The roles of relevance and expectation for the control of attention in visual search.

Berggren, Nick; Eimer, Martin · 2019 · Crossref

DOI: 10.1037/xhp0000666

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study investigates how task relevance and probabilistic expectations jointly influence the control of attention during visual search. Specifically, it addresses whether preparatory attentional templates—mental representations of target-defining features—are sensitive only to the relevance of features or if they can be adaptively adjusted based on the likelihood of those features appearing. While previous research established that multiple feature templates can be maintained in parallel, it remained unclear if the activation strength of these templates varies with expectation. The researchers conducted two experiments using a two-color visual search task. Participants searched for a target bar of a specific orientation among distractors, where the target could appear in one of two colors. In an 80-20 probability condition, one color appeared in 80% of trials (expected) and the other in 20% (unexpected). Search displays were preceded by spatially uninformative color cues matching either the expected or unexpected target color. Experiment 1 measured behavioral reaction times (RTs) to assess spatial cueing effects. Experiment 2 replicated this design while recording electroencephalography (EEG) to measure the N2pc component, an electrophysiological marker of attentional allocation to task-relevant stimuli. This allowed for a direct assessment of attentional capture by cues and target selection independent of motor response stages. The results revealed a critical dissociation between the effects of relevance and expectation. Behavioral and EEG data showed that attentional capture by color cues was identical regardless of whether the cue matched the expected or unexpected target color. Both cue types elicited significant spatial cueing effects and comparable N2pc components, indicating that preparatory attentional templates are activated based solely on feature relevance and are insensitive to probability expectations. However, expectations significantly modulated the speed of target selection within the search displays. RTs and N2pc components for targets in the unexpected color were delayed compared to those in the expected color. This demonstrates that while preparation is blind to probability, the subsequent selection process is facilitated by expectations. These findings suggest that relevance and expectation operate through distinct mechanisms in visual attention. Task relevance appears to be specified at the level of individual features, creating binary "on/off" states for attentional templates that cannot be graded by probability. In contrast, expectations likely operate in an object-based fashion, influencing the efficiency of target selection after initial attentional capture. This challenges the view that expectations modulate early sensory-perceptual stages of template activation, supporting a model where relevance guides initial attentional allocation while expectation facilitates later stages of target identification and response.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-25; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• attention selective divided