AN ANALYSIS OF THE RECOGNITION PROCESS
Bennet B. Murdock Jr.1
UNIVERSITY OF VERMONT
The emphasis in the present paper will be on recognition as a method of measuring retention. All we shall attempt to do is to suggest some possible hypotheses about the nature of the recognition process, state them quantitatively, test them, and then explore some further implications of the most promising hypothesis. The perceptual aspect of recognition as manifested, for instance, in the recognition of ambiguous stimuli (see Binder and Feldman, 1960) is beyond the scope of the present paper and will not be considered here. No attempt will be made to survey either studies comparing recognition and recall as methods of measuring retention ( Postman and Rau, 1957) or studies assessing the reliability and validity of multiple-choice tests ( Kinney and Eurich, 1932).
Recognition is one of the traditional methods of measuring retention, and it tests the individual's ability to identify previously learned material. Three different types of recognition tests may be distinguished. The first, essentially a variant of the method of single stimuli, requires the individual to make a binary choice (usually "yes" or "no") about a single stimulus item ( Seward, 1928). Although not usually cast in this form, a true-false test question is essentially a recognition test of this type. The second type of recognition test requires the individual to select the one "correct" stimulus from a group of several stimuli ( Postman and Rau, 1957). This, of course, is the familiar multiple-choice type of test. The third type of recognition test requires the individual to select all correct stimuli from a larger number of stimuli ( Luh, 1922). This is essentially a test of knowledge of list membership and is often used in____________________