The influence of cue probability on item and source judgments in item method directed forgetting

ABSTRACT

The current study examined how selective rehearsal strategies in item method directed forgetting are influenced by the probability of remember or forget cues from different sources. In four experiments, study words were presented by one of two sources in an item method directed forgetting paradigm. In all experiments, one source was mostly-remember (presenting twice as many remember as forget words) and the other source was mostly-forget (presenting twice as many forget as remember words). Participants completed item recognition tests (providing cue tags in Experiment 2) with source judgments. Item recognition of forget words was generally greater for the mostly-remember source than for the mostly forget source, whereas recognition of remember words was largely unaffected by source cue probability. Source judgments were consistent with heuristic guessing based on memory strength and knowledge of source cue probability. Experiment 4 analysed overt rehearsal, and showed that words from the mostly-remember source were more likely to be rehearsed prior to the memory cue. Results are discussed in terms of the influence that source cue probability knowledge has on selective rehearsal strategies, recognition decisions, and source memory attributions.

KEYWORDS:

• Directed forgetting
• source memory
• recognition memory
• intentional forgetting

Acknowledgements

I am thankful to Victoria Kavanagh for her work with data collection and coding in Experiment 4. I am grateful to Colin M. MacLeod for feedback on an earlier version of this article, and to Brandon Slaney, Landon Churchill, Sarah Hogan, Chelsea Hudson, Molly MacMillan, and Whitney Willcott-Benoit for their assistance with data collection.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 A pilot experiment was conducted in which both sources were presented at the centre of the screen, rather than on the left and right. Results were generally consistent with the current experiment, indicating that participants can acquire source memory cue probability information with sources differing only on the basis of identity information, rather than needing both identity and location information to differentiate the sources.

2 The cue × source interaction approached significance at p = .061, with the difference in confidence between mostly-remember and mostly-forget sources being numerically larger for forget than remember items. The next smallest p-value for any of the other interactions was .387.

3 Recognition accuracy was analysed twice, once including five of these six participants (data from one participant were lost due to computer error) and once excluding them. The results did not differ, so only the analyses including the 32 participants whose rehearsal data were analysed are included here for parsimony.

Additional information

Funding

This research was supported by a Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grant to Kathleen L. Hourihan.