Abstract
Anxiety has consistently been found to potentiate the attentional processing of physically salient stimuli. However, a recent study demonstrated that experimentally-inducing anxiety reduces attentional capture by previously reward-associated stimuli, suggesting that anxiety does not globally increase distractibility but rather its influence depends on the nature of the eliciting stimulus. To decipher how the processing of threat interacts with the processing of visual reward cues, we probed the neural mechanisms of oculomotor capture by previously reward-associated stimuli with and without the threat of unpredictable electric shock (blockwise manipulation, within-subjects), combining the value-driven attentional capture paradigm with the translational threat of shock paradigm in an fMRI study. Our eye-movement data replicated prior behavioral findings demonstrating an interaction between threat and distractor condition, with oculomotor capture by previously reward-associated distractors being reduced under threat. In our neuroimaging data, significant main effects of threat and distractor condition on stimulus-evoked blood-oxygen-level-dependent (BOLD) activation were evident in the extrastriate visual cortex, frontal eye field, intraparietal sulcus, and caudate tail, collectively referred to as the value-driven attention network (VDAN). In addition, we found an interaction between distractor condition and threat within each region of the VDAN. Surprisingly, the direction of this interaction was opposite that of the behavioral interaction, with the previously reward-associated distractor evoking especially strong neural activation under threat. Our neuroimaging data are consistent with the Arousal-Biased Competition (ABC) model of information processing in the visual system, although our corresponding behavioral results suggest that, at least under certain circumstances involving reward cues, such biased competition can be leveraged to facilitate more efficient ignoring at the level of eye movements. Our findings inform our understanding of value-driven attention, emotion-attention interactions, and mechanisms of signal suppression.