Social attention biases in adults with and without Autism Spectrum Disorders: selecting and following eye-gaze and arrow cues in real-world scenes

[Background] Despite obvious impairments in following another person’s eye-gaze during social interactions, individuals with Autism Spectrum Disorders (ASD) show typical gaze-following on standard attention tasks. Most attention tasks do not differentiate gaze-selection (focusing on someone’s eyes) and gaze-following (following another person’s gaze) components of social attention and may not be sensitive to ASD differences. [Aims] The goal of this study was to first effectively characterize attention orienting during gaze-following in adults with typical development (TD) and then explore any relative differences in individuals with ASD. Participants were allowed to select cues in a manner that revealed their priorities. In particular, biases for eye-gaze vs. arrow cues were compared using a flicker task to present real-world scenes. [Results] In Experiment 1 when participants were shown either eyes or an arrow, TD adults demonstrated no preference for gaze-following over arrow-following, suggesting that single cues do not reveal biases. In Experiment 2 TD participants viewed competing eyes and arrows and showed an initial preference for gaze-following. As the task progressed, gaze-following diminished, suggesting that the behavior may be susceptible to conscious influence. Experiment 3 involved a forced choice response after variable durations of viewing the scene (i.e. short, medium, long) in order to examine the time course of gaze-following. Eyes were selected at short viewing durations, followed at medium durations, and re-selected at long durations. A different pattern was found for arrows, suggesting that they are attended differently. In Experiment 4, the visual saliency of arrows was reduced and arrows were no longer followed; arrow following may rely upon visual saliency, whereas gaze-following likely relies upon social saliency. In Experiment 5, gaze-following was examined in adolescents and young adults with and without ASD. Performance of participants with ASD differed from comparisons in two ways, they: 1. showed no preference to select eyes over arrows and, 2. did not follow eye-gaze. [Conclusion] Findings suggest that TD adults prioritize eyes and then have a flexible bias to follow another person’s eye-gaze. Preliminary ASD findings suggest that within the context of a flicker task eyes are neither prioritized nor followed. Implications for research methodology are discussed.