Update

Multiple-trace theory and semantic dementia: Response to K.S. Graham (1999)

Patients with semantic dementia (SD), who have an incontrovertible deficit in semantic memory, are reported to show good day-to-day memory for recent events; but experimental evidence on their anterograde episodic memory/new learning is somewhat sparse and does not always tell a consistent story. We describe the performance of five SD patients, relative to controls, on (a) a range of semantic memory measures that predictably revealed substantial impairment, and (b) a newly designed naturalistic and incidental episodic task, which included information regarding the items and context of the semantic tasks. As a group, the patients’ episodic memory for these natural events was good, even after a 24-h delay, although case-by-case analysis revealed some heterogeneity in performance. These findings are discussed with regard to the neural substrate of episodic memory and psychological models of long-term memory.

Studies of autobiographical memory in semantic dementia (SD) have yielded either a reversed temporal gradient or spared performances across the entire lifetime. This discrepancy might be owing to the fact that these studies did not take into account disease severity. Our aim was to study patterns of autobiographical memory impairment according to disease severity and to unravel their mechanisms in 14 SD patients, using an autobiographical memory task assessing overall and strictly episodic memories across the entire lifetime. We divided our patients in 2 subgroups of 7 patients each, one mild and one moderate according to their level of disease severity. The results indicated for the mild subgroup selective preserved performances for the most recent time period (last 12 months period) for both autobiographical memory scores. In the moderate subgroup, performances were impaired for both scores whatever the time period. Within-group comparisons across time periods showed a recency effect and a reminiscence bump in the mild subgroup and only a less important recency effect in the moderate subgroup, suggesting that with disease severity, old memories (reminiscence bump) tend to vanish and even recent memories are less well retrieved. A correlation analysis was carried out on the entire group, between the overall autobiographical memory score and performances provided by a general cognitive evaluation (semantic memory, executive functions, working and episodic memory). The results of this analysis reflect that mechanisms of disruption of autobiographical memory in SD predominantly involve a deficit of storage of semantic information in addition to faulty executive retrieval strategies. Finally, our result and those of the literature suggest the existence of 3 distinct autobiographical memory impairment patterns in SD according to disease severity: firstly preserved performances whatever the time period, secondly a reversed temporal gradient with a reminiscence bump and thirdly the appearance of a “step-function”.

Episodic or autobiographical recollection involves re-experiencing a past event that is specific in time and place, while semantic recollection is concerned with facts and general knowledge about the world. Several prominent memory theories posit that the hippocampus differentiates between these two types of memories, mediating episodic, but not semantic, retrieval. In this chapter we explore a different view of hippocampus, one that emphasizes a singular response of the hippocampus during memory encoding and retrieval of both episodic and semantic memories, based on an amalgam of two existing theories of hippocampal function, multiple trace theory (MTT; Nadel & Moscovitch, 1997), and cognitive map theory (O'Keefe & Nadel, 1978). We review neuropsychological and neuroimaging literature suggesting that both semantic and episodic memory retrieval engages the hippocampus, at least within the normally functioning brain. We then describe an updated version of MTT that incorporates these new findings. Finally, we explore the notion that differences in the role the hippocampus plays in these forms of memory reflect two critical factors - the nature of the information being retrieved, and the requirements of the retrieval task.

We examined autobiographical memory performance in two patients with semantic dementia using a novel measure, the Autobiographical Interview [Levine, Svoboda, Hay, Winocur, & Moscovitch (2002). Aging and autobiographical memory: Dissociating episodic from semantic retrieval. Psychology and Aging, 17, 677–689], that is capable of dissociating episodic and personal semantic recall under varying levels of retrieval support. Earlier reports indicated that patients with semantic dementia demonstrate autobiographical episodic memory loss following a “reverse gradient” by which recent memories are preserved relative to remote memories. We found limited evidence for this pattern at conditions of low retrieval support. When structured probing was provided, patients’ autobiographical memory performance was similar to that of controls. Retesting of one patient after 1 year indicated that retrieval support was insufficient to bolster performance following progressive prefrontal volume loss, as documented with quantified structural neuroimaging. These findings are discussed in relation to theories of limbic-neocortical interaction in autobiographical memory.

The time-scale of medial temporal lobe (MTL) involvement in storage and retrieval of episodic memory is keenly debated. To test competitive theories of long-term memory consolidation, the present work aimed at characterizing which cerebral regions are involved during retrieval of recent and remote strictly episodic autobiographical memory. Using positron emission tomography (PET), we examined mental retrieval of recent (0–1 year) and remote (5–10 years) autobiographical memories, controlling for the nature of the autobiographical memories (i.e., specificity, state of consciousness, vividness of mental visual imagery, emotion) retrieved during scanning by behavioral measures assessed at debriefing for each event recalled. Cognitive results showed that specificity and emotion did not change with time interval although both autonoetic consciousness and mental image quality were significantly higher for recent memories, suggesting an underlying shift in the phenomenal experience of remembering with the passage of time. The SPM analysis revealed common activations during the recollection of recent and remote memories that involved a widespread but mainly left-sided cerebral network, consistent with previous studies. Subtraction analysis demonstrated that the retrieval of recent (relative to remote) autobiographical memories principally activated the left dorsolateral prefrontal cortex whereas the retrieval of remote (relative to recent) autobiographical memories activated the inferior parietal cortex bilaterally. ROIs analysis revealed more hippocampal activity for remote memories than for recent ones and a preferentially right-sided involvement of the hippocampal responses whatever the remoteness of autobiographical memories. New insights based on higher hippocampal response to the remoteness of episodic autobiographical memories challenge the standard model and are less discrepant with the multiple trace theory.

Conclusions about the duration of hippocampal contributions to our autobiographical record of personal episodes have come under intense scrutiny in recent years. Interpretation is complicated by such factors as extent and site of lesions as well as test sensitivity. We describe the case of an amnesic person, K.C., with large, bilateral hippocampal lesions who figured prominently in the development of theories of remote memory due to his severely impoverished autobiographical memory extending across his entire lifetime. However, the presence of lesions in higher-order visual cortex raises the possibility that K.C.’s retrograde autobiographical amnesia is mediated by loss of long-term visual images, whereas widespread frontal lesions suggest that his impairment may relate to deficits in strategic retrieval rather than storage. Normal performance on an extensive battery of visual imagery tests refutes the imagery loss interpretation. To test for deficits in strategic retrieval, we used a more formal autobiographical memory test requiring generation of personal events under varying levels of retrieval support. However, even with rigorous contextual prompting, K.C. produced few pre-injury recollections; all were schematic, lacking the richness of detail produced by control participants, raising doubt that his deficit is one of retrieval. Findings are discussed in the context of theories concerning the duration of hippocampal-neocortical interactions in supporting autobiographical re-experiencing. The approach we used to investigate the effects of different lesions on memory provides a framework for dealing with other patients who present with an interesting functional deficit whose neuroanatomical source is difficult to specify due to widespread lesions.