Abstract
Theories of attention hypothesize the existence of an “attentional template” that contains target features in working or long-term memory. It is frequently assumed that the template contains a veridical copy of the target, but recent studies suggest that this is not true when the distractor set is linearly separable from the target (e.g., all distractors are “yellower” than an orange-colored target). In such cases, target representations may shift “off-veridical” from actual target features (e.g., be represented as “redder” than it truly is). The shifts of the target template are often considered adaptive because they are presumed to increase the representational distinctiveness of targets from distractors, which leads to better attentional selection. However, some have argued that the shifts may be entirely explained by perceptual biases created by simultaneous contrast effects in the visual system. To address this debate, we asked if target shifting occurs only when it is adaptive and increases visual search efficiency or if it occurs whenever linearly separable distractors are present. Participants searched for a target that could be identified based on either its color or orientation and we separately manipulated the linear separability of one dimension (color) and the visual similarity of the other (orientation). We found that target shifting along the linearly separable (color) dimension occurred more strongly when the other distractor feature (orientation) was highly similar to the target and hard to distinguish. Furthermore, a post-experiment strategy questionnaire suggested that participants were aware of using color more when orientation was similar and hard than when orientation was dissimilar and easier to use. We conclude that target shifting is task-adaptive and that the contents of the target template are shifted to increase target-to-distractor distinctiveness when doing so confers advantages to visual search efficiency.