Visual search for schematic affective faces: Stability and variability of search slopes with different instances

Abstract

The threat-advantage hypothesis that threatening or negative faces can be discriminated preattentively has often been tested in the visual search paradigm with schematic stimuli. The results have been heterogeneous, suggesting that the choice of particular stimuli have profound effects on search efficiency. Because this conclusion is hampered by differences in experimental procedure, I selected examples from past literature and presented replicas of stimulus pairs (schematic positive and negative faces) in a within-participants design. Although there was a consistent advantage for angry-face targets, search efficiency varied between 8 and 35 ms/item, yielding no clear evidence for the threat-advantage hypothesis. Furthermore, search efficiency for negative- and positive-face targets was highly correlated over stimulus pairs, which implies that whatever complicates the search for the negative face of a pair also complicates the search for the positive face. This results pattern argues against the hypothesised preattentive detector.

Acknowledgements

I am indebted to Nadine Hübner, Nadine Potthast, and Nadine von Rothkirch, for collecting the data, Lily Silny for assistance in manuscript preparation, and to Stefanie Becker, Michael Niepel, Murray White, and two anonymous reviewers, for their helpful comments on an earlier version of the manuscript.

Notes

¹Note that throughout the article, I refer to covert shifts of attention that need not necessarily coincide with overt shifts, i.e., eye movements (cf. Posner, Synder, & Davidson, 1980).

²In fact, unpublished data from our lab show that search efficiency is a function of the curvature of the mouth.

³In a yet unpublished paper, Horstmann, Bergmann, Burghaus, and Becker report on evidence that perceptual grouping is indeed a main factor for the facial valence effect with schematic faces.