Convex hull test of the linear separability hypothesis in visual search
Related Experiment Video
Updated: Oct 19, 2026

A Psychophysics Paradigm for the Collection and Analysis of Similarity Judgments
Published on: March 1, 2022
B Bauer1, P Jolicoeur, W B Cowan
1Department of Psychology, University of Waterloo, Ontario, Canada.
Abstract:
Visual search for a colour target in distractors of two other colours is dramatically affected by the configuration of the colours in CIE (x, y) space. To a first approximation, search is difficult when a target's chromaticity falls directly between (i.e. is not linearly separable from) two distractor chromaticities, otherwise search is easy (D'Zmura [1991, Vision Research, 31, 951-966]; Bauer, Jolicoeur, & Cowan [1996a, Vision Research, 36, 1439-1466]; Bauer, Jolicoeur, & Cowan [1996b, Perception, 25, 1282-1294]). In this paper, we demonstrate that the linear separability effect transcends the two distractor case. Placing a target colour inside the convex hull defined by a set of distractors hindered search performance compared with a target placed outside the convex hull. This is true whether the target was linearly separable in chromaticity only (Experiments 1 and 2), or in a combination of luminance and chromaticity (Experiments 3 and 4).
Related Concept Videos
Fisher's Exact Test
Depth Perception and Spatial Vision
Causes of Similarity-Dissimilarity Effect
Behrens–Fisher Test
This test is...
Divergence Theorem in 3D Space
Test for Homogeneity

