Abstract
Theories of attention hypothesize the existence of an "attentional template" that contains target features in memory. It is often assumed that the template contains veridical target features, but recent studies have found that the template is flexible and shifts away from anticipated distractors to increase the target-to-distractor distinctiveness. Here, we investigated if except for shifting, the target representation can also be sharpened in response to distractor competition. In two experiments, participants were instructed to search for a target colored circle among three distractors on visual search "training" trials. Separate target identification "probe" trials were interleaved to measure the target representation. On these trials observers judged whether a single colored stimulus (sampled from both sides of the target color) was the target; "Yes" responses were modeled using a split normal distribution, and taken as an estimate the target template. In Experiment 1, we found that distractor predicability resulted in observers being more likely to misidentify colors in the direction opposite to distractors as the target, compared to a control group. The predictable directionality of visual search distractors produced a shift in the target representation away from distractors. In Experiment 2, we manipulated the strength of distractor competition in addition to distractor directionality. Participants always saw distractors from one side of the target in color space, but the similarity of those distractors to the target increased gradually over 5 blocks of visual search. We found that the magnitude of shift remained constant across blocks, but there was a continuous increase in the sharpening of the template between the target and visual search distractor colors. This suggests that shifting and sharpening are two separable mechanisms that increase target's salience. Shifting occurred in response to directionality of distractor features, while sharpening occurred in response to the strength of competition from the distractors.
Meeting abstract presented at VSS 2018