Visual search processes and the multivariate point symbol

Elisabeth S. Nelson, David Dow, Christopher Lukinbeal, Ray Farley

Research output: Contribution to journalArticlepeer-review

11 Scopus citations

Abstract

This study reviews the major theories of visual search processes and applies some of their concepts to searching for multivariate point symbols in a map environment. The act of searching a map for information is a primary activity undertaken during map-reading. The complexity of this process will vary, of course, with symbol design and map content. Multivariate symbols, for example, will be more difficult to search for efficiently than univariate symbols. The purpose of this research was to examine the cognitive processes used by map readers when searching for multivariate points symbols on a map. The experiment used Chernoff Faces as the test symbol, and a symbol-detection task to assess how accurately and how efficiently target symbols composed of different combinations of facial features could be detected. Of particular interest was assessing the role that different combinations of symbol dimensions and different combinations of symbol parts played in moderating search efficiency. Subject reaction times and error rates were used to evaluate the efficiency of the searches. Results suggested all searches employed serial search processes, although feature searches (those in which a target symbol consists of a unique feature) were by far the easiest for subjects to complete. It was also demonstrated that hierarchical relationships could be manipulated within symbols to increase search efficiency for searches in which the target does not have a unique feature (conjunctive search).

Original languageEnglish (US)
Pages (from-to)19-33
Number of pages15
JournalCartographica
Volume34
Issue number4
DOIs
StatePublished - 1997

ASJC Scopus subject areas

  • Earth-Surface Processes

Fingerprint Dive into the research topics of 'Visual search processes and the multivariate point symbol'. Together they form a unique fingerprint.

Cite this