Visual search for schematic emotional faces risks perceptual confound

Abstract

Several studies have used a visual search task to demonstrate that schematic negative-face targets are found faster and/or more efficiently than positive ones, with these findings taken as evidence that negative emotional expression is capable of guiding attentional allocation in visual search. A common hypothesis is that these effects should be disrupted by face inversion; however, this has not been consistently demonstrated, and raises the possibility of a perceptual confound. One candidate confound is the feature of “closure” (see Wolfe & Horowitz, 2004) caused by the down-turned mouth adjacent to edge of the face. This was investigated in the present series of experiments. In Experiment 1, the speed advantage for upright negative faces was replicated. In Experiment 2, the effect was not disrupted with inversion, and an efficiency advantage emerged, suggesting that perceptual features could be causing the advantage. In Experiment 3, speed and efficiency effects were seen when this perceptual characteristic remained but face features were scrambled. Taken together, these findings suggest that visual search using schematic faces containing a curved-line mouth feature cannot provide a valid test of guided search by negative facial emotion unless this confound is controlled.

Keywords:

• Emotion
• Attention
• Faces
• Facial expression
• Visual search

Acknowledgements

The research and manuscript preparation were supported by Discovery grants from the Natural Sciences and Engineering Research Council of Canada (NSERC) awarded to the second and third authors and the Australian Research Council awarded to the second author, and a post-graduate scholarship from NSERC to the first author. The authors would like to thank Elaine Fox for providing the stimuli for use in these experiments, Le-Anh Ngo and Michelle Albornoz for research assistance, and reviewers for helpful comments.

Notes

¹the stimuli used in these studies all contain a schematic representation of the outside of the face, and internal features of eyes, nose and mouth. However, the stimuli used by Öhman et al. (2001b) also included eyebrows, angled up for positive faces and angled down for negative faces.