In the presence of conflicting gaze cues, fearful expression and eye-size guide attention

ABSTRACT

Humans are social beings that often interact in multi-individual environments. As such, we are frequently confronted with nonverbal social signals, including eye-gaze direction, from multiple individuals. Yet, the factors that allow for the prioritisation of certain gaze cues over others are poorly understood. Using a modified conflicting gaze paradigm, we tested the hypothesis that fearful gaze would be favoured amongst competing gaze cues. We further hypothesised that this effect is related to the increased sclera exposure, which is characteristic of fearful expressions. Across three experiments, we found that fearful, but not happy, gaze guides observers’ attention over competing non-emotional gaze. The guidance of attention by fearful gaze appears to be linked to increased sclera exposure. However, differences in sclera exposure do not prioritise competing gazes of other types. Thus, fearful gaze guides attention among competing cues and this effect is facilitated by increased sclera exposure – but increased sclera exposure per se does not guide attention. The prioritisation of fearful gaze over non-emotional gaze likely represents an adaptive means of selectively attending to survival-relevant spatial locations.

KEYWORDS:

• Gaze cueing
• attentional bias
• social attention
• shared attention
• fear
• sclera

Acknowledgements

We thank the following students from the Cognitive × Affective Behaviour and Integrative Neuroscience (CABIN) lab at Northern Michigan University: Amy Abel, Isabella Elya, Ashley Finfrock, Vanessa Hiley, and Rachel Simpson for assisting in data collection; Timothy Egedus and Beth Thiele for image editing; and Bob Torrence for assisting with the pilot study.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 Faces were presented on the top and bottom of the screen, which was necessary to allow for the presentation of leftward and rightward targets. Although the faces themselves were equidistant from the central fixation cue, the eyes of the bottom face were positioned in closer proximity to fixation than the eyes of the upper face. To ensure that the results were not influenced by this difference, dot-location was included as a control factor in our analyses.