Comparing different tests to detect early manifestation of prospective memory decline in aging

Abstract

Objective

Prospective memory (PM) is the ability to remember to perform future intentions. Previous studies have demonstrated that, compared to a younger cohort, healthy older adults have impairments in PM. Considering the importance of early detection of age-related PM decline, the present study aims to compare the performance of healthy older adults using three well-known PM tests commonly used in clinical settings.

Method

In the present study, we tested 70 older adults (65–95 years old) using the Cambridge Prospective Memory Test (CAMPROMPT), the Memory for Intentions Screening Test (MIST) and the Royal Prince Alfred Prospective Memory Test (RPA-ProMem). In order to compare performance across tests and the interaction between age and cues, we performed a linear mixed model with random intercept and random slopes. Moreover, additional mixed models with random intercept were run for analyzing the additional information provided by MIST and RPA-ProMem regarding delay responses, response modality effects and type of errors committed.

Results

Our data showed a drop in PM performance as age increased detected by all three tests. Furthermore, CAMPROMPT was the most sensitive test to identify differences in PM for event-and time-based cues, at least for participants with 65–77 years old. When data were analyzed in term of delay responses, participants were more accurate for 2 min delay (MIST) and 30 in delay (RPA-ProMem). Participants were less accurate when response modality was “verbal” compared to “action” (MIST) and made more PM errors as age increased.

Conclusions

Overall, the study provides important information regarding age-related PM decline and can help researchers as well as clinicians in deciding the preferred test to evaluate PM performance.

KEYWORDS:

• Prospective memory
• aging
• CAMPROMPT
• MIST
• RPA-ProMem

Acknowledgments

The information in this manuscript and the manuscript itself has never been published either electronically or in print. There are no financial or other relationships that could be interpreted as a conflict of interest affecting this manuscript. This research received no specific grant from any funding agency, commercial or not-for-profit sectors. This work was carried out within the scope of the project “use-inspired basic research”, for which the Department of General Psychology of the University of Padova has been recognized as “Dipartimento di Eccellenza” by the Ministry of University and Research. The authors gratefully acknowledge the participants, their families, and friends who participated in this study; dott. Giovanna Stornati for her support in editing the final version of this manuscript and prof. Andrea Spoto for his statistical support.

Disclosure statement

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

Notes

1 The age from when the decline can be observed depends on the parameter adopted by the study. For example, Kliegel et al. (2008) and Uttl (2011) considered the participants to be old when they were over 60, whereas Henry et al.’s (2004) “old” group was considered such when the average age was higher than 55 years.

2 The Rivermead Behavioural Memory Test (RBMT) (Wilson et al., 1989) is one of the first tools used in clinical and experimental settings to investigate PM. It has four alternative versions, all of which include three event-based PM tasks that are administered along with other measures of everyday memory. While the total RBMT score appears to be a valid measure of “everyday” memory function, Shum et al. (2002) concluded that there was little evidence to support the reliability or validity of the PM items separately. The lack of time-based tasks or tasks extending beyond the test session (i.e., long-term tasks) prevented us from including the RBMT in the present study.

3 The first and the last authors independently translated RPA-ProMem. After that we consulted a native English speaker that helped us with grammar editing and to back translate it.

4 Controls’ PM performance for CAMPROMPT and MIST was not reported in the table or in the text (Fleming et al., 2009).