Reduced Attentional Capture Following More Variable Rewards

Value-driven attentional capture is a robust phenomenon in which a reward-associated stimulus automatically captures our attention despite being irrelevant to the task at hand. Last year, we investigated the occurrence of value-driven attentional capture in the presence of reward variability. We did not observe a difference in attentional bias for stimuli associated with consistent or variable reward (expected value equated). Rather, our results suggested that individual preferences for variable or consistent reward, as revealed by a decision-making task, predicted which reward schedule more strongly influenced attention. This year, we will present a follow-up study that probes the relative strength of value-driven attentional biases when the reward variance is considerably more extreme. In this follow-up study, participants were presented with color squares predictive of reward feedback during a training phase. One color was associated with a consistent amount of reward (five cents) while another color predicted reward with high variance (one dollar or no reward), with expected value equated between colors. After training, attentional bias was measured using a task in which participants fixated on a target circle while ignoring a square distractor, each of which could appear in the colors experienced during training. Finally, a manual decision-making task was presented where participants selected between the same colors experienced during training in an effort to maximize earnings. The results showed a significant attentional bias towards the consistent reward color, and a marginally significant correlation between the frequency of distractor fixations and reward preference during the decision-making task. These results replicate our previous finding in which attentional bias and choice preference are linked. Strikingly, in contrast to some more recent reports, we also demonstrate that more variable rewards do not always lead to more robust attentional biases, and that with sufficient variability, consistent rewards will more strongly drive attention.