The elusive link between eye-movement patterns and facial expression recognition

Different stereotypical facial expressions convey unique signals of muscular activity in distinct face regions. For example, the recognition of anger relies on decoding the downward drawing of the brows, while disgust recognition relies on decoding nose wrinkling and upper lip raising. Therefore, it is reasonable to assume that one’s ability to identify a facial expression is strongly influenced by the facial features that gaze was fixated upon. Indeed, a large body of research supports the claim that facial fixation distributions and emotion recognition are strongly associated: fixating information-rich regions enhances emotion recognition, while suboptimal face-scanning impedes it. Here, we offer a critical assessment of whether the data support this link. We critically review several bodies of research, examining the relationship between gaze exploration of facial expressions and emotion recognition, across different cultures, neuropsychological conditions, age, gender, as well as personal tendencies of facial visual exploration and different experimental conditions. Integrating results from these diverse fields suggests that compelling evidence regarding the manner that emotion recognition relies on visual scanning patterns has yet to emerge. While it is clear that certain facial features are differentially diagnostic for stereotypical emotion recognition, how that information is sampled and extracted remains elusive. We suggest that the dominant approach of examining the distribution of fixation locations may not be sufficient for solving this puzzle. Rather, using more complex eye-tracking analyses and taking into account the extrafoveal processing are crucial for revealing the contribution of visual scanning to emotion recognition.