Decoding Covert Visual Attention in Space and Time from Neural Signals

Hamed, Suliann Ben · 2025 · Crossref

DOI: 10.1146/annurev-vision-101322-011902

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 article by Suliann Ben Hamed addresses the neural mechanisms underlying covert visual attention, specifically focusing on how attentional control is encoded and decoded from prefrontal neural signals in both spatial and temporal domains. The research is motivated by the need to understand how the brain prioritizes relevant stimuli in complex environments through top-down (goal-directed) and bottom-up (stimulus-driven) mechanisms. While overt attention involves eye movements, covert attention allows for focus without gaze shifts, relying on distinct but interacting dorsal and ventral attention networks. The paper aims to synthesize current knowledge on how decoding methods can track these internal cognitive processes in real time, serving as a neurophysiological proxy for the "attentional spotlight." The paper reviews methodologies for decoding covert attention from neural signals, acknowledging significant challenges such as the lack of time-locked external markers, noise correlations among neurons, and the multiplexing of cognitive signals with other variables like task engagement. It outlines discrete, continuous, and two-step decoding strategies used to infer cognitive variables from neuronal activity. The review integrates findings from electrophysiological recordings, including single-unit activity and local field potentials, alongside causal manipulation studies such as microstimulation, inactivation, and optogenetics in animal models. These methods help distinguish between downstream correlates and necessary causal drivers of attention, particularly highlighting the role of the prefrontal cortex in goal-directed control and the parietal cortex in spatial integration. Key findings indicate that decoding prefrontal activity allows for high-resolution tracking of the attentional spotlight, revealing its rhythmic nature and dynamic shifts. The review contrasts views on attentional shifts, noting evidence for rhythmic neural activity in the beta frequency range during top-down serial search, as well as continuous exploration at default alpha frequencies. It details how priority maps in the parietal and prefrontal cortices integrate sensory saliency with task relevance to guide target selection and suppress distractors, either reactively or proactively. Furthermore, the paper highlights that recurrent neural networks in the prefrontal cortex support these attention dynamics, balancing focus and flexibility across short and longer timescales of sustained attention. The significance of this work lies in providing a comprehensive model of attention that integrates dynamic prioritization processes. By demonstrating that decoded neural signals accurately reflect target cognitive states, the review underscores the validity of using decoding as a tool to interrogate the functional roles of specific cortical areas. This approach offers deeper insights into stimulus selection, distractor suppression, and the rhythmic nature of attentional saccades. Ultimately, the paper advances the understanding of how the brain allocates limited cognitive resources efficiently, bridging the gap between neural activity and behavioral relevance in complex visual environments.

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