The neural basis of attentional control in visual search

Eimer, Martin · 2014 · Crossref

DOI: 10.1016/j.tics.2014.05.005

archive: archived pipeline: cataloged verified

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

Summary

This review by Martin Eimer (2014) addresses the neural mechanisms underlying attentional control in visual search, specifically how humans localize and identify target objects among distractors when target locations are unpredictable. The paper synthesizes neuroscientific findings from human and monkey studies to propose a four-stage model of attentional processing: preparation, guidance, selection, and identification. This framework posits that attention is not a single control system but emerges from the coordinated, real-time operation of distinct neurocognitive mechanisms closely linked to working memory and recurrent feedback loops. The author analyzes existing literature, including electrophysiological (ERP), fMRI, and single-neuron recording studies, to delineate the temporal and functional characteristics of each stage. During the **preparation** phase, search goals are represented in working memory as "attentional templates." Neural evidence shows that visual cortical areas (e.g., inferior temporal cortex) and prefrontal cortex exhibit sustained, goal-selective activation prior to stimulus onset. These preparatory baseline shifts are position-independent, allowing for search without precise spatial cues. In the **guidance** phase, following stimulus onset, feature-based attention operates globally across the visual field. Neurons in areas like V4 and MT enhance activity for task-relevant features (e.g., color, motion) regardless of location, providing parallel guidance signals. The **selection** stage involves the transition from global feature processing to spatially specific attention. Guided by priority maps in posterior parietal cortex or frontal eye fields, recurrent top-down signals enhance neural responses in ventral visual cortex for candidate targets. This process is marked by the N2pc ERP component (~180 ms post-stimulus). The final **identification** stage requires integrating features and matching them to search goals, a process dependent on visual working memory. This is reflected by a sustained posterior contralateral negativity (CDA), indicating the maintenance of selected objects via recurrent feedback loops. The review also discusses the debate between serial and parallel selection, suggesting that both strategies may be employed depending on task demands, such as target discriminability and distractor density. The significance of this work lies in providing a unified temporal framework for interpreting visual search data. It clarifies the distinct roles of working memory (holding templates vs. maintaining selected objects) and recurrent feedback (sustaining representations) at different processing stages. The model highlights that visual search efficiency depends on the complex interplay of these mechanisms rather than a single bottleneck. Furthermore, it identifies critical outstanding questions, such as the causal role of preparatory baseline shifts and the conditions determining serial versus parallel selection, offering a roadmap for future research into the neural basis of selective attention.

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.

• visual search
• attention selective divided