Acute stress, either social or physical, alters the priority of salient feared distracters but not neutral salient distracters

Abstract

Stress increases involuntary attention capture by emotional stimuli. It is unclear, however, how the predictability of feared stimuli modulates this effect relative to a neutral stimulus. Participants completed a visual search task that could include feared (spider) or neutral (butterfly) salient distracters following a stress induction and appropriate control intervention. Stress was induced either using the cold pressor test (CPT, n=31) or the Trier Social Stress test (TSST, n=27) prior to the visual search task. Salient distracters captured attention under all conditions, such that reaction times (RT) were slower when a salient distracter was present than when absent. In blocks where the valence of the salient distracter was predictable, but not blocks where either valence was equally likely, there was a three-way interaction between stress state (stress vs. control), valence (feared vs. neutral), and the presence of a salient distracter on RT (F (1,57)=6.74, p=.011, η²=.11), which did not change as a function of stress type (CPT vs. TSST, p=.777). Neutral distracters captured attention similarly under both the stress and control conditions (present vs. absent, F (1,57) = 57.7, p < .001, η² = .50; interaction with stress vs. control, p =.577). Capture by feared distracters was influenced by stress (present vs. absent × stress vs. control, F (1,57) = 12.16, p < .001, η² = .18), such that it was greater under stress than control. However, compared to neutral distracters, capture by a feared distracter was both greater under stress (spider vs. butterfly, M_Δ = 25 > 18 ms) and less (M_Δ = 11 < 21 ms) under control. These results indicate that stress, either social or physical, not only increases the priority of feared salient stimuli, but also abolishes some protective suppression from the effect of feared salient stimuli present under control conditions when its appearance is predictable.

Acknowledgement: U.S. Army Research Office Grant W911NK-09-0001