Psychometric properties of the Actual Week test: a naturalistic prospective memory task

Prospective memory is the ability to 'remember to remember' and a facet of memory important to everyday functioning. For older adults, prospective memory slips are a common concern. In the present study, we conducted an initial validation of a paper-and-pencil adaptation of the Actual Week test, and reported on internal consistency, inter-rater and test-retest reliability, convergent and divergent validity, as well as veridicality of the task.

Fifty-eight healthy, community-dwelling older adults were recruited from a larger randomized controlled trial and tested at baseline. The Actual Week test was a naturalistic five-day prospective memory task where participants were assigned eight hypothetical tasks to remember per day for five days. Tasks were either time-cued or event-cued and regular (i.e. occurring daily) or irregular (i.e. varied each day). The proportion of tasks that were recorded as on time and accurate was used as the primary measure of performance.

The Actual Week test had good internal consistency (Kuder-Richardson: r > .8), intra-test (intraclass correlation: α > .9) and test-retest reliability (r = .76). There was also evidence for convergent and divergent validity. Task performance was associated with age, but not years of education or sex.

The Actual Week test demonstrated strong psychometric qualities and promising evidence for validity as a performance-based measure of everyday prospective memory in older adults. Avenues for future studies include extending the evidence for convergent validity and evaluating feasibility and utility with other clinical populations.