Patterns of impairment in autobiographical memory in the degenerative dementias constrain models of memory

https://doi.org/10.1016/j.neuropsychologia.2006.01.030Get rights and content

Abstract

Detailed study of the autobiographical memory (ABM) impairments seen in different forms of degenerative dementia, in particular Alzheimer's disease (AD) and semantic dementia (SD) can inform neuropsychological models of memory. A modified ABM questionnaire which allowed more detailed analysis of episodic and semantic ABM was used to study the pattern of deficits in patients with minimal to mild Alzheimer's disease (AD) and in two patients with mild and moderate semantic dementia (SD). The questionnaire tested both cued and free recall. A group of healthy elderly was also tested. AD patients differed from controls in all measures. There was no clear temporal gradient for episodic ABM, but a modest gradient was observed for semantic ABM. The mild SD patient performed at control level for episodic ABM but showed a deficit within the range of the AD patients for semantic ABM except for the most recent life period. In contrast the moderate SD patient was impaired within the range of the AD patients for both episodic and semantic ABM. The evidence for differential impairment of episodic and semantic ABM retrieval in AD and SD is interpreted as supporting the multiple trace model of memory.

Introduction

The study of autobiographical memory (ABM) deficits has made a significant contribution to the evidence supporting neuropsychological models of memory. ABM refers to memory for one's own past, and includes memories of both an episodic and semantic nature (Conway & Pleydell-Pearce, 2000). Episodic ABMs are incidents that happened to an individual in the past and that can be recalled in detail and re-experienced “as if one were there”. Semantic ABMs are facts related to one's life such as an address, former jobs, names of friends, etc. that can be retrieved from a “database” (Tulving, 1999). The cognitive processes implementing ABM have usually been thought capable of interpretation in terms of the standard model of memory function (Squire & Alvarez, 1995). The standard model proposes that mediotemporal structures, especially the hippocampus, support memory encoding with a subsequent period of ‘consolidation’ of memories involving a distributed network of neocortical structures. Over time memories are transferred to and stored in the neocortex and the mediotemporal structures are no longer necessary in the recollection process. In this account, therefore, lesions of the mediotemporal structures result in temporally graded amnesia with older memories of any type better preserved than recent ones. The standard model, however, cannot easily account for the empirical observation of dissociated semantic and episodic memory impairments (Tulving, 1999). An alternative account of memory function is offered by the multiple trace model (Nadel & Moscovitch, 1997). This model postulates that the retrieval of old memories will be carried out by different brain structures depending on whether they are of an episodic or semantic type. Within such a framework, the process of recovery of older episodic memories occurs by way of a reconstructive interaction between the hippocampus and the neocortex, while semantic memories are ‘corticalised’ through the creation of multiple memory traces, which are then repeatedly activated.

These theories about memory can be tested by studying the patterns of memory impairments presented by patients with different forms of degenerative dementia, in particular Alzheimer's disease (AD) and semantic dementia (SD). In AD the mediotemporal regions are damaged early in the course of the disease and a typical anterograde amnesic syndrome appears (Braak & Braak, 1991). In SD the most severe damage involves the inferior temporal gyrus and anterior temporal pole. The more medial temporal lobe structures are thought to be less severely affected with the capacity to encode events in everyday life relatively preserved (Garrard & Hodges, 2000; Graham & Hodges, 1997). Volumetric MRI studies, however, have shown that the hippocampal formation is atrophic in both degenerative disorders, but asymmetrically (left worse than right) in SD (Galton et al., 2001), with possible preservation of posterior hippocampal structures (Chan et al., 2001; Chan, Fox, & Rossor, 2002).

A few studies have investigated impairment of ABM retrieval in degenerative dementia using a variety of methods (Dall’Ora, Della Sala, & Spinnler, 1989; Della Sala, Laiacona, Spinnler, & Trivelli, 1993; Graham & Hodges, 1997; Greene, Hodges, & Baddeley, 1995; Kopelman, 1989; Kopelman, Stanhope, & Kingsley, 1999; Nestor, Graham, Bozeat, Simons, & Hodges, 2002; Piolino et al., 2003b; Sagar et al., 1988, Sagar et al., 1991; Westmacott, Leach, Freedman, & Moscovitch, 2001). Studies of AD have shown a deficit in the retrieval of ABM with a shallow temporal gradient indicating more successful retrieval of earlier memories (Greene et al., 1995, Kopelman, 1989, Piolino et al., 2003b, Sagar et al., 1988). Across studies, however, ABM type (episodic/semantic) could not be consistently associated with a distinctive gradient profile. Other studies have interpreted the ABM deficit in AD as a consequence of executive impairments affecting organisation and memory search strategies rather than loss of stored memory representations (Dall’Ora et al., 1989, Della Sala et al., 1993, Kopelman, 1991). Investigations of ABM in SD patients have reported an inverse temporal gradient, with better preservation of memories from more recent years and loss of very old memories (Graham & Hodges, 1997; Graham, Kropelnicki, Goldman, & Hodges, 2003; Nestor et al., 2002; Snowden, Griffiths, & Neary, 1996), although whether this pattern is truly representative of the ABM deficit in SD has been challenged by more recent reports (Moss, Kopelman, Cappelletti, de Mornay Davis, & Jaldow, 2003; Piolino et al., 2003b, Westmacott et al., 2001).

In the face of these anatomical and neuropsychological uncertainties, the detailed study of patterns of ABM impairment in AD and SD might help clarify the role played by temporal and mediotemporal structures in the encoding and retrieval of ABM, and help test the explanatory power of proposed relationships between brain structures and memory processes in these respects.

The Autobiographical Memory Interview (AMI) is the instrument most widely used to obtain a general evaluation of ABM and to distinguish episodic from semantic ABM (Kopelman, Wilson, & Baddeley, 1989; Kopelman, Wilson, & Baddeley, 1990). The AMI is a valuable instrument for case assessment, especially for patients with focal lesions. The test is perhaps less suitable for the assessment of progressive neurodegenerative disorders. In particular, the presence of insidious encoding problems might introduce a bias in the evaluation of the most recent time period resulting in a spurious temporal gradient. Further, the distinction between ‘true’ context rich, episodic memories and more ‘general’ memories, which are schemata or summaries of many episodes (Conway, 2001), might be important in the evaluation of ABM to disclose more fine-graded dissociations. The ‘true’ episodes should ideally reflect episodic memory ability more accurately since more general memories may rely largely on semantic information. Finally, there is evidence that executive deficits might hamper the retrieval of ABM independent of any overt memory deficit, and it is important that this potential confound is addressed by any method of assessment of ABM. This is particularly relevant in the study of ABM in different neurodegenerative disorders which might differ with respect to executive impairment.

The aim of this study was to investigate whether episodic and semantic ABM would dissociate in AD and SD patients using a new questionnaire based on the general framework of the Kopelman et al. AMI (1990). The AMI allows a scored distinction between personal context rich incidents and semantic knowledge, but not a finer grained scoring of more general as against ‘true’ detailed incidents. The modifications were designed therefore, to enhance discrimination within episodic ABM using specific prompts designed to elicit more detailed recollection, to provide a finer temporally graded assessment of ABM and to evaluate ABM retrieval so far as possible, independent of any contribution from concurrent executive deficits by comparing ABM retrieval in cued and free recall.

Section snippets

Participants

A group of AD patients, a group of healthy elderly controls and two SD patients participated in this study. The 20 right-handed AD patients (12 male and eight female) had a mean age of 76 years (S.D. 7) and a mean education of 10 years (S.D. 1). All patients fulfilled the NINCDS-ADRDA clinical criteria for probable AD (McKhann et al., 1984). Their mean Mini Mental State Examination (MMSE) score was 22 (S.D. 3, range 16–26). They were subdivided into a minimal subgroup (mean MMSE 24.67, S.D.

Results

The AD group was significantly older and less educated than the control group. A significant age difference was also found between the two AD subgroups of patients. Age and education were, therefore, included as covariates in all analyses. The two SD patients were compared with the AD and control groups using a computerised version of the Sokal and Rolph (1981) modified t-test (SINGLIMS, http://www.abdn.ac.uk/∼psy086/dept/psychom.htm). Each SD patient's scores were compared to those of the AD

Discussion

This study found that patients with AD and SD showed different patterns of impairment when their episodic and semantic ABM performances were independently assessed. AD patients had a more marked deficit in the recall of specific episodes from their past than in the recall of semantic ABM. The patient in the mild stage of SD showed a substantial impairment in recalling semantic ABM information but episodic memories were better preserved. The moderate SD patient showed pronounced impairment of

Acknowledgments

The authors thank Sergio Della Sala for early discussion of this study. This study was funded by grant no: 066349/Z/01/Z from The Wellcome Trust to AV. AI was funded by “Fondation Saint Luc”, the Saint Luc University Clinic, Brussels, Belgium. JMC was funded by a Royal Society of Edinburgh/Lloyds TSB studentship. AV and JMC were employed by the University of Aberdeen and MFS by Grampian Primary Care Trust when data for this study were collected. Preliminary findings from this study were

References (42)

  • A. Baddeley et al.

    Doors and people

    (1994)
  • H. Braak et al.

    Neuropathological stageing of Alzheimer-related changes

    Acta Neuropathologia

    (1991)
  • D. Chan et al.

    Differing patterns of temporal atrophy in Alzheimer's disease and semantic dementia

    Neurology

    (2002)
  • D. Chan et al.

    Patterns of temporal lobe atrophy in semantic dementia and Alzheimer's disease

    Annals of Neurology

    (2001)
  • M.A. Conway

    Sensory-perceptual episodic memory and its context: autobiographical memory

    Philosophical Transactions of the Royal Society of London Series B-Biological Sciences

    (2001)
  • M.A. Conway et al.

    The construction of autobiographical memories in the self-memory system

    Psychology Review

    (2000)
  • S. Della Sala et al.

    Autobiographical recollection and frontal damage

    Neuropsychologia

    (1993)
  • B.H. Dritschel et al.

    Autobiographical fluency: a method for the study of personal memory

    Memory and Cognition

    (1992)
  • A.W. Ellis et al.

    Real age-of-acquisition effects in lexical retrieval

    Journal of Experimental Psychology-Learning Memory and Cognition

    (1998)
  • C.J. Galton et al.

    Differing patterns of temporal atrophy in Alzheimer's disease and semantic dementia

    Neurology

    (2001)
  • P. Garrard et al.

    Semantic dementia: clinical, radiological and pathological perspectives

    Journal of Neurolinguistics

    (2000)
  • Cited by (70)

    • Episodic and autobiographical memory: Function, dysfunction and behavioral evaluation

      2021, Encyclopedia of Behavioral Neuroscience: Second Edition
    View all citing articles on Scopus
    View full text