Neuro-cognitive models of single-trial EEG measures describe latent effects of spatial attention during perceptual decision making

Ghaderi-Kangavari, Amin; Rad, Jamal Amani; Parand, Kourosh; Nunez, Michael D. · 2022 · Crossref

DOI: 10.1101/2022.04.07.487571

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 spatial top-down attention influences the latent cognitive components of perceptual decision-making, specifically distinguishing between visual encoding time (VET), evidence accumulation, and other non-decision processes such as motor execution. While spatial attention is known to improve behavioral performance, previous models could not separately identify its effects on these distinct stages. The authors hypothesized that spatial prioritization would affect VET and other non-decision times, but not the evidence accumulation process itself. The researchers utilized an open-source dataset from a face-car perceptual decision-making task involving 16 participants. The experimental design featured a 2x2 factorial manipulation of spatial prioritization (informative one-way cue vs. uninformative two-way cue) and stimulus coherence (high vs. low). EEG data were recorded using 64 channels. To extract single-trial neural measures, the authors applied Singular Value Decomposition (SVD) to EEG epochs time-locked to stimulus onset, isolating N200 latencies (approximately 125–225 ms post-stimulus) as a marker for the onset of evidence accumulation. These single-trial N200 latencies, along with behavioral response times and accuracy, were integrated into hierarchical Bayesian neuro-cognitive drift-diffusion models. Four competing models were fitted to test whether spatial attention influenced VET (linked to N200 latency), other non-decision times, or both. Model selection was informed by simulation studies and convergence diagnostics. The results provided evidence that spatial top-down attention manipulates both visual encoding time and other non-decision time processes, such as motor execution, but does not affect the evidence accumulation rate. Model comparison indicated that spatial prioritization shifts the non-decision time components associated with N200 latencies as well as residual non-decision times unrelated to N200. This finding suggests that the performance benefits of spatial attention arise from faster sensory encoding and post-decision processes rather than a more efficient accumulation of sensory evidence. These findings refine the understanding of the neural mechanisms underlying spatial attention in decision-making. By demonstrating that attention affects pre- and post-decision stages rather than the core decision process, the study challenges assumptions that attention primarily modulates evidence accumulation. The use of single-trial EEG measures within a hierarchical Bayesian framework allows for a more precise dissociation of cognitive components than traditional averaged ERP analyses. This approach highlights the utility of neuro-cognitive modeling in linking electrophysiological dynamics to specific cognitive parameters, offering a robust method for future research into the latent effects of attention and other cognitive factors.

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-nvidia on 2026-08-10; verification: verified.

Topics

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

• neuro workload indices