Instructed forgetting: Rehearsal control or retrieval inhibition (repression)?

In a Brown-Peterson short-term memory task three different incentive conditions (remember and rehearse, remember without rehearsing, or forget) were cued by varying the color background of a trigram slide. A self-paced interpolated task was used so that differences in covert rehearsal could be inferred from backward counting rates under the various motivational conditions. It was concluded that S's poor recall in the forget condition was due to a slower rehearsal rate than in the other conditions. However, this explanation was insufficient in accounting for poorer recall in the forget as compared to nonrehearse condition, where an interpretation involving differential attention to the trigram at presentation was offered. Each difference in memory performance was matched by a corresponding difference in counting performance. Results of a postexperimental recognition test and questionnaire supported these conclusions.