Abstract
The ability to rapidly detect threatening situations would seem to offer an organism an important survival advantage, and hence be something likely to enter its genetic blueprint. History suggests that one major source of threat for humans is another human intending to do harm (either socially or physically); typically evidenced by their facial expression of anger. Thus, an argument has been made that humans have evolved the ability to more rapidly detect angry faces than faces baring other expressions. Initial results in support of this ‘anger superiority effect’, using pictures of real faces as stimuli, have since been undermined by evidence that the effect is likely due to low-level visual artefacts. In an attempt to control for these artefacts, some researchers have experimented with schematic face images (simple line drawings). Reports have emerged of a strong and consistent search advantage for angry faces. This has been seen as significant because the low-level artefact counter-argument is a much more difficult case to mount. Recently we reported (Coelho, Cloete, and Wallis, 2010) that the search advantage could still be found using images which no longer resembled faces, but maintained similar line structures to the schematics. Nonetheless, by only testing for this effect using a constant set-size, these results could be construed as being due to response biases rather than detection speed per se. This follow-up study specifically addressed this concern by testing over multiple set sizes. The expected variations in search slope were measured for both schematic and abstract line stimuli, confirming that the effect is due to a difference in detection speed. The results confirm therefore, that the search advantage exists, but that it is likely due to low-level visual features rather than a projection of threat. In a follow-up study we describe investigations into the source of this strong search asymmetry.
Meeting abstract presented at VSS 2013