Search

The performance of a goal-directed task is often interrupted by task-irrelevant distractors. This distractor interference has well been demonstrated in a large body of studies employing the Eriksen flanker paradigm. A notable finding regarding this issue is that distractor interference is attenuated by increased perceptual load or the presence of non-target stimuli diluting distractors. Besides increased perceptual load or the presence of diluters, we hypothesized that either suppression of distractor or enhancement of target via top-down attentional control would also contribute to modulating distractor interference. To test this, we had participants identify a target stimulus while ignoring a distractor under three different conditions; under low load, only the target and distractor was presented, while under high load, the target was surrounded by non-target letters. In the dilution condition, the target was accompanied by non-targets, whose colors were distinct from the target. Importantly, following the task stimuli, a probe stimulus was presented either in the target location, non-target location, or distractor location. As results, under low load, attention was captured by the target stimulus in a bottom-up manner at the early stage of processing. By contrast, in the high load and dilution conditions, attention was focused on the target in a top-down manner. Taken together, we suggest that increasing perceptual load or presenting diluters incentivizes the establishment of top-down bias toward the target stimulus, which plays a role in attenuating distractor interference.

Neuroimaging studies for human participants have shown that the activity in the multiple-demand (MD) network is associated with various kinds of cognitive demand. However, surprisingly, it remains unclear how this MD network is related to a core component of cognition, the process of searching for a target among distractors. This was because previous neuroimaging studies of visual search were confounded by task difficulty or time on task. To circumvent these limitations, we examined human brain activity while participants perform two different visual search tasks. The performance of a task was limited by increased attentional demand, while the other task was primarily limited by poor quality of input data or neural noise. Throughout the MD network, increased activity and strengthened functional connectivity among the MD regions were observed under the search task recruiting capacity-limited attentional resources. The present findings provide unequivocal evidence that the MD network mediates visual search, as well as other capacity-limited cognitive processes.