How many trees does it take to make a forest?

Research output: Contribution to journalArticlepeer-review


In order to study the relative perceptual availability of global and local features in very sparse patterns, subjects were asked to make 'same'/'different' judgments on pairs of geometrical figures and the times needed to detect global and local differences were compared. With triangular patterns a global precedence was found which could be attributed to size differences. With rectangular patterns global precedence was larger, not accounted for by size differences, and indifferent both to the number of elements and to their spacing. Thus it was demonstrated that global precedence may hold for patterns with as few as four elements. Patterns with smooth edges could be compared much more quickly than patterns with serrated eges. It is proposed that configurational properties of some of the patterns interfered with the encoding of their global structures or with comparing them. It is argued that the results support a principle of global addressability which postulates that visual schemata are mainly addressed through their global constituents.

Original languageEnglish
Pages (from-to)239-254
Number of pages16
Issue number3
StatePublished - 1983

ASJC Scopus subject areas

  • Experimental and Cognitive Psychology
  • Ophthalmology
  • Sensory Systems
  • Artificial Intelligence


Dive into the research topics of 'How many trees does it take to make a forest?'. Together they form a unique fingerprint.

Cite this