Task-irrelevant faces capture attention regardless of perceptual load
26.522, Saturday, 17-May, 2:45 pm - 6:45 pm, Pavilion
Shiori Sato1, Jun Kawahara1; 1Chukyo University
According to load theory (Lavie, 2010), the perceptual load of cognitive tasks determines the extent of flanker interference such that the ability to identify a central target letter decreases due to interference from peripheral flanker distractors under low perceptual load, whereas no such interference occurs under high perceptual load. It has been unclear whether load theory holds when face distractors are irrelevant to central identification tasks. The present study examined whether human face distractors captured attention under conditions of high perceptual load when the faces were entirely irrelevant to the task. Participants identified a target letter (N or X) in a circular array of five non-target letters. In 10% of the trials, a face distractor simultaneously appeared above or below the search array. In another 10% of the trials, a phase-scrambled control image appeared. No distractor was presented in the remaining trials. Perceptual load was manipulated by varying the type of non-target letters (low load: homogenous non-targets, "o"; high load: heterogeneous non-targets that shared features with the target). Although load theory predicts no or reduced attentional capture, the reaction times for letter identification were delayed in trials with a peripheral face distractor compared with those with a control image. This result suggests that attentional capture by an entirely task-irrelevant distractor occurred regardless of perceptual load. Additional experiments revealed that the capture effect was specific to faces; no capture was obtained with inverted face images or meaningful non-face images (e.g., food). Moreover, Experiment 4 replicated face-specific attentional capture even when the distractor did not share the temporal component of abrupt onset with the search array. These results extend the notion that faces are exceptions to load theory when faces and search items share no common display features.