Representation and selection of relative position.

Heathcote, Andrew; Mewhort, D. J. K. · 1993 · Journal of Experimental Psychology Human Perception & Performance

DOI: 10.1037//0096-1523.19.3.488

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 the representation and selection of relative position information in visual search, addressing the theoretical debate between early- and late-selection models of attention. Early-selection theories, such as Feature Integration Theory (FIT), posit that simple features are processed preattentively, but their relative positions require serial, attentional processing. Consequently, these models predict that search for targets defined by shape (relative position) should yield increasing search functions. In contrast, late-selection theories allow for the preattentive processing of complex properties, predicting that such searches could exhibit "pop-out" (flat search functions). The authors aim to determine whether simple shape information, defined solely by the relative positions of components, is represented preattentively and can guide selection. The study is divided into two parts. Part 1 focuses on representation, testing whether targets distinguishable only by the relative position of their components produce pop-out effects. The authors utilize consistent-mapping, varied-mapping, and odd-man search paradigms to examine search functions (response time versus display size). They critique previous findings that supported localization-by-attention, arguing that confounds such as heterogeneous distractors and low discriminability may have obscured preattentive shape processing. Part 2 addresses selection, demonstrating strong learning in a varied-mapping paradigm to show that preattentive shape information can be used for target selection. The authors propose the "group scale hypothesis" to account for this learning, suggesting that grouping processes facilitate selection, and present a final experiment to test this hypothesis. The results demonstrate pop-out for targets defined by the relative position of their components, indicating that simple shape information is indeed represented preattentively. This finding challenges the core assumption of early-selection models that attention is necessary for localizing features to construct shape. Furthermore, the study shows that this preattentive shape information can be utilized for selection, evidenced by strong learning effects in varied-mapping tasks. The final experiment supports the group scale hypothesis, confirming that grouping processes play a critical role in visual search efficiency. These findings align with Duncan and Humphreys’ attentional engagement theory, which emphasizes the importance of grouping in visual search. The significance of this work lies in its falsification of strict early-selection models regarding shape processing. By showing that relative position is coded preattentively, the authors provide strong support for late-selection positions or modified early-selection models that incorporate preattentive grouping. The findings imply that the visual system can segregate and select objects based on simple shape differences without serial attentional scanning, provided the stimuli are sufficiently discriminable and grouped. This extends the understanding of visual search by highlighting the role of grouping mechanisms and preattentive shape representation in overcoming capacity limitations.

Key finding

Simple shape information defined by the relative position of components is represented preattentively and can be used for visual selection, as evidenced by pop-out effects and learning in varied-mapping search tasks.

Methodology

lab_experiment

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 author_sweep_intake on 2026-05-28.

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