Stefanie Becker¹, Zachary Hamblin-Frohman², Koralalage Don Raveen Amarasekera¹; ¹The University of Queensland, Australia, ²The University of Toronto, Canada

Research has shown that attention is often biased to the relative feature of the target in visual search (e.g., redder / larger / darker) rather than its absolute feature values, in line with a Relational Account of Attention (Becker, 2010). However, it is currently unknown how tuning to relative features is achieved. If we know the feature value of the target (e.g., orange), can the visual system rapidly assess the dominant feature in the visual scene (e.g., red or yellow) and compute the relative feature of the target prior to selection (e.g., yellower or redder)? Or do we need on-task experience to learn how the target differs from the context? Another important question is how search progresses through the search items when there are multiple distractors. When an orange target is redder than the majority of other items, do we first select the reddest item, then the next-reddest item and so forth, until we find the target? The present study tested these questions in a 36-item search display with multiple differently coloured distractors and variable target and non-target colours. The first fixations on a trial showed that these displays still reliably evoked relational search, even when observers had no knowledge of the context or relative feature of the target. This indicates that information about the relative target feature can be rapidly extracted and guide attention prior to the first eye movement. Moreover, the first five fixations within a trial revealed that we tend to select the most extreme items first (e.g., red), followed by the next-extreme (e.g., red-orange), etc., until the target is found. This shows that attention is first guided by relative features and only hones in on the exact target colour after multiple fixations on relatively more extreme distractors.

Acknowledgements: Australian Research Council