Abstract
Ensemble processing allows the visual system to condense visual information into useful summary statistics (e.g., average size), thereby overcoming capacity limitations to visual processing. Here we explored a novel dissociation between task relevant (i.e., a size cue paired with an average size ensemble task) and irrelevant (i.e., a color cue paired with an average size task) attentional cues in ensemble processing by creating a new paradigm that merged the action effect (a manipulation of attention) with ensemble-processing tasks. Participants made a simple action (action condition) if a task relevant word cue (“Large” or “Small”) corresponded to the size of a subsequent object (large or small rectangle) and made no action (viewing condition) if the cue did not correspond to the subsequent object size. Immediately after, they were shown an ensemble display of 8 ovals of varying sizes and were asked to report either the average size of all ovals (ensemble task) or the size of a single oval from the set (single task). On congruent trials the word cue corresponded to the average size of the ensemble display, while on incongruent trials the word cue did not correspond to the average size. An action effect occurred for the ensemble task, whereby cue congruency differences were found in the action but not the viewing condition. In contrast, no action effect was present in the single task. Experiments 2 and 3 examined if task irrelevant cues (colour instead of size) would also generate action effects in ensemble processing. No action effects were found in these experiments. Overall, task relevant cues that elicit an action can influence ensemble processing, but task irrelevant cues cannot, suggesting that attention is involved in ensemble processing but only when it is directed towards a task relevant feature.