Electrophysiological evidence for parallel and serial processing during visual search

Luck, Steven J.; Hillyard, Steven A. · 1990 · OpenAlex-citations

DOI: 10.3758/bf03211606

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 the neural mechanisms underlying visual search, specifically testing whether detection of targets relies on parallel or serial processing models as proposed by Treisman’s feature integration theory. The authors aimed to distinguish between competing theories that predict similar behavioral outcomes (reaction times) but propose different underlying cognitive processes. By recording event-related potentials (ERPs) alongside behavioral measures, the study sought to determine if set size effects on reaction time are due to perceptual processing stages or post-perceptual decision processes. The experiment involved twelve participants performing a visual search task with two conditions: "feature-present" (parallel search) and "feature-absent" (serial search). Stimuli consisted of arrays of triangles, where targets were distinguished either by the presence of a unique line or by the absence of a line common to distractors. Set sizes varied (4, 8, or 12 items). ERPs were recorded from multiple scalp sites, focusing on the P3 component, which indexes stimulus recognition and classification. The analysis compared P3 latency and amplitude against reaction times to isolate processing stages. Results indicated distinct processing strategies for the two conditions. In the feature-absent condition, reaction times increased linearly with set size, consistent with serial search. Crucially, the slope of the function relating P3 latency to set size was identical to that of reaction time for target-present trials. This finding suggests that the increased reaction time is driven entirely by prolonged stimulus identification and classification, rather than post-perceptual processing. Furthermore, P3 amplitude in this condition increased with set size and was influenced by the target status of the preceding trial, patterns consistent with a serial, self-terminating search model. In contrast, the feature-present condition showed shallow reaction time slopes and P3 latencies largely independent of set size, supporting parallel processing. Additionally, differences in ERP scalp distributions suggested that distinct physiological processes underlie the detection of feature presence versus absence. The findings provide electrophysiological evidence supporting the serial self-terminating search model for feature-absent targets, contradicting parallel processing accounts for this specific paradigm. The study demonstrates that ERP measures can dissociate perceptual from post-perceptual contributions to reaction time, resolving ambiguities in behavioral data. It confirms that different neural mechanisms are engaged depending on whether a target is defined by the presence or absence of a feature, validating aspects of feature integration theory while highlighting the utility of ERPs in mapping cognitive processes.

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