Dependent or independent? The role of working memory in prospective memory

Prospective memory (PM) refers to the processes associated with the planning and execution of delayed intentions. PM involves a complex balance between the memory of an intention and the maintenance of concurrent activities, thus requiring different cognitive mechanisms. Recent studies have examined the involvement of WM in the PM in order to investigate whether these two processes share cognitive resources or they are independent mechanisms. It was shown that PM is supported by the central executive component of working memory (WM), which organizes the execution of joined actions through selection of relevant information and suppression of the irrelevant ones. According to the preparatory attention and memory model (PAM), PM task performance is always influenced by the availability of WM resources, since controlled attentional processes are needed. Differently, the multiprocess framework posits that a delayed intention can be automatically retrieve, at least under certain circumstances, suggesting that PM and WM are distinct mechanisms.This study aims to examine the role of WM in PM, by comparing the performance of MB, a patient with a disexecutive syndrome and proved PM difficulties acquired after brain damage, to that of nine healthy participants. Two experiments were developed, in which subjects had to perform an ongoing activity together with a PM switching task. We administered two distinct ongoing tasks for the Experiments, differing for the amount of cognitive resources required: a simple arithmetic (Experiment 1) and a high-load WM task (PASAT test, Experiment 2). As a further variable, the modality of PM retrieval was manipulated by introducing both time and event-based conditions for each experiments. In the time-based conditions, the PM task had to be performed at a given time, whereas in the event-based condition PM retrieval occurred following cue presentation. MB’s performances in PM tasks significantly differ from the controls only in the PASAT time-based condition. By the comparison of Experiment 1 and Experiment 2 results in time and event-based conditions, significant differences also arose in the effect size between MB and healthy subjects, which were selective for time-based conditions.Our results indicated that WM has an influential role on PM performance only in complex tasks when active self-retrieval is required (PASAT-time), suggesting that PM and WM are partially independent. Such findings provided support to the multiprocess framework, highlighting how WM and PM can be considered as distinct mechanisms at least when the prospective retrieval is associated with a cue (event-based conditions). Moreover, they emphasized PM as a complex high-level metacognitive function, which requires coordination of future intentions through adoption of strategic and controlled processes.