Abstract
In five experiments with sample sizes ranging from 47 to 104, subjects were tested for recognition memory on photographs of faces that were untransformed or transformed (laterally reversed) between presentation and test. Scores were consistently higher on untransformed than on transformed faces and on those faces looking to the observer's left than on those looking to the right. In three of the experiments, women performed better on female than on male faces, and there was some evidence that the effect of reversal was mitigated on own-sex faces. The findings are briefly discussed in terms of schema theory, and it is concluded that the two sides of the face are not symmetrical in the memory representation, where the left may be more important than the right.

This publication has 32 references indexed in Scilit: