Abstract
Rewards can help teach selective attention what is relevant in a complex world. Previous studies have demonstrated that specific features consistently associated with high reward capture attention more than those associated with low reward. How reward influences attentional capture at the level of entire feature dimensions remains unclear. Here we test the hypothesis that learning to attend to a highly rewarding feature in one dimension results in attentional capture for other unrewarded features within that same dimension, compared to features in other dimensions. Participants viewed a series of stimuli ("tokens") that differed in color and orientation, and "collected" each token via a button press in order to earn its point value. For each block of trials, tokens with one specific feature (e.g., red) provided high reward with 90% probability and low reward with 10% probability. Tokens without this feature provided high reward with 10% probability and low reward with 90% probability. Which feature was rewarded changed on every block. Participants were randomly assigned to either a 'color' group or an 'orientation' group, and all rewarded features were drawn from the corresponding dimension. To probe attention, we occasionally interrupted the token collection task with a visual search task. Each search array contained both a color singleton and an orientation singleton, one of which was randomly chosen each trial to be the search target. We hypothesized that a learned dimensional attention bias would facilitate pop-out and thus search for singleton targets in the dimension containing the highly rewarded feature. Consistent with this hypothesis, the color group had faster response times for color versus orientation singleton targets. No differences were observed for the orientation group. These findings suggest that reward can drive attentional capture at the dimensional level, at least for color.
Meeting abstract presented at VSS 2017