JOURNAL OF VERBALLEARNINGAND VERBALBEHAVIOR, 12, 140-149 (1973) Strategy Control and Directed Forgetting~ WALTER RHTMAN, JANE TANNER MALIN, ROBERT A. BJORK AND BARBARA HIGMAN The University of Michigan, Ann Arbor, Michigan 48104 Signals to subjects to forget presignal items (F-items) can completely eliminate the pro- active interference postsignal items (R-items) would normally suffer from the F-items. To determine what happens to F-items, a strategy-control procedure was developed to test memory for F-items without destroying forget-signal credibility. The subjects saw short lists of paired associates, Some containing a signal to forget presignal pairs. After each list, a single stimulus probe tested memory for the appropriate response. Tested pairs were always R-pairs unless a prearranged cue informed the subjects that the stimulus was from an F-pair. Although F-pairs did not interfere with recall of R-pairs, there was substantial recall of and interference among F-pairs. The results support a set differentiation mechanism of directed forgetting. The processes by which information no longer needed is eliminated or set aside are as fundamental to the efficient functioning of an information processing system as are the pro- cesses by which information is acquired. Any limited capacity system without the means to select and eliminate is doomed to an unfor- tunate and incoherent end; without some mechanism to prevent old information from interfering with the processing of current information, the system's output will eventu- ally bear no sensible relation to its input. One promising approach to the problem of information elimination is the recent active research on directed forgetting (for reviews see Bjork, 1972; Epstein, 1972). This approach centers on the use of signals to subjects that they can forget some or all of the information We are indebted to J. E. K. Smith for his advice on the design of this study and the analysis of the data, and to Judith Reitman, Donald Norman, and Gordon Bower for a close reading and thoughtful comments on an earlier draft. We also acknowledge support received from the National Institutes of Mental Health, the National Science Foundation, and the Mental Health Reseach Institute of The University of Michi- gan. Requests for reprints should be sent to Walter Reitman, The University of Michigan, 205 North Washtenaw Place, Ann Arbor, Michigan 48104. Copyright © 1973by AcademicPress, Inc. All rights of reproduction in any form reserved. they have been presented. Such signals (F-cues) can have remarkable effects; subjects are able, for example, to take advantage of F-cues in some situations in a way that completely eliminates the proactive interference that in- formation presented subsequent to the F-cue would normally suffer from the preceding to- be-forgotten information. The present study concerns itself with the fate of to-be-forgotten information (F-items). Rather than to in- directly investigate the effects of an F-cue on F-items by looking at whether or not F-items interfere with to-be-remembered information (R-items), the present study employs direct tests of subjects' memory for F-items. The reasons one would want to test subjects' memory for F-items are obvious: The extent to which F-items are or are not recallable or recognizable in different situations provides a basis for inferring the mechanisms by which F-items are marked, segregated, erased, or whatever. Unfortunately, there are obvious and formidable problems in attempting to test F-items. Given that the directed-forgetting paradigm is based on instructing subjects that an F-signal means forget the designated items, any test of F-items violates fair play to some: extent. 140