Several previous studies have shown that whereas young adults perform better than older adults on prospective memory (PM) tasks in the laboratory, this superiority is often reversed in real-life PM tasks. The present studies investigated this paradox by creating a laboratory task in the form of a board game (Virtual Week) that mimicked many features of daily living. It was hypothesized that older adults might use strategies derived from their more structured lives to outperform young adults on the board game. However, contrary to our prediction, it was found that younger adults were superior. In Experiment 2 we had participants perform very similar PM tasks in real life (Actual Week), and found that now the older adults were generally superior to their younger counterparts. Possible reasons are discussed for this striking age-related difference between laboratory-based and naturalistic PM tasks.The present report is concerned with age-related differences in prospective memory (PM) performance, and how these differences are modi®ed by the setting in which testing takes place. Schon®eld (1982) drew researchers' attention to possible declines in PM with increasing age from young to older adulthood. In particular he suggested that older people have dif®culty`remembering to remember' and cited some earlier work of his own (reported by Welford, 1958) in support of this suggestion. Interestingly, in the same volume, Moscovitch (1982) reported an informal study in which participants of various ages were asked to phone the laboratory from home at various designated times; in this case the older adults outperformed their younger counterparts Ð a result directly opposing Schon®eld's suggestion. However, it seemed that the older participants in Moscovitch's study performed well largely through their greater use of external aids and reminders. Subsequent research has generally shown that older adults do perform less well than younger adults in laboratory tests of PM, but the few studies carried out in naturalistic settings have tended to ®nd an age-related improvement in PM performance. The studies described in the present article explored this possible difference between PM in the laboratory and in real life.Adult age differences in laboratory-based PM tasks are reasonably well documented at the present time. Initial reports suggested that age-related differences were slight (e.g. Einstein and McDaniel, 1990), but subsequent studies have usually found that performance declines with increasing age. Craik (1983, 1986) suggested that older adults have dif®culty with cognitive tasks that require a lot of self-initiated mental activity, and in line

The contrasting age-related trends on laboratory and naturalistic prospective memory (PM) studies were investigated with the same participants. In the first two experiments, 380 participants in three age groups (20s, 60s, and 80+) were given a naturalistic PM task of logging the time at four set times for one week. There were six between-subjects regimens that varied the complexity of the time schedule, and the opportunity to use conjunction cues and external aids. The 60s and 80+ age groups did not differ and both older adult age groups were consistently superior to the young adults on all regimens. In Experiment 3, the same participants showed a significant age-related decline on retrospective memory tasks, and on event-based and time-based laboratory PM tasks embedded within the retrospective memory tasks. The study confirmed the paradoxical age-related trends on laboratory and naturalistic PM tasks.

Although until recently much of the evidence for pregnancy-related deficits in memory was anecdotal or based on self-report, a number of studies have now been conducted that have tested whether these subjective appraisals of memory difficulties reflect objective impairment. However, these studies have failed to yield consistent results. A meta-analysis of the 14 studies that have been conducted over the past 17 years comparing pregnant and/or postpartum women with healthy matched controls on behavioral measures of memory was conducted. The results indicate that pregnant women are significantly impaired on some, but not all, measures of memory, and, specifically, memory measures that place relatively high demands on executive cognitive control may be selectively disrupted. The same specific deficits associated with pregnancy are also observed postpartum. These findings highlight the need for exploration of the etiologies and functional consequences of pregnancy-related memory difficulties and may help to guide the interpretation of neuropsychological data for the purpose of determining cognitive status in individuals who are pregnant or postpartum.

Multiple sclerosis (MS) is a white matter disease associated with neurocognitive diffi culties. More recently the potential for white matter pathology to also disrupt important aspects of emotion understanding has been recognized. However, no study to date has assessed whether capacity for facial affect recognition and theory of mind (ToM) is disrupted in MS, or whether any observed defi cits are related to more general cognitive impairment. In the present study MS participants ( n = 27) and nonclinical controls ( n = 30) were administered measures of facial affect recognition, ToM, and cognitive functioning. MS participants were signifi cantly impaired on the ToM task, and also presented with specifi c defi cits decoding facial emotions of anger and fear. Performance on the measures of facial affect recognition and ToM were related to general cognitive functioning, and in particular, measures sensitive to executive dysfunction and information processing speed. These data highlight the need for future research to more fully delineate the extent and implications of emotion understanding diffi culties in this population. ( JINS , 2009, 15 , 277-285 .)

The current study confirms the age-prospective memory-paradox within one sample and with carefully matched laboratory and naturalistic tasks. Additionally, it takes an important step forward in clarifying the role of different factors in understanding age effects across these different contexts. The results indicate that the relative importance of different factors vary as a function of assessment context, with conceptual as well as applied implications.