Word Frequency and Short-Term Recognition Memory

In Experiment I, high-frequency words resulted in poorer recognition performance than did low-frequency words only when the 'new' words on the test were also high-frequency words. When they were lowfrequency words, recognition was nearly errorless. These findings are interpreted by assuming that a frequency attribute dominates recognition performance and that frequency information for high-frequency lists was influenced by implicit associational responses. Since performance on the recognition test deteriorated somewhat during testing, Experiment II used retention intervals of up to 6 min. and showed that this effect for low-frequency words was not due to the testing itself but to forgetting over time; for highfrequency words, the results were less clearcut. The evidence suggests that forgetting may be retarded by increased temporal discrimination between 'old' and 'new' words. A number of studies have shown that word frequency, as determined by Thorndike and Lorge, is related to recognition; Gorman's study is an example.' More particularly, recognition for words of high frequency is poorer than recognition for words of low frequency. However, a theoretical analysis of recognition leads to expectations that are not in agreement with this finding for all situations. Experiment I, below, was a test of these expectations, which may be understood by indicating the type of recognition test used and the nature of the theoretical assumptions. The subject was presented a list of 50 words for a single study trial, with full instructions about the nature of the test. On the test, each of the words presented for study ('old' words) was paired with Received for publication March 6, 1970. This work was supported by Contract N00014-67-A-0356-0010, Project NR 154-057, between Northwestern University and the Office of Naval Research. Reproduction in whole or in part is permitted for any purpose by the United States Government. 1 E. L. Thorndike and I. Lorge, The Teacher's Word Book of 30,000 Words, 1944; A. M. Gorman, Recognition memory for nouns as a function of abstractness and frequency, J. exp. Psychol., 61, 1961, 23-29.