The Impact on Emotion Classification Performance and Gaze Behavior of Foveal versus Extrafoveal Processing of Facial Features

Abstract

At normal interpersonal distances all features of a face cannot fall within one’s fovea simultaneously. Given that certain facial features are differentially informative of different emotions, does the ability to identify facially expressed emotions vary according to the feature fixated and do saccades preferentially seek diagnostic features? Previous findings are equivocal. We presented faces for a brief time, insufficient for a saccade, at a spatial position that guaranteed that a given feature – an eye, cheek, the central brow, or mouth – fell at the fovea. Across two experiments, observers were more accurate and faster at discriminating angry expressions when the high spatial-frequency information of the brow was projected to their fovea than when one or other cheek or eye was. Performance in classifying fear and happiness (Experiment 1) was not influenced by whether the most informative features (eyes and mouth, respectively) were projected foveally or extrafoveally. Observers more accurately distinguished between fearful and surprised expressions (Experiment 2) when the mouth was projected to the fovea. Reflexive first saccades tended towards the left and center of the face rather than preferentially targeting emotion-distinguishing features. These results reflect the integration of task-relevant information across the face constrained by the differences between foveal and extrafoveal processing (Peterson & Eckstein, 2012).