A comparison of recognition and savings as retrieval measures: A reexamination

Abstract

The present study was designed to extend the findings of Nelson (1978) in determining the relative sensitivities of recognition and savings for detecting small amounts of information in memory. Subjects learned a list of 20 number-noun paired associates to a recognition criterion. After a retention interval of 2, 3, or 4 weeks, subjects returned for tests of recall and recognition. Then, at relearning, some of the nonretrieved items remained the same as at original learning, and some new items were created by a re-pairing of numbers and nouns. After one relearning study trial, retention tests were again given. Statistical tests showed significant savings effects on all dependent measures. It was concluded that savings is a more sensitive retrieval measure than recognition and that subthreshold memory traces can exist and can be incremented by the process of concatenation. It was suggested that a reinstitution of old mediators may also play a role in the savings effect. © 1980, The Psychonomic Society, Inc.. All rights reserved.