When a face is (or is not) more than the sum of its features: Configural and analytic processes in facial temporal integration

Abstract

To investigate temporal integration in face recognition, top and bottom halves of pictures of famous people were presented sequentially, either upright or inverted, with varying temporal intervals between the two halves. The inversion effect, a marker of configural processing, was comparable across 0–400 ms intervals, but decreased at intervals exceeding 400 ms (Exp. 1). When an interfering stimulus appeared during the interval between the two face parts (Exp. 2), it disrupted the integration of the parts but not their perception. This is the first report of such an effect. Thus, performance equalled the combined accuracy of each part when presented alone, which in turn was worse than when they were integrated. Our findings indicate that (a) configural processing of faces depends on integration of face parts that are maintained temporarily in a visual buffer; (b) without integration, identification depends on recognition of individual parts whose contributions are additive; and (c) an interfering visual stimulus can obstruct integration, but leaves perception of individual parts intact. The ability to integrate temporally separated face parts into a unified representation is discussed in light of theories of face perception and temporal integration.

Acknowledgements

We express our deep appreciation to Philippe Schyns and three anonymous reviewers for their helpful comments and suggestions on earlier versions of this paper. We thank Marilyne Ziegler, Irina Nica, and Aliza Goodman for their help in programming, running, and analysing the experiments.

Notes

¹One might claim that higher accuracy was obtained in the top only condition than in the bottom only condition because of masking effects, as the former condition was presented 417 or 1017 ms before the mask, while the mask preceded the latter condition by only 200 ms. However, as noted earlier, our results clearly show that masking effects are no longer apparent 200 ms (and even earlier) after the stimulus presentation. For example, in Exp. 2 in the 200 ISI condition an interfering mask appeared 150 ms after the bottom part of the face. Nevertheless, the accuracy in this condition was equal to the no mask condition where the whole-face mask appeared 200 ms after the bottom part. The same claim could be demonstrated in the 800 ISI condition, where no differences were found between the three condition. Of course it could be claimed that at ISIs of 150–200 ms masking effects still exist and the small difference between the two ISIs precluded us from seeing any differences. However, since perceptual masking occurs at short ISIs this temporal difference should be significant. Furthermore, in Exp. 1 we have ISIs that are similar to the long ISIs in the top only condition. Specifically, the ISI between the top part and the mask in Exp. 1 varies between 217 (in the 0 ISI) and 1817 ms (in the 1600 ISI). If an active masking is operating at ISIs longer than 150–200 ms we should have seen an increase in accuracy as the ISI lengthens, since the ISI between the mask and the top part varies. This was not found.