Purchase this article with an account.
Andrew Leber, Bo-Yeong Won; Spatial reward guides choice, not visual search. Journal of Vision 2016;16(12):1139. doi: 10.1167/16.12.1139.
Download citation file:
© 2017 Association for Research in Vision and Ophthalmology.
The influence of reward on human behavior is pervasive, extending to virtually all aspects of cognition. Much recent research has focused on how reward influences attention, producing consistent demonstrations that valuable stimuli command greater attention. However, one conspicuous exception has been spatial location; only a few studies have reported attentional prioritization of valuable spatial locations, while several others have failed to find any prioritization. These failures suggest that visual search is largely insensitive to spatial value. Here, we provide more rigorous tests of this possibility, comparing spatial reward learning during visual search to a novel "visual choice" task. During Visual Search (Experiment 1), participants searched for a target T among 15 Ls, responding with a speeded mouse click on the target. Targets in a pre-designated "high value" quadrant yielded 20 points on 75% of trials and 1 point on the remaining trials; targets in the other "low value" quadrants yielded the opposite payoff ratio. We informed participants that points translated to monetary earnings but did not describe the spatial value contingencies. Results replicated previous work, showing no search bias toward the high-value quadrant. The Visual Choice task (Experiment 2) used virtually identical displays, except all 16 items were Ls. Participants clicked on any L and received a reward determined by the same quadrant contingencies as Experiment 1. Here, participants rapidly developed a strong choice bias toward the high-value quadrant. In Experiment 3, to rule out potential effects of priming, task context, and learning differences, we alternated 16-trial mini-blocks of the Search and Choice tasks. Strikingly, the participants preserved no spatial bias during search and a strong bias during choice. Thus, even if spatial rewards are learned (i.e., during visual choice), they are not transferred to visual search. Consequently, our visual search apparatus appears to be insensitive to spatial value.
Meeting abstract presented at VSS 2016
This PDF is available to Subscribers Only