Information and viewpoint dependence in face recognition

How we recognize faces despite rotations in depth is of great interest to psychologists, computer scientists and neurophysiologists because of the accuracy of human performance despite the intrinsic difficulty of the task. Three experiments are reported here which used three-dimensional facial surface representations to investigate the effects of rotations in depth on a face recognition task. Experiment 1, using "shape only" representations, showed that all the views used (full-face, three-quarter and profile) were equally well recognized when all had been learned. Performance was better when the same views were presented in an animated sequence rather than at random, suggesting that structure-from-motion provides useful information for recognition. When stimuli were presented inverted, performance was worse and there were differences in the recognizability of views, demonstrating that the familiarity of upright faces affects generalization across views. Experiments 2 and 3 investigated generalization from single views and found performance to be dependent on learned view. In both experiments, generalization from learned full-face fell off with increasing angle of rotation. With shape only stimuli, three-quarter views generalized well to each other, even when inverted but for profiles generalization was equally bad to all unlearned views. This difference may be explained because of the particular relationship of the profile to the axis of symmetry. In Experiment 3, addition of information about superficial properties including color and texture facilitated performance, but patterns of generalization remained substantially the same, emphasizing the importance of underlying shape information. However, generalization from the three-quarter view became viewpoint invariant and there was some evidence for better generalization between profiles. The results are interpreted as showing that three-dimensional shape information is fundamental for recognition across rotations in depth although superficial information may also be used to reduce viewpoint dependence.