Commission errors in delay–execute prospective memory tasks

ABSTRACT

Individuals frequently retrieve an intention, but the execution of the task needs to be delayed due to ongoing task demands - so-called delay-execute prospective memory (PM) tasks. We investigated commission errors in the delay-execute paradigm. Participants were told that a PM task is finished (PM task has been executed and is now finished for a final phase) or cancelled (PM task has been cancelled immediately after introduction). We observed commission errors and ongoing task performance in the final phase which included several irrelevant PM cues. In two experiments, we observed significantly more commission errors for cancelled compared to the finished intentions. In Experiment 2, commission errors were eliminated if the final phase required divided attention, regardless of PM task status. In addition, we observed significantly more PM cue interference on the ongoing task in the cancelled compared to the finished group, indicating that the PM task was retrieved in the cancelled group but not in the finished group. As retrieval and execution of the PM task were separated by a delay, the results indicate that commission errors are not always the result of a quick, spontaneous retrieval-execution sequence and may also occur when retrieval and execution are temporally separated.

KEYWORDS:

• Delay–execute paradigm
• Prospective memory
• Commission errors

Acknowledgements

We thank Daniel Diethei, Armin Winkler, and Tristan Wehner for assistance with data collection. We appreciate the helpful comments of three anonymous reviewers. The research was presented at the 55th Tagung experimentell arbeitender Psychologen (TeaP) in Giessen, Germany, April 2014, the 4th International Conference on Prospective Memory in Naples, Italy, May 2014, and the 56th Tagung experimentell arbeitender Psychologen (TeaP) in Hildesheim, Germany, March 2015.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Philipp Schaper http://orcid.org/0000-0001-5960-7921

Notes

1. For the analysis of the delay phase of non-interrupted PM blocks we calculated separate per-participant averages for reaction times for trials before the red screen (pre-phase: Trials 2–5) and after the red screen, excluding the trial immediately after the red screen (post-phase: Trials 7–20; cf. Ball et al., 2013). We calculated a 4 (PM status: finished, cancelled, active PM, no‐PM) × 3 (phase: pre-phase, post-trial, post-phase) mixed ANOVA. There were effects of phase, F(2, 128) = 91.59, p < .001,  = .59, and group, F(3, 64) = 5.18, p = .003,  = .20, as well as an interaction, F(6, 128) = 4.69, p < .001,  = .18. Bonferroni-corrected post hoc tests showed significant differences between pre-phase and post-trial (p < .001) and between post-trial and post-phase (p < .001), but not between pre-phase and post-phase (p = .615). Separate one-way ANOVAs for the pre-phase, F(3, 73) = 0.67, p = .576,  = .03, and the post-phase, F(3, 73) = 1.79, p = .157,  = .07, did not reveal any group differences, indicating that there was no active monitoring or maintenance of the intention in the pre-phase and the post-phase, respectively.

Additional information

Funding

This work was supported by the Deutsche Forschungsgemeinschaft [grant number GR 3970/1-1].