When Hints Hurt Memory: The Influence of the Number of Part-Set Cues on Free Recall

Abstract

The present study explored the effects of part-set cues on retention in a free recall task across 5 experiments. In the part-set cueing literature, researchers typically provide half of the to-be-remembered items as cues at test; accordingly, little is known about the effect of the number of part-cues on retention, particularly when very few or very many cues are presented at test. These experiments examined the effects of very few cues (2, 3, 4, 5, and 6 cues), half cues (15), and very many cues (25, 27, 28, and 29 cues) out of a list of 30 words. Put differently, the percentage of part-set cues in the very few condition ranged from 7–20% of the list, the half-cue condition was 50% of the list, and the very many cue condition ranged from 83–97% of the list. The very few cue conditions demonstrated null effects (all ps > .26), which suggests that there might be a minimum percentage of cues required before part-set cueing influences memory performance. In contrast, significant part-set cueing impairment occurred in all conditions where at least half of the cues were present (all ps < .03, except one marginally significant, p=. 06, when comparing the 0 vs. 27 cue conditions). Generally, these results are consistent with predictions derived from the retrieval strategy disruption hypothesis (Basden & Basden, 1995). Furthermore, the patterns of results reported in the half-cue and very many cue conditions were consistent with similar conditions in Slamecka (1968), but the results of these two studies differed when few cues were present.