Measures of Information Processing in Concept Identification

Document Type



Behavioral and Organizational Sciences (CGU)

Publication Date



Cognitive Psychology | Psychology | Social and Behavioral Sciences


Developed a method to show multiple-hypothesis behavior and states of information during concept identification (CI). The method was tested with 34 undergraduates. Compared to M. Levine's blank-trial method, the new method provided more detail about information held by S at each stage of a problem and allowed examination of process assumptions of current CI models. Most Ss attempted to process information on all dimensions simultaneously; some initially selected a subset of dimensions to process. Processing errors were made following positive as well as negative feedback. No CI model could account for the data in detail, although protocols from individual Ss were consistent with aspects of the various models. A random-reversal assumption is suggested as a mechanism to account for departures from perfect processing.

Rights Information

© 1974 American Psychological Association

Terms of Use & License Information

Terms of Use for work posted in Scholarship@Claremont.