Measuring icon complexity: An automated analysis

A Forsythe, N Sheehy, M Sawey

Research output: Contribution to journalArticlepeer-review

50 Citations (Scopus)


Measures of icon designs rely heavily on surveys of the perceptions of population samples. Thus, measuring the extent to which changes in the structure of an icon will alter its perceived complexity can be costly and slow. An automated system capable of producing reliable estimates of perceived complexity could reduce development costs and time. Measures of icon complexity developed by Garcia, Badre, and Stasko (1994) and McDougall, Curry, and de Bruijn (1999) were correlated with six icon properties measured using Matlab (MathWorks, 2001) software, which uses image-processing techniques to measure icon properties. The six icon properties measured were icon foreground, the number of objects in an icon, the number of holes in those objects, and two calculations of icon edges and homogeneity in icon structure. The strongest correlates with human judgments of perceived icon complexity (McDougall et al., 1999) were structural variability (r(s) = .65) and edge information (r(s) =.64).

Original languageEnglish
Pages (from-to)334-342
Number of pages9
JournalBehavior research methods instruments & computers
Issue number2
Publication statusPublished - May 2003

ASJC Scopus subject areas

  • Psychology(all)
  • Psychology (miscellaneous)
  • Experimental and Cognitive Psychology


Dive into the research topics of 'Measuring icon complexity: An automated analysis'. Together they form a unique fingerprint.

Cite this