Visual attention and stimulus identification.

Pashler, Harold; Badgio, Peter C. · 1985 · Journal of Experimental Psychology Human Perception & Performance

DOI: 10.1037//0096-1523.11.2.105

archive: archived pipeline: cataloged verified

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

Summary

This paper investigates whether visual attention requires a serial scan for stimulus identification or if multiple familiar stimuli can be identified in parallel. The authors address a critical limitation in previous visual search research: standard detection tasks (e.g., finding a target among distractors) may rely on simple feature extraction rather than full pattern recognition, allowing subjects to reject distractors without identifying them. To resolve this ambiguity, Pashler and Badgio designed a novel "highest digit" task requiring exhaustive identification of all items in a display, alongside standard character search tasks. The study employed five experiments manipulating display size (two, four, or six items) and visual quality (high vs. low contrast/intensity). In Experiments 1 and 2, subjects performed character search tasks. Experiment 1 used a standard target (A among distractors), while Experiment 2 used a "potential illusory conjunction" condition (E among F and L) to force accurate feature conjunction, theoretically requiring serial processing if Treisman and Gelade’s model were correct. Experiments 3–5 utilized the highest digit task, where subjects named the numerically highest digit in an array, necessitating the conceptual encoding of every item. Reaction times and error rates were measured to test predictions of serial versus parallel encoding models. Specifically, the authors tested whether the effects of display size and visual quality were additive (supporting parallel processing) or multiplicative (supporting serial processing). The results demonstrated that in both the standard search tasks and the exhaustive highest digit task, the effects of display size and visual quality were largely additive. In the highest digit task, reaction times increased with display size and decreased visual quality, but these factors did not interact significantly in a way that would suggest serial encoding. Even in the feature-conjunction search task, which theoretically demanded serial attention, the data showed additivity rather than the multiplicative interaction predicted by serial models. The performance in the highest digit task was strikingly similar to conventional search tasks in terms of reaction time slopes and display size effects. These findings strengthen the case for parallel identification of multiple familiar stimuli, suggesting that full-scale pattern recognition can occur simultaneously rather than sequentially. The additivity of display size and visual quality implies that these factors affect different, parallel stages of processing. The authors conclude that parallel identification does not necessarily entail late selection theory’s claim of unlimited capacity; rather, identification may be parallel but capacity-limited. This challenges early selection views that posit serial encoding for complex pattern recognition and suggests that visual search paradigms may indeed reflect parallel identification processes, not just simple feature detection.

Key finding

The additive interaction between display size and visual quality in an exhaustive identification task provides evidence for parallel, rather than serial, encoding of multiple visual stimuli.

Methodology

lab_experiment

Sample size: 40

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. Discovered via openalex_abstract on 2026-05-08.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

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

• visual search