Judgments of 3D Bars in Depth

Évaluations de la Profondeur des Barres en Infographie 3D

PDF

Authors
  1. Hollands, J.G.
  2. Parker, H.A.
  3. Morton, A.
Corporate Authors
Defence R&D Canada - Toronto, Toronto ONT (CAN)
Abstract
Twenty participants judged the relative size of two bars portrayed at different locations in a 3D bar graph. Bars were co-located, placed side by side (near adjacent), or the small bar occluded the large (near or far occluded). Error was greater for the far occluded condition and there was greater variability in bias scores. To account for observed error, we proposed a model that distinguishes between cyclical bias commonly observed in proportion judgments and bias resulting from improper size-distance scaling. The model was fit to data, and the results indicated that the absolute value of size-distance scaling parameter gamma was greater in the far occluded condition. Inclusion of gamma increased R2 for far and near occluded conditions only. Bar location did not affect cyclical bias. Thus, judgments of the relative sizes of bars in 3D bar graphs showed increased error when the bars were separated, due to inaccurate size-distance scaling.

Il y a un résumé en français ici.

Keywords
Bar graphs;3D graphs;Linear perspective;Judgment bias;Proportion judgments;Stevens law;Quantitative models;Size-distance scaling;Size constancy
Report Number
DRDC-TORONTO-SL-2002-112 — Scientific Literature
Date of publication
01 Oct 2002
Number of Pages
5
DSTKIM No
CA023166
CANDIS No
520362
Format(s):
Electronic Document(PDF);CD ROM

Permanent link

Document 1 of 1

Date modified: