Finding the face in a crowd: Relationships between distractor redundancy, target emotion, and target gender

Abstract

The authors developed and tested a novel theory to reconcile opposing findings in the literature on visual search for emotional target faces among neutral distractor faces. We predicted and found that the nature of the distractor stimuli was very important with overall better search performance when distractors were redundant (i.e., “familiar” because they were drawn from a small rather than large stimulus set). This effect interacted with target gender and target emotion. Female happy target faces were always more quickly detected than female angry target faces. Similarly, a happy target face advantage was obtained with male target faces when the distractors were non-redundant. However, more rapid detection of angry than happy target faces was observed with male targets and redundant distractors. Thus, our results demonstrate that claims of an anger superiority effect in visual search for faces are empirically valid, but only under circumscribed boundary conditions.

Keywords:

• Facial emotion
• Attention
• Anger superiority
• Visual search
• Distractor redundancy

Acknowledgements

The research reported in this article was supported by grants to the first author from the Swedish Research Council, the Bank of Sweden Tercentennial foundation, and from the US National Institute of Mental Health (P50 MH 72850) to the Center for the Study of Emotion and Attention at the University of Florida, Gainesville, USA. Arne Öhman is also affiliated with the Stockholm Brain Institute and the NIMH Center for Research on Emotion and Attention, University of Florida-Gainesville. The first draft of the article was written while he was a fellow at the Center for Advanced Studies in the Behavioral Science, Stanford.

The authors want to express their gratitude to Drs Jennifer Eberhardt and Gunilla Bohlin for invaluable comments on an earlier draft of this article.

Notes

¹Specifying the units of the implicit stimulus set involves a somewhat arbitrary decision between defining the stimulus set units in terms of faces (i.e., counting multiple emotional expressions by the same persons as separate units) or in terms of individuals (ignoring expressions). Empirical findings (e.g., Schweinberger & Soukup, 1998), and theory (e.g., Haxby, Hoffman, & Gobbini, 2000), as well as intuition, concur in viewing individual identity as a primary unit when perceiving faces. Nonetheless, which one to choose in the present case, in which we simply distinguish small from large implicit stimulus sets, is probably of little practical consequence.