Horstmann, G. & Bauland, A. (2006). Search asymmetries with real faces: Testing the anger-superiority effect. Emotion, 6, 2, 193-207
http://www.uni-bielefeld.de/psychologie/ae/AE02/EJournals.html
:
Conclusion
:
We have also pointed out that the general threat-advantage hypothesis (that it is of evolutionary advantage to detect threatening faces easily) does not enforce the acceptance of the preattentive-threat-detector hypothesis (that facial threat is a preattentively available feature similar to the known basic perceptual features). The general idea is also compatible with a sensory-bias hypothesis (that facial expressions have adapted to the extant perceptual mechanisms of the intended observer). In our view, the evidence―including the present Experiments 3 and 4―favors the sensory bias hypothesis, which is also easier to bring in line with present theories of visual search. Finally, consistent with most of the previous research, facial expressions of anger were not searched for efficiently; thus, we obtained no strong evidence for a preattentive detection of threat.