Positive effects of subclinical depression in prospective memory and ongoing tasks in young and old adults

ABSTRACT

Results reported in the literature show that depression can have either negative or neutral effects on prospective memory (PM). The goal of the present study was to broaden the analysis of depression-related effects on PM, with regard to the possibility that subclinical depression may have positive influence on PM. A total of 120 participants from four groups (young/old, subclinically depressed/non-depressed) completed event- and time-based PM tasks embedded in the linear orders task or stories task, respectively. In the event-based PM task no effects of depression were found, whereas depressed participants were more accurate in the time-based PM task, where higher monitoring during the last minute of the task was observed. It was also found that depressed participants built a mental model in the linear orders task more accurately than controls. Results of the present study are discussed with reference to the analytical rumination hypothesis.

Keywords:

• Time-based prospective memory
• Event-based prospective memory
• Subclinical depression
• Aging
• Ongoing tasks

Notes

¹We re-analyzed PM accuracy, taking into account also the results of participants who did not monitor the flow of time even once – similar pattern of results was obtained, with the main effect of depression less significant (p = .11) than in the original analysis. However, as noted above, including non-monitoring participants makes the interpretation of the results difficult, as it is impossible to identify the cause of lack of monitoring behavior. When participants check the flow of time even once, there is proof that they understood the instruction, and correctly initiated the time monitoring processes. Thus, we share the view that excluding non-monitoring participants leads to more conclusive results.

²Again, similar pattern of results was observed when participants who did not monitor flow of time even once were included into the analysis, with the crucial Mood × Time segment interaction slightly less significant (p = .02) than in the original analysis (p = .012).