Abstract
Attending to visual stimuli associated with a high probability of procuring rewards is greatly adaptive. Exactly how reward interacts to guide visual attention is presently unclear. Recent research suggests high-value, but contextually irrelevant stimuli, capture attention as a consequence of reward learning and that this capture occurs independently of task goals and visual salience. Here we investigated whether stimulus reward associations learned in a visual search task, would influence performance in a subsequent spatial cueing task. The aim was to test whether reward learning would attached to specific features (feature hypothesis), or whether stimulus reward associations are specific to the context in which they are learned (context hypothesis). We further explored the conditions in which stimulus-value associations are formed by examining whether value learning requires rewarded features to be the explicit goal of search (explicit-reward hypothesis), or if value attaches to all features of an attended object (implicit-reward hypothesis). Consistent with previous reports of value-driven capture, we found significant cueing effects for high-reward stimuli (feature-hypothesis). Furthermore, significant cueing effects for high-reward stimuli were observed when the rewarded stimuli did not define the search target during the learning phase (implicit-reward hypothesis). Here we extend on previous finding by demonstrating that reward attaches to the feature defined in training, rather than the context and that these stimulus reward associations may be learned implicitly.
Meeting abstract presented at VSS 2012