Saturday December 18, 11:00 - 12:00

Prospective Memory across the Lifespan

Elizabeth A. Maylor

Research on prospective memory (PM: remembering to do something in the future without being prompted) has been relatively neglected until recently, particularly with respect to changes in PM across the lifespan. Although there have been some mixed results, it will be shown that PM does develop in childhood, and decline in old age, with further decline in dementia. Both in development and in normal aging, PM appears to be more affected when the attentional demands of the task are high. For example, it seems that effects are greater when attention-switching is required between the background task and PM task demands. Data will be presented to show that PM is no more affected by dementia than retrospective memory (RM), although patients’ PM failures are more frustrating to carers than their RM failures. In terms of the activation of PM tasks during the retention interval between encoding and execution, PM tasks are less represented in long-term memory by a heightened level of activation in old age and dementia than in young adults, which could account for at least some of the deficits shown. When PM is successful in old age and dementia, it is achieved at the expense of greater effort than in the young, for example, in terms of more checks on the clock in a time-based task, or more thoughts about the task in an event-based task. The issue of whether there is anything special about PM will be discussed, and, in particular, the question of whether we can simply predict PM changes across the lifespan on the basis of what we know about RM changes. Initial studies suggested that PM was exceptionally spared in old age but subsequent studies have shown age deficits in PM performance that are largely consistent with Craik’s (1986) framework. Thus, deficits are greater when self-initiated activity is high and environmental support is low, both for PM and RM. It certainly seems the case that findings from the RM literature can help in understanding PM data. For example, the development of temporal memory can account for some of the errors seen in data to be presented from PM experiments in children.