Activation comparisons were between retrieval of autobiographical events and general semantic knowledge. There was no difference between age groups in prefrontal cortical activation during retrieval, but there were differences between groups in hippocampal activation. As in previous studies of autobiographical retrieval, there was significant activation of the left hippocampus in young participants. For the old participants, however, there was significant activation of both left and right hippocampi, suggesting that the older adults recruited additional circuits when recalling episodes from specific times and contexts. This AZD6244 result
may suggest a neural compensatory process for recall of detailed episodes, or different strategies used for recall in the older adults. Regardless, it is likely that this difference in regional activation is initiated because
of functional changes within the circuits responsible for these behaviors. One of the most replicated results in the cognitive aging literature is that cognitive processes that rely on frontal cortical areas are particularly vulnerable to the effects of aging. In particular, maintaining a representation through working memory is reliably affected (e.g., Alexander et al., 2012; Störmer et al., 2012). Older adults show a decline in performance on tasks that require updating items in working memory (e.g., Hartman et al., 2001), in accuracy during trials with larger memory loads (e.g., Cappell et al., 2010) and in responding after a delay (e.g., Lyons-Warren et al., 2004). Similarly, aged nonhuman primates learn more and rats also show deficits in tasks that require working memory (for review Bizon et al., 2012). That is, when a delay is incorporated into the design of the task, aged animals are particularly disadvantaged (e.g., Bartus et al., Loperamide 1978; Rapp & Amaral, 1989; Muir et al., 1999; Grottick & Higgins, 2002; Ramos et al., 2003; Smith et al., 2004; Bizon et al., 2009). Two widely used working-memory tasks implemented in monkey experiments include the delayed response task (DR), which relies on the dorsolateral prefrontal cortex (PFC; Goldman & Rosvold, 1970; Passingham, 1985; Funahashi
et al., 1993) and the delayed nonmatching-to-sample (DNMS) task, which relies on the ventromedial PFC (Arnsten & Goldman-Rakic, 1990; Fig. 2C). In the DR task, a monkey is required to remember a spatial location on a screen over a brief delay period, after which it must make a saccade towards that location in order to receive a juice reward. Aged monkeys are slower to acquire the task and are impaired when longer delays are imposed (e.g., Bartus et al., 1978; Rapp & Amaral, 1989; Bachevalier et al., 1991). In the DNMS task, a monkey is first exposed to one object that it displaces to receive a reward. After a delay period, the monkey is exposed to two objects and the task requires that the novel object is displaced for the ‘nonmatch’ requirement of the task.