Blinded by fear? Prior exposure to fearful faces enhances attentional processing of task-irrelevant stimuli

Abstract

Threatening information has been shown to both capture attention and enhance sensory processing. Recent evidence has also suggested that exposure to fearful stimuli may enhance perceptual processing of subsequently presented information, as well as increase attentional capacity. However, these results are inconsistent with other findings that fearful stimuli reduce task-irrelevant distraction and improve selective attention. Here, we investigated the effect of prior exposure to fearful faces on performance in the Eriksen flanker task. Across experiments, fearful cues led to increased task-irrelevant distraction for items positioned across visual space, in contrast to other emotional expressions and inverted face items, and under conditions of attentional load. Findings support the view that fearful images enhance attentional capacity, allowing one to attend to as much visual information as possible when danger is implied. Conflicting findings on the effect of fear and selective attention are discussed.

Keywords:

• Emotion
• Selective attention
• Response competition

The research reported in this paper was supported by a 1 + 3 ESRC (Economic and Social Research Council) PhD studentship awarded to N.B. and carried out under the supervision of N.D. N.D. is also supported, in part, by a Visiting Research Associate fellowship from St John's College Research Centre at the University of Oxford.

Notes

¹ To examine any possible differences between our fearful and angry face stimuli on the basis of their emotional arousal, we obtained ratings from 32 naïve participants (10 male; mean age 29 years, SD = 6). Participants filled out questionnaires requiring them to look at the neutral, fearful, and angry faces used in the present study and rate their level of arousal using the Self-Assessment Mannequin (a widely used pictorial rating scale displaying levels of emotional arousal; Lang, 1980). A within-subjects ANOVA showed that emotional expression significantly modulated arousal ratings, F(2, 62) = 138.05, p < .001, η_p² = .82. Post hoc comparisons showed that both angry (M = 53, SD = 11) and fearful faces (M = 48, SD = 12) were rated as more emotionally arousing than neutral (M = 19, SD = 8; both ts > 10, ps < .001, ds > 2.8). Furthermore, angry faces were rated as significantly more emotionally arousing than fearful faces, t(31) = 4.49, p < .001, d = 0.45. Thus, there was no evidence that fearful faces may imbue greater emotional arousal in participants than do angry faces, with in fact evidence to the contrary.

² One such alternative account is perceptual redundancy; fearful cues could have increased the likelihood of attending to the distractor merely when it was the same stimulus as the target (i.e., compatible). We tested this possibility on 19 participants (8 male; age M = 25 years, SD = 3), using fearful and neutral cues: methods similar to those in Experiment 2; Trials (96) × Blocks (4). To control for perceptual differences, targets were M or W (i.e., the same image, a W letter, either upright or inverted), and distractors were M, W, or a shape made up of the same features (appearing similar to “XX”) acting as a neutral distractor. Consistent with previous studies (e.g., Lavie, Hirst, de Fockert, & Viding, 2004, Experiment 1), we calculated the incompatible minus neutral (I – N) distractor cost and neutral minus compatible (N – C) facilitatory/redundancy effect and entered this into a 2 × 2 ANOVA with expression, finding a significant interaction on RTs, F(1, 18) = 4.72, p < .05, η_p² = .21; no effects were found on error rates (F < 1). Fear increased I – N costs (7 ms to 18 ms), t(18) = 2.32, p < .05, d = 0.63, but did not affect facilitation/redundancy effects, if anything reducing them (10 ms to 4 ms), t(18) = 1.43, p > .15. It should be noted that facilitation effects were at floor and missed significance in general, t(18) = 1.77, p > .05. Regardless, these findings suggest that our results are not due to fear increasing perceptual redundancy to matching information on target–distractor compatible trials.