Responding to social and symbolic extrafoveal cues: cue shape trumps biological relevance

Abstract

Social cues presented at visual fixation have been shown to strongly influence an observer’s attention and response selection. Here we ask whether the same holds for cues (initially) presented away from fixation, as cues are commonly perceived in natural vision. In six experiments, we show that extrafoveally presented cues with a distinct outline, such as pointing hands, rotated heads, and arrow cues result in strong cueing of responses (either to the cue itself, or a cued object). In contrast, cues without a clear outline, such as gazing eyes and direction words exert much weaker effects on participants’ responses to a target cue. We also show that distraction effects on response times are relatively weak, but that strong interference effects can be obtained by measuring mouse trajectories. Eye tracking suggests that gaze cues are slower to respond to because their direction cannot easily be perceived in extrafoveal vision. Together, these data suggest that the strength of an extrafoveal cue is determined by the shape of the cue outline, rather than its biological relevance (i.e., whether the cue is provided by another human being), and that this shape effect is due to how easily the direction of a cue can be perceived in extrafoveal vision.

Mental states modulate gaze following, but not automatically

“…When, by contrast, participants were instructed to perform a target discrimination task (i.e., Experiment 2), the target was rapidly fixated (< 500 ms), and often within one or two fixations, illustrating that the actor's gaze direction was processed within a relatively short time period. Moreover, participants rarely fixated the actor's face prior to fixating the target, suggesting that this gaze-induced attentional effect occurred even when gaze and head direction were processed peripherally (Hermens, Bindemann, & Burton, 2015). Thus, in the target discrimination scenario, rapid, or automatic-like, processing did not occur with respect to what the model could see.…”

Eye movements while judging faces for trustworthiness and dominance

“…These participants were also recruited using the online participant recruitment system, and also received course credit for taking part. Recruitment and reimbursement differed between the first and the latter two groups, but based on past experience with different recruitment methods and reimbursements within a single set of experiments (and with consistent results, Hermens, Bindemann & Burton (2017) ), we do not expect these differences to have an effect. Participants all provided written consent for the study that was approved by the local ethics committee (University of Aberdeen, UK), in agreement with guidelines of the British Psychological Society and the declaration of Helsinki.…”

Does agreement mean accuracy? Evaluating glance annotation in naturalistic driving data

“…There are also indications that foveal vision is specifically used in situations where shape information is difficult to extract from extrafoveal vision. For example, when asked to report the direction of a pair of eyes (embedded inside a face, likely to cause visual crowding) or a pointing hand presented away from fixation, participants more often make an eye movement to the pair of eyes than the hand when allowed to, and perform more poorly in a direction discrimination task of the pair of eyes than the pointing hand when not to allowed to look at the stimuli (Hermens, Bindemann & Burton, 2017 ). Peripheral vision, however, needs to serve a purpose.…”

“…These studies therefore provide less information about the consequences of asking actors to look back to center, also because they focus on automatic attention shifts, and less on estimating the direction of a perceived glance shift. Social attention studies have also largely focused on eye-glance shifts, with fewer studies examining whole-head glance shifts, whereas whole-head glance shifts tend to result in stronger attention shifts in the observer, particularly when the stimuli are looked at from peripheral vision (Burton et al, 2009 ; Hermens et al, 2017 ). Future studies should therefore examine what the effects of using dynamic versus static stimuli are, and whether showing the section of the video in which the actor looks back at the center position affects the findings.…”