Laboratory vs. naturalistic prospective memory task predictions: young adults are overconfident outside of the laboratory

ABSTRACT

This study investigated whether individuals can predict their future prospective memory (PM) performance in a lab-based task and in a naturalistic task. Metacognitive awareness was assessed by asking participants to give judgments-of-learning (JOLs) on an item-level for the prospective (that something has to be done) and retrospective (what to do) PM component. In addition, to explore whether giving predictions influences PM performance, we compared a control group (without predictions) to a prediction group. Results revealed that giving predictions did not change PM performance. Moreover, participants were underconfident in their PM performance in the lab-based task, while they were overconfident in the naturalistic task. In addition, item-level JOLs indicated that they were inaccurate in predicting what items they will recall or not, but only for the prospective component of the PM task. As for the retrospective component, they were equally accurate in both task settings. This study suggests a dissociation of metacognitive awareness for PM according to both task setting and processing component.

KEYWORDS:

• Prospective memory
• metacognition
• judgment-of-learning

Acknowledgements

We thank Mathilde Bastien, Alexandre Caddoux, Asli Erdemli and Vanessa Marti for assistance with data collection.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Stéphanie Cauvin http://orcid.org/0000-0001-7017-6210

Notes

1 Sample size had initially been calculated in the context of a study design comparing young and older adults. Here, an a priori power analysis indicated that a total sample size of N = 176 (88 young and 88 older adults) is large enough to detect a medium effect of ɳ2 = 0.06 (f = 0.25) with an alpha probability of 0.05 and a power of 0.90 (all power analyses were conducted using G*Power 3.10). Due to administrative reasons, the older cohort could not be tested with the present protocol and the present paper, therefore, focuses on the younger cohort only. Note, that we tested 90 younger participants altogether but had to exclude 3 participants because their native language was not French.

Additional information

Funding

Preparation of this manuscript was funded by a joint grant from the Swiss National Science Foundation (SNSF), the Agence Nationale de Recherche (ANR; France) to MK, CM, KS and CS. CM gratefully acknowledges the support of the Institut Universitaire de France.