Title: Sign-dependent detectability of change in curvature as an empirical measure of perceptual closure
Authors: Vandekerckhove, Joachim
Panis, Sven
Wagemans, Johan #
Issue Date: 2005
Publisher: Pion ltd
Conference: Paper presented at the 28th European Conference on Visual Perception (ECVP), A Coruña, Spain. Abstract published in Perception, 34 (Supplement), 174b
Abstract: In 2-D image perception, changes in shape are more salient if they involve a change in a concavity than in a convexity. As the distinction between concavities and convexities is conditional on the assignment of figural status, this observation can be employed to measure figural status, and hence closure. Experiment 1 involved a temporal 2AFC change-detection task in which observers viewed two meaningless polygonal contours or silhouettes. They were either identical, or one vertex had been removed (or added). The size of the change was also manipulated. We confirmed the advantage for concavities found by Barenholtz et al (2003 Cognition 89 1-9). A strong effect of change size also emerged, with no difference between contours and silhouettes. In experiment 2, a 'concavity effect' was similarly obtained with spline-smoothed (non-angular) contour stimuli. In experiment 3, we used the same paradigm to disentangle two theoretical explanations for the effect: a local effect of 'mere' concaveness versus a global effect of part structure being dependent on the sequence of apexes (or curvature extremes) in the contour. In a new manipulation, we either removed (or added) a vertex, or we changed the extremity of a single vertex, thereby preserving part structure. It is concluded that both the local and the global components independently contribute to the concavity effect. In experiment 4, we applied the procedure to examine two traditional measures of closure. Using polygonal contours with a single gap, we manipulated gap size (corresponding to the C measure--Elder and Zucker, 1994 Vision Research 34 3361-3370) and the relative orientation of the terminating edges (corresponding to the relatability criterion--Kellman and Shipley, 1991 Cognitive Psychology 23 141-221). Only when both measures predicted closure, did the concavity effect emerge. We conclude that both measures indicate necessary conditions for closure.
ISSN: 0301-0066
Publication status: published
KU Leuven publication type: IMa
Appears in Collections:Laboratory for Experimental Psychology
Department of Managerial Economics, Strategy and Innovation (MSI), Leuven
Quantitative Psychology and Individual Differences
# (joint) last author

Files in This Item:

There are no files associated with this item.


All items in Lirias are protected by copyright, with all rights reserved.