یک بررسی پتانسیل برانگیخته حالت پایدار بصری حافظه و سالخوردگی

Old age is generally accompanied by a decline in memory performance. Specifically, neuroimaging and electrophysiological studies have revealed that there are age-related changes in the neural correlates of episodic and working memory. This study investigated age-associated changes in the steady state visually evoked potential (SSVEP) amplitude and latency associated with memory performance. Participants were 15 older (59–67 years) and 14 younger (20–30 years) adults who performed an object working memory (OWM) task and a contextual recognition memory (CRM) task, whilst the SSVEP was recorded from 64 electrode sites. Retention of a single object in the low demand OWM task was characterised by smaller frontal SSVEP amplitude and latency differences in older adults than in younger adults, indicative of an age-associated reduction in neural processes. Recognition of visual images in the more difficult CRM task was accompanied by larger, more sustained SSVEP amplitude and latency decreases over temporal parietal regions in older adults. In contrast, the more transient, frontally mediated pattern of activity demonstrated by younger adults suggests that younger and older adults utilize different neural resources to perform recognition judgements. The results provide support for compensatory processes in the aging brain; at lower task demands, older adults demonstrate reduced neural activity, whereas at greater task demands neural activity is increased.

It has been well documented that old age is accompanied by changes in memory and cognition (e.g. Budson & Price, 2005). Specifically, older adults experience difficulties with episodic memory, defined as the explicit and declarative recollection of previously experienced personal events (Tulving, 1985). Age-associated episodic memory deficits have been demonstrated during the performance of recognition tasks, where older adults tend to exhibit greater difficulties recalling the context of an item than the item itself (Glisky, Rubin, & Davidson, 2001). Old age is also associated with a reduction in the efficiency of working memory processes (Hartley, Speer, Jonides, Reuter-Lorenz, & Smith, 2001). Working memory involves short-term memory combined with the executive processes involved in the temporary storage, maintenance and manipulation of information (Baddeley, 1992). In addition to a reduction in performance, ageing is accompanied by increased response times on memory tasks, thought to be the consequence of elongated processing times in the brain (Christensen, 2001). Neuroimaging studies utilizing techniques such as positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) have provided evidence that older adults recruit additional resources mediated by the prefrontal cortex (PFC) to encode, store and retrieve information from memory in order to compensate for reduced neural processing efficiency in the brain (Cabeza et al., 2002 and Reuter-Lorenz et al., 2001). Specifically, age-associated increases in bilateral activity of the PFC during memory performance are thought to reflect such compensatory processes (Cabeza et al., 2002). With younger adults, spatial working memory has typically been found to be lateralised in the right hemisphere and verbal and object working memory to be lateralised in the left hemisphere (Courtney et al., 1996 and Smith et al., 1995). In contrast, older adults have been found to show a reduction in the lateralisation of neural activity during working memory activation (Mattay et al., 2006), regardless of the nature of the processed material (Reuter-Lorenz et al., 2000). Similarly, studies have demonstrated that older adults show a more bilateral pattern of PFC activity during episodic encoding and retrieval than younger adults (Grady et al., 2002, Madden et al., 1999 and Morcom et al., 2003). Collectively, these findings provide evidence that older adults recruit more neural resources in the PFC to perform the same memory processes as younger adults (Cabeza, 2002). The PFC has been found to play an essential role in executive function (D’Esposito et al., 1998), with the dorsolateral region specifically implicated in the storage and manipulation of the contents of working memory (Smith & Jonides, 1997). Consistent with the premise that memory deficits in older adults stem from reductions in frontal lobe function (West, 1996), the findings from neuroimaging studies suggest that decreases in the activity of the right dorsolateral prefrontal cortex (DLPFC) during retrieval contribute to age-associated deficits in the executive processes underlying working memory (Rypma and D’Esposito, 2000, Rypma and D’Esposito, 2001 and Rypma et al., 2001). In line with a compensatory approach to ageing in the brain, increases in bilateral frontal activity have been proposed to compensate for such inefficient PFC processes (Cabeza, 2002 and Cabeza et al., 2002). Additionally, several studies have provided evidence that older adults engage additional frontal resources to compensate for less efficient activity in posterior neocortex (Grady et al., 1994 and Schiavetto et al., 2002). Similarly, the results from electrophysiological research have revealed that both the frontal and posterior regions of the fronto-parietal network involved in spatial working memory are vulnerable to the effects of ageing (McEvoy et al., 2001 and Muller and Knight, 2002). Electrophysiological studies of the event related potential (ERP) old/new recognition effect have provided insights into the temporal characteristics of episodic retrieval in both young and older adults (Friedman, 2000). In the old/new recognition memory task, participants discriminate between previously studied (old) and unstudied (new) items. Correctly recognised old stimuli elicit larger N400 and P600 components of the ERP waveform than items correctly identified as new (Curran, 1999). Wilding and Rugg (1996) initially identified two distinct old/new effects when participants recognised items and later recalled whether words were presented by a male or female voice. The first effect was maximal in amplitude over the left parietal and temporal scalp and had an onset of around 400–500 ms post stimulus, with a duration of approximately 400 ms. A second effect had a later onset and exhibited a more sustained time course (>1 s), with a maximum amplitude over the right frontal scalp. The ERP parietal effect is thought to represent a simple judgment of prior occurrence (Morcom & Rugg, 2004) and parietal activity observed in neuroimaging studies of recognition has been interpreted in a similar manner (Cansino et al., 2002 and Iidaka et al., 2006). In contrast, the frontal effect mediated by right prefrontal regions is involved in monitoring the outcome of a retrieval attempt (Morcom & Rugg, 2004), as well as integrating information about an item’s previous occurrence with its initial contextual features (Rugg, Fletcher, Chua, & Dolan, 1999). In older adults, the parietal old/new effect tends to be attenuated and occurs later in time than it does for younger adults, whereas the frontal effect is relatively preserved (Fjell et al., 2005, Li et al., 2004 and Morcom and Rugg, 2004). On the basis of this finding, it has been suggested that the recruitment of the PFC during successful episodic retrieval is not necessarily compromised by age (Li et al., 2004). This view is at odds with the behavioural findings that during episodic retrieval, older adults tend to exhibit greater difficulty with memory for context than memory for items (Glisky et al., 2001). In contrast, several studies have failed to identify the frontal ERP effect in older adults and this has been attributed to a deficit in source memory mediated by the PFC (Fabiani et al., 1999 and Wegesin et al., 2002). To date, the neural networks underpinning episodic and working memory have been specifically investigated using PET and fMRI neuroimaging techniques, as well as electrophysiological techniques such as ERPs. The steady state visually evoked potential (SSVEP) elicited by a task irrelevant 13 Hz light flicker has also been found to be sensitive to a range of rapidly changing cognitive processes such as attention (Silberstein, Line, Pipingas, Copolov, & Harris, 2000b), working memory (Ellis et al., 2006 and Silberstein et al., 2001), recognition memory (Pipingas & Silberstein, 1995) and long term memory (Silberstein, Harris, Nield, & Pipingas, 2000a). A 13 Hz visual flicker has been used to target the low frequency resonant system (alpha band) described by Regan (1989). SSVEP (13 Hz) amplitude changes have been described as being akin to changes in the alpha band (8–12 Hz) (Silberstein, 1995). For example, reduced 13 SSVEP amplitude was found during an attentional task (Silberstein et al., 1990) and increased 13 Hz SSVEP amplitude was found during the hold period of a working memory task (Silberstein et al., 2001). These changes are consistent with the many reports of reductions in alpha amplitude during attentional processing (Klimesch, 1999) and increased alpha activity during a rejection task (Ray & Cole, 1985), respectively. A specific advantage of the SSVEP is that it enables the assessment of both sustained task related cognitive processes, such as visual vigilance (Silberstein et al., 1990) and more transient, fast, occurring processes associated with memory (Silberstein, 1995). More recently, the SSVEP has been found to be useful for isolating neural activity associated with “hold” period of a working memory task (Silberstein et al., 2001). PET and fMRI neuroimaging methods are limited in their ability to resolve such brief cognitive activity and ERP activity is generally not maintained for time periods exceeding 1 s following stimulus presentation (Perlstein et al., 2003). SSVEP investigations allow continuous monitoring of cognitive processes over extended periods with the possibility of resolving activity down to 1/13 s.