, Madelain, Buquet, Charlier, and Miossec (1999) found that smooth eye movements can be enhanced using an auditory reinforcer contingent to the eye velocity. Schütz, Lossin, and Gegenfurtner (2015) showed that, in presence of two moving textures, pursuit direction choice is mainly determined by the target's visual salience at initiation, but 300-400 ms after motion onset, pursuit steered toward the rewarded direction, and the salience effects gradually disappeared. On the other hand, Joshua and Lisberger (2012) reported an influence of precued reward contingencies already on vector average pursuit initiation in monkeys.…”

Theoretical perspectives on active sensing

“…Active and preplanned strategies are not mutually exclusive in that actual patterns of sensor movements seem to be influenced by some mixture of them. For example, eye movements have been found to be best predicted by a combination of several bottom-up (such as saliency) and top-down (such as reward) factors [34, 35, 36] where the contribution of these different factors can depend on the timing requirement of the task and the time course and type of eye movement [37, 38, 39, 40, 41].…”

Smooth Pursuit Eye Movement of Monkeys Naive to Laboratory Setups With Pictures and Artificial Stimuli

“…The pursuit system does not depend solely on low level sensory features. For example, the size of the reward can determine the gain of the pursuit ( Joshua and Lisberger, 2012 ; Brielmann and Spering, 2015 ; Schutz et al, 2015 ). Therefore, we expected that objects that were significant to the monkeys before training would potentiate pursuit.…”