Cognitive Impairment in Preclinical Alzheimer's

Neuropsychology 2005, Vol. 19, No. 4, 520 –531

Copyright 2005 by the American Psychological Association 0894-4105/05/$12.00 DOI: 10.1037/0894-4105.19.4.520

Cognitive Impairment in Preclinical Alzheimer’s Disease: A Meta-Analysis Lars Ba¨ckman

Sari Jones, Anna-Karin Berger, and Erika Jonsson Laukka

Karolinska Institute, Stockholm Gerontology Research Center, and Max Planck Institute for Human Development

Karolinska Institute and Stockholm Gerontology Research Center

Brent J. Small University of South Florida To determine the size of the impairment across different cognitive domains in preclinical Alzheimer’s disease (AD), a meta-analysis based on 47 studies involving 9,097 controls and 1,207 preclinical AD cases was conducted. There were marked preclinical deficits in global cognitive ability, episodic memory, perceptual speed, and executive functioning; somewhat smaller deficits in verbal ability, visuospatial skill, and attention; and no preclinical impairment in primary memory. Younger age (⬍ 75 years) and shorter follow-up intervals (⬍ 3 years) were associated with larger effect sizes for both global cognitive ability and episodic memory. For global cognitive ability, studies that used population-based sampling yielded larger effect sizes; for episodic memory, larger differences were seen in studies that preidentified groups in terms of baseline cognitive impairment. Within episodic memory, delayed testing and recall-based assessment resulted in the largest effect sizes. The authors conclude that deficits in multiple cognitive domains are characteristic of AD several years before clinical diagnosis. The generalized nature of the deficit is consistent with recent observations that multiple brain structures and functions are affected long before the AD diagnosis. Keywords: preclinical, Alzheimer’s disease, cognition, dementia, meta-analysis, moderating factors

Almkvist, & Ba¨ckman, 1997), attention (e.g., Linn, Wolf, Bachman, & Knoefel, 1995), and perceptual speed (e.g., Fabrigoule et al., 1998). Given this apparent breadth of impairment, it is only logical that similar deficits have been observed for global indicators of cognitive functioning, such as the Mini-Mental State Examination (MMSE; e.g., B. J. Small, Viitanen, & Ba¨ckman, 1997) and composite measures of cognitive ability (e.g., Fabrigoule et al., 1998). Knowledge pertaining to the preclinical phase of AD is important for theoretical and clinical reasons alike. Theoretically, knowledge regarding the transition from normal aging to dementia is vital in furthering an understanding of how the disease evolves. From a clinical perspective, identifying individuals at risk for developing AD as early as possible is imperative for maximizing treatment efficacy (Flicker, 1999; Post, 1999). Although multiple cognitive domains are affected in preclinical AD, several studies have suggested that tasks assessing episodic memory (e.g., word recall, face recognition) may be particularly effective at identifying at-risk individuals (Elias et al., 2000; B. J. Small, Herlitz, et al., 1997; Tierney, Szalai, Snow, Fisher, Nores, et al., 1996). These observations are consistent with findings that brain structures known to be critical to episodic remembering, such as the hippocampus and neighboring regions (Nyberg, McIntosh, Houle, Nilsson, & Tulving, 1996; Vargha-Khadem et al., 1997), show pathological alterations long before clinical diagnosis (Braak & Braak, 1995; Fox et al., 1996; S. A. Small, Perara, DeLaPaz, Mayeux, & Stern, 1999). A recurrent debate in the literature on episodic memory in preclinical AD has centered on the issue of whether some episodic memory tasks (e.g., free recall) are more effective than others (e.g., recognition) in signaling an impending

During the past decade, considerable evidence has accumulated indicating that individuals who will develop Alzheimer’s disease (AD) exhibit cognitive deficits several years before a clinical diagnosis of dementia can be rendered (e.g., Chen et al., 2001; Elias et al., 2000; B. J. Small, Fratiglioni, Viitanen, Winblad, & Ba¨ckman, 2000; for an overview, see Ba¨ckman, Small, & Fratiglioni, 2004). This is evidenced by a performance decrement at a baseline assessment point among persons who are currently without dementia but who will receive an AD diagnosis after a followup interval. Such preclinical deficits have been observed across multiple cognitive domains, including episodic memory (e.g., Ba¨ckman, Small, & Fratiglioni, 2001), executive functioning (e.g., Albert, Moss, Tanzi, & Jones, 2001), verbal ability (e.g., Jacobs et al., 1995), visuospatial skill (e.g., B. J. Small, Herlitz, Fratiglioni,

Lars Ba¨ckman, Aging Research Center, Division of Geriatric Epidemiology, Neurotec, Karolinska Institute, Stockholm, Sweden; Stockholm Gerontology Research Center, Stockholm, Sweden; and Center for Lifespan Psychology, Max Planck Institute for Human Development, Berlin, Germany. Sari Jones, Anna-Karin Berger, and Erika Jonsson Laukka, Aging Research Center, Division of Geriatric Epidemiology, Neurotec, Karolinska Institute and Stockholm Gerontology Research Center. Brent J. Small, School of Aging Studies, University of South Florida. Preparation of this article was supported by grants from the Swedish Research Council and the Swedish Council for Working Life and Social Research to Lars Ba¨ckman. A complete list of the articles included in this meta-analysis is on the Web at http://dx.doi.org/10.1037/0894-4105.19.4.520.supp. Correspondence concerning this article should be addressed to Lars Ba¨ckman, Aging Research Center, Karolinska Institute, Box 6401, S-113 82, Stockholm, Sweden. E-mail:[email protected] 520

COGNITIVE IMPAIRMENT IN PRECLINICAL AD

dementia disease (Ba¨ckman & Small, 1998; Howieson et al., 1997; B. J. Small, Herlitz, et al., 1997). However, the primacy of episodic memory as an early harbinger of AD has not been assessed quantitatively using all available information. In addition, the extent to which the degree of preclinical cognitive impairment in AD is influenced by various study and task characteristics remains unknown. In this article, we report a meta-analysis that synthesizes the empirical research on cognition in preclinical AD. Meta-analytic techniques allow for quantitatively combining data from several studies, thereby increasing the statistical power. Another advantage is that individual studies are given differential weights depending on sample size. Finally, by pooling the results from multiple studies, it is possible to provide more exact information concerning the influence of various moderating variables on effect sizes (Cohn & Becker, 2003; Hedges & Olkin, 1985). Our goal in the present study was threefold. First, we sought to determine the magnitude and pattern of impairment across multiple cognitive domains in preclinical AD. Thus, baseline cognitive performance of individuals who would and would not develop AD across a follow-up period constituted the outcome measure in the current analysis. The following domains were identified: global cognitive ability, episodic memory, executive functioning, verbal ability, visuospatial skill, attention, perceptual speed, and primary memory. Of chief interest was whether there would be systematic differences among different cognitive domains in the size of the preclinical impairment. Second, we examined the influence of study characteristics on the pattern of outcome in tests of global cognitive ability and episodic memory. These analyses involved comparisons across (a) participant age, (b) the time between preclinical assessment and diagnosis, and (c) the nature of the study sample. With regard to the latter, we examined differences between studies that used population-based versus convenience samples, as well as between studies in which at-risk individuals were identified a priori on the basis of some classificatory system (e.g., mild cognitive impairment, or MCI) versus those in which no such procedures were adopted. A third goal was to determine whether there were differences in effect size as a function of task characteristics (i.e., retention interval, retrieval demands, nature of the learning materials) within the key domain of episodic memory.

Method Search Strategy and Selection Criteria Articles published between January 1985 and February 2003 were identified through an extensive literature search of online databases (MEDLINE and PsycINFO). The year 1985 was chosen as a starting point because it corresponds to the introduction of more systematic and reliable diagnostic criteria for AD (McKhann et al., 1984). The search was performed in March 2003 and was limited to published English-language articles with human participants. The keywords used were Alzheimer’s disease, Alzheimer disease, Alzheimer’s, Alzheimer, or AD in combination with preclinical, subclinical, prodromal, prediagnostic, pre-diagnostic, presymptomatic, pre-symptomatic, early stages, early symptoms, early diagnosis, early detection, mild cognitive impairment, MCI, cognitive impairment no dementia, CIND, conversion, prediction, transition, incidence AND cognition, incidence AND longitudinal, or incidence AND follow-up. In addition, we examined the tables of contents of relevant journals published within the 12 months prior to the completion of the literature review in order to minimize the possibility of overlooking studies

521

that may not yet have been included in computerized databases. The journals that were searched were American Journal of Epidemiology; American Journal of Psychiatry; Annals of Neurology; Archives of General Psychiatry; Archives of Neurology; Brain; Brain and Cognition; British Journal of Psychiatry; International Journal of Geriatric Psychiatry; Journal of the American Geriatrics Society; Journal of the American Medical Association; Journal of Clinical and Experimental Neuropsychology; Journal of the International Neuropsychological Society; Journals of Gerontology, Series B: Psychological Sciences and Social Sciences; Journals of Gerontology, Series A: Biological Sciences and Medical Sciences; Journal of Neurology, Neurosurgery, and Psychiatry; Neurology; Neuropsychology; Psychology and Aging; and Psychological Medicine. Studies included in the meta-analysis had to satisfy several criteria. These criteria were implemented to ensure that the studies were sufficiently similar to justify their contribution to the current analysis. First, participants had to be free of clinical dementia at a baseline assessment point, with a portion of the sample receiving, after a follow-up period, a diagnosis of AD according to standardized clinical criteria (e.g., National Institute of Neurological and Communication Disorders and Stroke—Alzheimer’s Disease and Related Disorders Association [NINCDS–ADRDA; McKhann et al., 1984], the revised third edition of the Diagnostic and Statistical Manual of Mental Disorders [DSM–III–R; American Psychiatric Association, 1987], the fourth edition of the DSM [DSM–IV; American Psychiatric Association, 1994], and the 10th edition of the International Statistical Classification of Diseases and Related Health Problems [ICD-10; World Health Organization, 1992]). Note that we only selected studies in which AD was the diagnostic outcome. Studies that used dementia as the outcome were excluded because of the potential heterogeneity of this category. Second, all participants must have received an assessment of cognitive functioning at baseline. Finally, the studies had to include sufficient statistical information to allow for effect sizes to be calculated, which could include means and standard deviations, exact p values, t values, or F values, and sample sizes. In cases in which this information was not originally reported, we contacted the study authors directly to obtain the relevant statistics. In total, we obtained results from 50 studies that met the criteria for inclusion in the present meta-analysis. However, in some cases, the retrieved articles came from the same research group and were not derived from completely independent samples. Specifically, studies from the Kungsholmen Project (Ba¨ckman & Small, 1998; Ba¨ckman, Small, & Fratiglioni, 2001; Berger, Fratiglioni, Forsell, Winblad, & Ba¨ckman, 1999; B. J. Small et al., 2000; B. J. Small, Herlitz, et al., 1997; B. J. Small, Viitanen, & Ba¨ckman, 1997), studies from the University of California, San Diego research group (Bondi, Salmon, Galasko, Thomas, & Thal, 1999; Lange et al., 2002; Olichney et al., 2002), and work by Tierney and colleagues (Tierney, Szalai, Dunn, Geslani, & McDowell, 2000; Tierney, Szalai, Snow, & Fisher, 1996; Tierney, Szalai, Snow, Fisher, Nores, et al., 1996; Tierney, Szalai, Snow, Fisher, Tsuda, et al., 1996) exhibited enough similarities, with the exception of differences in the numbers of participants, to raise concerns about the independence of the study samples. In these cases, the inclusion of studies was determined by whether each study contributed unique information. As a result, three studies (Berger et al., 1999; Tierney, Szalai, Snow, & Fisher, 1996; Tierney, Szalai, Snow, Fisher, Tsuda, et al., 1996) were excluded. All remaining studies contributed unique information. Thus, a total of 47 studies contributed data to the meta-analysis, comprising a total of 1,207 preclinical AD cases and 9,097 controls. The basic characteristics of each of the studies included are shown in Table 1. A full reference list of all studies included in the meta-analysis is provided on the Web at http://dx.doi.org/10.1037/0894-4105.19.4.520.supp.

Cognitive Outcome Measures The outcome measure was baseline cognitive performance for incident AD cases and controls. An inspection of the studies included revealed that

¨ CKMAN, JONES, BERGER, LAUKKA, AND SMALL BA

522

Table 1 Characteristics of the 47 Studies Included in the Meta-Analysis n Author

Year published

Cognitive ability domain(s) examined

Albert et al. Andreasen et al. Arnaíz et al. Ba¨ckman & Small Ba¨ckman et al. Bacon et al. Bizdan & Bizdan Bondi et al. Chen et al. Convit et al. Daly et al. de Leon et al. Devanand et al. Fabrigoule et al. Fowler et al. Fox et al. Grober & Kawas Hall et al. Howieson et al. Huang et al. Jacobs et al. Jacobson et al. Jelic et al. Kogure et al. Lange et al. Lindeboom et al. Lindsay et al. Linn et al. Meyer et al. Nielson et al. Okamura et al. Olichney et al. Riemenschneider et al. Rubin et al. Schaub et al. Schmand et al. Small, Herlitz, et al. Small, Viitanen, et al. Small et al. Smith et al. Tabert et al. Tierney et al. Tierney et al. Tro¨ster et al. Tuokko et al. Visser et al. Wolf et al.

2001 1999 2001 1998 2001 1998 2002 1999 2000 2000 2000 1993 2000 1998 2002 1998 1997 2000 1997 2000 1995 2002 2000 2000 2002 2002 2002 1995 2002 1999 2002 2002 2002 1998 2002 2000 1997 1997 2000 1998 2002 1996 2000 1994 1991 2001 2000

AT, EM, EX, PM, PS, VA, VS GC GC PM EM, PM GC GC EM AT, EM, EX, GC, VA, VS EM, GC, VA EM, EX GC GC EM, GC, PS, VA EM, GC, VA, VS EM, GC, PS, VA EM, GC, VA EM EM, GC, PM, VA, VS GC EM, VA EM, VA, VS GC EM, GC, PM, VS GC GC GC EM, PM, VA GC AT, EM, GC, VS GC GC EM, GC AT, EM, PM, PS, VA, VS GC EM, GC GC EM, PM, VA, VS GC GC GC AT EM, GC, VA GC EM EM, GC, VA GC

Preclinical Diagnostic Age AD Controls criteria (years) 23 16 9 24 15 8 19 7 120 14 23 25 19 16 9 10 20 35 16 14 41 20 14 32 20 24 30 55 35 102 8 7 10 16 17 25 32 26 71 35 23 29 29 4 18 23 8

32 15 11 134 105 62 185 86 483 32 100 47 58 1134 31 53 60 293 31 17 402 20 16 45 84 204 308 990 23 2307 22 6 10 55 17 169 189 179 457 215 69 94 98 59 27 44 19

NIH DSM/NIH DSM/NIH DSM DSM DSM/NIH DSM DSM/NIH DSM/NIH DSM NIH DSM/NIH DSM/NIH DSM/NIH NIH NIH NIH DSM/NIH NIH NIH DSM/NIH NIH NIH DSM/NIH DSM/NIH DSM/NIH DSM NIH DSM DSM/NIH NIH NIH NIH NIH DSM/NIH DSM DSM DSM DSM DSM/NIH NIH DSM/NIH DSM/NIH DSM/NIH DSM/NIH DSM/NIH DSM/ICD

71.8 71.5 62.3 82.8 81.2 74.2 75.5 70.8 75.6 72.4 71.8 71.1 65.7 72.9 59.0 44.7 79.1 80.3 86.8 62.2 73.3 75.2 59.1 69.8 74.6 73.3 73.3 72.5 73.7 72.8 71.4 74.3 69.5 71.7 — 73.5 83.5 83.5 80.6 79.8 67.2 72.7 71.7 67.2 68.7 68.9 72.0

Follow-Up interval Participant Sampling (years) characteristics method 3 1.1 3.2 2.9 6.7 — 5 3.7 1.5 3.2 3 3.9 1.6 2 2 2.6 4 2.2 2.8 2.1 2.1 1.3 1.8 2 2 2 5 6 3.9 3.9 3.1 2 1.5 1.5 3 3 3.1 3.1 6.7 3.5 2.0 2 2 5 1.3 5 2.4

intact MCI MCI intact intact intact cimprd intact intact cimprd cimprd cimprd cimprd intact intact/cimprd intact intact intact intact MCI intact intact MCI cimprd intact intact intact intact MCI intact MCI MCI MCI intact intact intact intact intact intact MCI MCI cimprd cimprd intact intact MCI MCI

C C C P P C C C P C C C C P C C P C C C P C C C C P P P C P C C C C C P P P P C C C C C C C C

Note. A complete list of the articles included in this meta-analysis can be found on the Web at http://dx.doi.org/10.1037/0894-4105.19.4.520.supp. Cognitive ability domains: AT ⫽ attention; EM ⫽ episodic memory; EX ⫽ executive functioning; GC ⫽ global cognitive ability; PS ⫽ perceptual speed; PM ⫽ primary memory; VA ⫽ verbal ability; VS ⫽ visuospatial skill. Diagnostic criteria: DSM ⫽ DSM-III-R (Diagnostic and Statistical Manual of Mental Disorders, 3rd ed., revised; American Psychiatric Association, 1987) or DSM-IV (4th ed.; American Psychiatric Association, 1994); ICD ⫽ ICD-10 (International Statistical Classification of Diseases and Related Health Problems, 10th ed.; World Health Organization, 1992); NIH ⫽ National Institutes of Health NINCDS–ADRDA (National Institute of Neurological and Communication Disorders and Stroke—Alzheimer’s Disease and Related Disorders Association) criteria (McKhann et al., 1984). Participant characteristics: intact ⫽ not classified with cognitive impairment; MCI ⫽ mild cognitive impairment; cimprd ⫽ cognitively impaired but no dementia. Sampling method: C ⫽ convenience sample; P ⫽ population-based sample.

there was sufficient information on enough individual cognitive tests that they could be organized into several broad domains of functioning. The tasks were organized into eight categories of cognitive ability on the basis of the typical association between tests and ability domains seen in the neurologic and neuropsychological literature (e.g., Lezak, 1995; Spreen &

Strauss, 1998). The specific domains identified (and associated tests) were as follows: global cognitive ability (e.g., the MMSE [Folstein, Folstein, & McHugh, 1975], the Cognitive test section of the Cambridge Examination for Mental Disorders of the Elderly [Roth, Huppert, Mountjoy, & Tym, 1988]); episodic memory (e.g., the California Verbal Learning Test [Delis,

523

COGNITIVE IMPAIRMENT IN PRECLINICAL AD Kramer, Kaplan, & Ober, 1987], the Benton Visual Retention Test [Benton, 1974], the Logical Memory subtest from the Wechsler Memory Scale [WMS; Wechsler, 1945], the Recall subtest of the Rey–Osterrieth Complex Figure test [Rey, 1941]); executive functioning (e.g., the Trail Making Test–Part B [Reitan, 1958], the Stroop interference task [Stroop, 1935], the alphabet span task [Craik, 1986]); verbal ability (e.g., the Controlled Oral Word Association Test [Benton & Hamsher, 1983], the Boston Naming Test [Kaplan, Goodglass, & Weintraub, 1983], the Vocabulary subtest from the Wechsler Adult Intelligence Scale—Revised [WAIS–R; Wechsler, 1981]); visuospatial skill (e.g., the Block Design subtest from the WAIS–R [Wechsler, 1981], the Copy subtest of the Rey–Osterrieth Complex Figure test [Rey, 1941], the clock drawing task [Christensen, 1984]); attention (e.g., the Trail Making Test–Part A [Reitan, 1958], the Mental Control subtest from the WMS [Wechsler, 1945]); perceptual speed (e.g., the Digit Symbol subtest from the WAIS–R [Wechsler, 1981], the letter cancellation task [Zazzo, 1974]); and primary memory (e.g., the Digit Span–Forward subtest from the WAIS–R [Wechsler, 1981], the primary memory score from the lag method [Tulving & Colotla, 1970]).

Data Extraction and Statistical Analysis The system for the meta-analysis was based on a random-effects procedure, as described by Hedges and Olkin (1985). From the data reported in each study, the effect-size estimate, g, was calculated, indicating the difference between the control and incident AD groups divided by the pooled standard deviation. Thus, g represents the standardized difference between the two groups within each study. In order to pool the results across studies, the effect size, d, was computed, which represents the standardized difference between the preclinical AD and nondementia groups weighted by the sample sizes of the individual studies. The weighting procedure affects the variance estimate from each study because, as Hedges and Olkin (1985) noted, “the variance estimates for studies with larger sample sizes are more precise than those for studies with smaller effect sizes” (p. 107). For both g and d, the direction of the effect size was positive if the performance of the incident AD group was worse than that of the control participants. In studies in which more than one dependent measure was present for a cognitive domain (e.g., multiple tests of episodic memory), an averaged effect size was calculated for the overall analysis to avoid having one study dominate the results (Heinrichs & Zakzanis, 1998). For example, if a study had tests of immediate and delayed episodic memory, these effect sizes were averaged to generate the overall effect size for episodic memory. However, when examining effect sizes as a function of retention interval, we treated the two tests separately. We also calculated the chi-square statistic Q, indicating homogeneity of results across effect sizes. If a significant Q value is observed, this indicates heterogeneity of results and may result in a search for potential moderating variables. In the moderator analysis, the QW statistic indicates the degree of homogeneity within a class of studies or a cognitive domain. The QB statistic refers to a test of differences between categories and, if statistically significant, suggests the influence of a moderating variable. Because of the large sample sizes in this study, an alpha level of .01 was adopted as the statistical criterion. The degree of overlap between the control and incident AD groups was determined using the U statistic (Heinrichs & Zakzanis, 1998). The overlap statistic (O/L%) represents the degree of overlap in the distributions of the cognitive scores for the incident AD and control groups. This statistic allows one to determine whether the two groups are clearly separable.

tics, we were interested in the effect of participant age (⬍ 75 years, ⱖ 75 years), length of follow-up period (⬍ 3 years, ⱖ 3 years), sampling method (population based, convenience), and initial cognitive status of participants (intact, impaired). For age and length of follow-up, the division was based on a median split across the studies included. With respect to sampling method, population-based studies were those that used epidemiologicbased methods to recruit persons from a specified population (e.g., Fabrigoule et al., 1998; B. J. Small, Herlitz, et al., 1997). By contrast, convenience samples included persons who were selected from clinical settings (e.g., Andreasen et al., 1999; Devanand et al., 2000) or recruited through appeals in the media or other public outlets (e.g., Grober & Kawas, 1997; Howieson et al., 1997). In terms of initial cognitive status, participants were considered impaired if they had been classified with cognitive dysfunction (i.e., MCI [e.g., Petersen et al., 1999] or memory impairment [e.g., Kogure et al., 2000]) at baseline. In addition to study characteristics, we examined the extent to which group differences in episodic memory functioning varied as a function of task characteristics. Specifically, we were interested in the effect of retention interval (immediate, delayed), study materials (verbal, nonverbal), and retrieval support (recall, recognition) on the magnitude of observed effects.

Results Overall Effect Sizes A concern in meta-analytic studies is the potential existence of publication bias and other biases related to the process of locating, selecting, and combining studies (Easterbrook, Berlin, Gopalan, & Matthews, 1991). One method for identifying outliers, and the possibility of biases, is the use of funnel plots, which assess the relationship between sample size and effect size. Using funnel plots (Egger, Davey Smith, Schneider, & Minder, 1997; Light & Pillemer, 1984), we addressed this issue for global cognitive ability and episodic memory, the domains that had the largest numbers of observations. As shown in Figure 1, for global cognitive ability, the majority of observations yielded g values between 0.5 and 2.0. A similar pattern was found for episodic memory, with g values ranging from 0.5 to 3.0. Note that the effect-size measures shown in Figure 1 are indexed by g, because data from individual studies are being plotted. The distributions approximated a funnel shape, with small-sample studies scattered at the bottom of the graphs and the peak converging on the average effect size for both cognitive domains. Thus, the shape of the plots provides no evidence for systematic bias within either cognitive domain.

Moderator Variables In addition to overall effect sizes, we examined the influence of several potential moderating variables, using categorical models. The moderator variables used were either at the level of study characteristics or in the nature of the episodic memory tests employed. Among study characteris-

Figure 1. Funnel plots of sample size and effect size (g) for global cognitive ability and episodic memory.


524

Effect sizes for the eight cognitive domains are shown in Table 2. The incident AD group exhibited statistically significant deficits across all domains except primary memory. Global cognitive ability, episodic memory, perceptual speed, and executive functioning yielded the largest effect sizes, all above 1.0. Significant group differences were also observed for verbal ability, visuospatial skill, and attention, with d values ranging from 0.62 to 0.79. The overlap statistic is also shown in Table 2. Of special note here is that even for global cognitive ability, episodic memory, perceptual speed, and executive functioning, almost half of the distributions between the control and incident AD groups overlapped. The QW statistic revealed significant heterogeneity of effect sizes for all domains except executive functioning. In terms of potential moderators of this variability, we examined only global cognitive ability and episodic memory. We did so for two principal reasons. First, the number of studies that included measures of these domains was sufficient to permit a more fine-grained analysis. Second, the a priori categorization of episodic memory tasks, in terms of retention interval, study materials, and retrieval support, lent itself to examining these factors as potential moderators.

Influence of Study Characteristics on Effect Sizes Table 3 shows the influence of the moderator variables on effect sizes for global cognitive ability and episodic memory. For both global cognitive ability and episodic memory, age and follow-up interval were associated with the magnitude of group differences. Specifically, younger age and shorter follow-up intervals resulted in larger differences between the incident AD and control groups. Studies in which population-based sampling was used revealed larger effect sizes for global cognitive ability, whereas there was a trend in the same direction for episodic memory ( p ⬍ .05). Finally, for episodic memory, studies in which participants were preselected on the basis of cognitive impairment at baseline yielded larger effect sizes.

Effect Sizes Within Episodic Memory Table 4 displays the effect sizes for episodic memory as a function of retention interval, retrieval support, and study materials. As evidenced by the QB statistic, the results indicated significant differences in effect size as a function of retention interval

and retrieval condition. Specifically, delayed testing and recallbased retrieval resulted in larger group differences. A trend was seen for study materials, with verbal materials resulting in somewhat larger group differences ( p ⬍ .05).

Discussion The results of this meta-analysis indicate a global impairment of cognitive functioning in the preclinical phase of AD. Specifically, for three of the specific cognitive domains targeted, the overall d values were 1.03 (episodic memory), 1.07 (executive functioning), and 1.11 (perceptual speed), reflecting large effect sizes according to the nomenclature of Cohen (1988). For verbal ability (d ⫽ 0.79), visuospatial skill (d ⫽ 0.64), and attention (d ⫽ 0.62), the group difference was of moderate size, whereas it was nonexistent for primary memory (d ⫽ 0.00). These findings are consistent with claims made by many investigators in this field (e.g., Ba¨ckman, Small, & Fratiglioni, 2001; Elias et al., 2000; Grober & Kawas, 1997; Linn et al., 1995; B. J. Small, Herlitz, et al., 1997) that episodic memory impairment is a cardinal feature of an impending dementia disease, likely reflecting the fact that memory-relevant regions in the medial-temporal lobe are affected prior to the time at which dementia can be clinically diagnosed (e.g., Fox et al., 1996; S. A. Small et al., 1999) At the same time, these results demonstrate that episodic memory does not have a unique status among categories of cognitive markers for identifying forthcoming AD. The confidence intervals around the effect sizes overlapped for episodic memory, executive functioning, and perceptual speed. For verbal ability, visuospatial skill, and attention, the effect sizes were somewhat smaller but still quite sizable. The only domain that deviated markedly from the pattern of clear preclinical impairment was that of primary memory. However, given that basic primary memory operations are well preserved in early clinical AD (e.g., Martin, Brouwers, Cox, & Fedio, 1985; Simon, Leach, Winocur, & Moscovitch, 1995), there is little reason to believe that they would be affected preclinically. The observation that the deficit in preclinical AD generalized across several major domains of cognition was corroborated by the fact that the category of global cognitive ability showed an effect size of 1.19. However, the large effect size obtained for this category does not imply that each cognitive domain represented in global assessment instruments such as the MMSE is equally impaired. For example, B. J. Small, Viitanen, and Ba¨ckman (1997)

Table 2 Effect Sizes for the Eight Domains of Cognitive Functioning Sample size Domain

Number of studies (k)

Controls

Cases

d

99% CI

O/L%

QW

Global Cognitive Ability Episodic Memory Verbal Ability Visuospatial Skill Primary Memory Attention Perceptual Speed Executive Functioning

35 24 15 9 8 5 4 3

6,509 6,694 3,625 3,159 1,567 2,969 1,274 547

868 704 438 363 205 289 65 166

1.19* 1.03* 0.79* 0.64* 0.00 0.62* 1.11* 1.07*

1.15–1.24 0.98–1.08 0.73–0.85 0.58–0.71 ⫺0.09–0.09 0.55–0.68 1.00–1.22 0.92–1.21

38.8 42.1 57.0 61.8 100.0 61.8 42.1 42.1

538.53* 768.85* 391.35* 24.44* 80.10* 20.50* 163.47* 2.45

Note. CI ⫽ confidence interval; O/L% ⫽ overlap statistic. * p ⬍ .01.

525


Table 3 Effect Sizes for Global Cognitive Ability and Episodic Memory Across Study Characteristics Sample size Moderator variable


Controls

Cases

d

99% CI

QW

QB

Global cognitive ability Age of participants ⬍ 75 years ⱖ 75 years Follow-up interval ⬍ 3 years ⱖ 3 years Sampling method Population based Convenience Participant characteristics Intact Impaired

27 8

4,872 1,808

538 347

1.32 0.91

1.26–1.37 0.83–0.99

237.22** 195.68**

110.35**

20 14

2,511 3,936

444 416

1.29 1.12

1.22–1.37 1.06–1.18

212.97** 302.42**

22.83**

9 26

5,311 1,198

440 428

1.22 1.08

1.17–1.27 0.98–1.17

363.77** 163.38**

11.37**

14 21

5,449 1,060

490 378

1.19 1.17

1.14–1.27 1.07–1.27

427.65** 110.67**

0.21

Episodic memory Age of participants ⬍ 75 years ⱖ 75 years Follow-up interval ⬍ 3 years ⱖ 3 years Sampling method Population based Convenience Participant characteristics Intact Impaired

17 7

5,523 1,171

452 252

1.06 0.89

1.01–1.12 0.79–0.99

696.24** 57.33**

15.28**

16 8

5,174 1,520

538 166

1.12 0.76

1.06–1.17 0.67–0.86

605.22** 91.00**

72.62**

9 15

5,824 870

419 285

1.04 0.95

1.00–1.10 0.83–1.06

546.66** 218.01**

4.18*

16 8

6,366 328

548 156

1.02 1.25

0.96–1.06 1.06–1.44

595.89** 163.72**

9.24**

Note. Participant characteristics: Intact ⫽ not classified with cognitive impairment; Impaired ⫽ mild cognitive impairment or cognitively impaired but no dementia. CI ⫽ confidence interval. * p ⬍ .05. ** p ⬍ .01.

demonstrated that three MMSE subscales, Orientation to Time, Orientation to Place, and Delayed Recall, were most effective in differentiating incident AD cases from controls 3 years before diagnosis. The finding of a multiple cognitive system breakdown is consistent with observations that conversion rates to AD over a 2.5year interval were considerably greater for persons who exhibited deficits in episodic memory and some other cognitive domain (e.g.,

verbal ability, visuospatial skill) at baseline than for persons who had isolated memory impairment (Bozoki, Giordani, Heidebrink, Berent, & Foster, 2001). Further, the breadth of cognitive impairment in preclinical AD is interesting to consider in view of the literature on normal cognitive aging. It is well established that episodic memory, executive functioning, and speed not only deteriorate markedly across the adult life span but also show strong associations in samples of individuals without dementia (e.g.,

Table 4 Effect Sizes for Episodic Memory Across Conditions Sample size Condition Retention interval Immediate Delayed Retrieval support Recall Recognition Study material Verbal Nonverbal * p ⬍ .05.

** p ⬍ .01.


Controls

Cases

d

99% CI

QW

QB

15 17

3,356 3,388

411 453

0.96 1.23

0.90–1.02 1.16–1.30

380.17** 516.89**

58.90**

22 6

6,493 937

656 194

1.02 .79

0.97–1.06 0.68–0.90

561.19** 74.92**

22.38**

24 8

6,694 2,476

704 187

1.05 .96

1.00–1.09 0.89–1.02

756.17** 431.70**

6.39*

526


Ba¨ckman, Small, & Wahlin, 2001; Park et al., 2002; Salthouse, 1996). Thus, the fact that the largest effect sizes were observed for these domains indicates continuity in patterns of cognitive impairment from normal aging to preclinical AD. In addition, it is of interest to consider recent brain imaging and histopathological findings comparing controls and incident AD cases. Emerging evidence indicates that multiple brain structures and functions in addition to those related to the medial-temporal lobe may be affected before the AD diagnosis. These findings include volume reductions of anterior cingulate and temporal sulcus (Killiany et al., 2000), posterior cingulate and neocortical temporoparietal regions (Fox et al., 2001), and frontal regions (van der Flier, van den Heuvel, Weverling-Rijnsburger, Bollen, et al., 2002); decreased blood flow in posterior cingulate and precuneus (Kogure et al., 2000); reduced glucose metabolism in temporoparietal regions (Arnaíz et al., 2001); and deposits of amyloid plaques in temporal (Morris & Price, 2001) and frontal (Yamaguchi, Sugihara, Ogawa, Oshima, & Ihara, 2001) cortex. In addition, more general alterations of brain functions in preclinical AD have been observed, including an increase of white-matter hyperintensities (Wolf, Ecke, Bettin, Dietrich, & Gertz, 2000) as well as a reduction of whole-brain glucose metabolism (Silverman et al., 2001). Thus, given that these studies indicate a rather widespread affection of brain structures and functions in preclinical AD, it should come as no surprise that a similar extent of impairment is observed at the behavioral level. This point is further strengthened by recent evidence indicating strong relationships of volumetric measures of the whole brain and the temporal and frontal lobes to cognitive performance (e.g., memory, verbal skill, executive functioning) in MCI and early AD, although no relationships were observed between regional brain damage and impairment of specific cognitive functions (van der Flier, van den Heuvel, WeverlingRijnsburger, Split, et al., 2002). Several interesting findings were obtained regarding the influence of study characteristics on effect sizes. For both global cognitive ability and episodic memory, the degree of impairment was greater (a) for persons younger than 75 years at baseline than for those who were older than 75 years at the outset and (b) for shorter (less than 3 years) than for longer (3 ⫹ years) follow-up periods. The result that differences were most pronounced among individuals who were 75 years of age or younger is consistent with findings in the literature on early clinical dementia indicating a negative relationship between onset age and level of cognitive impairment, findings possibly reflecting more widespread and severe brain lesions among younger cases (e.g., Hansen, De Teresa, Davies, & Terry, 1988; Mann, 1994). The current findings extend these observations to the preclinical phase of the disease. Another factor potentially contributing to this difference is related to the control group. It is well established that cognitive functioning deteriorates from early to later portions of senescence (e.g., Giambra, Arenberg, Kaws, Zonderman, & Costa, 1995; Korten et al., 1997; Lindenberger & Baltes, 1997; R. S. Wilson, Gilley, Bennett, Beckett, & Evans, 2000). Lower cognitive performance among the controls obviously reduces the room for further deterioration in preclinical AD. Not surprisingly, effect sizes were larger when the incident AD cases were examined less than 3 years before diagnosis than when baseline assessment occurred 3 or more years before diagnosis. However, note that the degree of impairment was quite sizable also

for studies that used longer retest intervals, with d values of 1.12 and 0.76 for global cognitive ability and episodic memory, respectively. This pattern of results is interesting in light of longitudinal evidence on the trajectory of cognitive decline in preclinical AD. In two studies, stability of cognitive impairment among incident AD patients was observed from 6 to 3 years prior to eventual diagnosis (Ba¨ckman, Small, & Fratiglioni, 2001; B. J. Small et al., 2000). In another study, disproportionate decline was seen among incident AD cases from 3.5 to 1.5 years before diagnosis (Chen et al., 2001). Thus, the preclinical period in AD appears to be characterized by an early onset followed by relative stability until a few years before diagnosis, when precipitous cognitive decline occurs (e.g., Fox et al., 1996; Rubin et al., 1998; B. J. Small, Viitanen, & Ba¨ckman, 1997). For global cognitive ability, studies in which participants were recruited through population-based sampling revealed larger effect sizes than did studies that used convenience sampling (e.g., clinical settings, advertisements). There was a trend in the same direction for episodic memory. This may reflect the fact that memory and other cognitive problems are common among people who actively seek participation in studies on aging and cognition irrespective of whether they are in a preclinical phase of dementia. As a consequence, memory performance would also be expected to be lower for the controls in these types of studies, resulting in smaller group differences. A key difference between studies in this area concerns whether participants have already been classified as cognitively impaired at the baseline assessment using categories such as MCI. On the one hand, it may appear reasonable that studies that follow a group of preclassified persons prospectively should yield larger effect sizes than studies that work retrospectively from diagnosis to baseline. On the other hand, recent evidence indicates that categories such as MCI are rather heterogeneous. For example, it has been shown that although a large portion of cognitively impaired older adults go on to develop AD within a few years, a sizable portion remain stable or even improve across the same time period (e.g., Helkala et al., 1997; Palmer, Wang, Ba¨ckman, Winblad, & Fratiglioni, 2002). Obviously, the latter fact speaks against prospective studies resulting in larger effect sizes. For global cognitive ability, the confidence intervals for the two types of studies were overlapping, whereas for episodic memory, larger effect sizes were observed for MCI-type studies. The latter finding was expected in view of the fact that episodic memory impairment constitutes a cardinal criterion for inclusion in the MCI category (Petersen et al., 1999). Within the domain of episodic memory, there were two significant effects pertaining to specific task characteristics. First, the effect size was larger for delayed than for immediate testing of memory. Provided that delayed testing taxes consolidation processes to a greater extent than does immediate testing (e.g., Haist, Gore, & Mao, 2001; Squire, 1986), this result is consistent with the view that failure in transferring information from temporary storage to a more permanent memory representation is a characteristic feature of the episodic memory impairment in preclinical AD (Ba¨ckman & Small, 1998; Martin et al., 1985). Second, effect sizes were larger for recall than for recognition. Given that the retrieval demands are typically greater in recall than in recognition (e.g., Craik, 1983; Craik & McDowd, 1987; Jacoby, Toth, & Yonelinas, 1993), these data suggest that retrieval problems, in addition to difficulties in encoding and consolidation, may be characteristic of


preclinical AD. There was also a trend in the direction of larger effect sizes for verbal than for nonverbal materials. This may reflect the fact that verbal materials (e.g., words, paired-associates) are typically poorer in terms of the features available at encoding than are nonverbal materials (e.g., faces, pictures). As a result, the requirement of self-initiated elaborative encoding operations is generally greater for verbal materials, which may be particularly handicapping for preclinical AD cases. In general, then, the moderating effects observed for episodic memory suggest that performance deficits in preclinical AD are exacerbated with increasing cognitive demands. However, despite the observed influences of study and task characteristics, it is important to note that the effect sizes for global cognitive ability and episodic memory were uniformly large. Thus, the differences in effect sizes observed in this meta-analysis should not conceal the fact that the preclinical cognitive impairment in AD was found to generalize across (a) several key domains of cognitive functioning, (b) major characteristics of research studies and sample composition, and (c) multiple aspects of episodic memory. Although the present meta-analysis revealed generally large effect sizes, the overlap in cognitive performance between cases and controls was quite substantial, with overlap scores ranging from 39% to 62% for those task domains yielding reliable group differences. Thus, a sizable proportion of the cases had scores that fell within normal ranges and vice versa. To be sure, it may not be realistic to expect nonoverlapping distributions of cognitive performance scores many years before diagnosis given the large interindividual variability in preclinical AD (e.g., Ba¨ckman, Jones, Small, Aguëro-Torres, & Fratiglioni, 2003; Fox et al., 1996; Palmer et al., 2002) as well among those who will remain without dementia (e.g., Anstey & Christensen, 2000; Ba¨ckman, Small, Wahlin, & Larsson, 1999; Hultsch, Hertzog, Dixon, & Small, 1998). There are numerous reasons for exhibiting cognitive impairment in old age in addition to being in a preclinical phase of AD. These include psychiatric (e.g., Ba¨ckman & Forsell, 1994), metabolic (e.g., Calvaresi & Bryan, 2001), immunological (e.g., C. J. Wilson, Finch, & Cohen, 2002), hormonal (e.g., Wahlin, Robins Wahlin, Small, & Ba¨ckman, 1998), and circulatory (e.g., Fahlander et al., 2000) disturbances, all of which could potentially result in false negatives. Indeed, there is evidence that some persons classified as cognitively impaired at baseline not only remained free of AD many years later but actually exhibited improved cognitive performance at follow-up (e.g., Palmer et al., 2002). It is important to note that these improvers had no increased risk of being diagnosed with AD at a subsequent follow-up examination. Further, persons who will be diagnosed with AD vary both in terms of onset of precipitous decline and rate of decline during the preclinical period (Ba¨ckman et al., 2003; Rubin et al., 1998). In particular, incident AD cases who exhibit relatively normal cognitive performance until shortly before follow-up, when decline occurs very rapidly, will be difficult to catch. As with the data on patterns of cognitive impairment in normal aging and preclinical AD discussed above, these observations are consistent with a continuity view of AD development. Specifically, labels such as “normal aging,” “preclinical AD,” “MCI,” and “early AD” may best be viewed as instances on a dimension of brain and cognitive functioning rather than as discrete categories. These facts obvi-

527

ously hamper the utility of cognitive tests in identifying persons eligible for pharmacological or other interventions designed to slow down the disease process. Having said that, we should note that the meta-analysis reported in this article provides a rather conservative test of differences between cases and controls in the sense that effect sizes were assessed in a univariate fashion (i.e., d values were estimated for single-task variables). There is evidence that the ability to identify persons at risk for developing AD increases substantially when tasks assessing different cognitive domains (e.g., episodic memory, executive functioning, verbal ability) are combined into the same prediction model (e.g., Albert et al., 2001; Chen et al., 2001; B. J. Small, Viitanen, & Ba¨ckman, 1997). Obviously, this was not feasible in the current meta-analysis. In addition, indicators of other domains of functioning may be useful in identifying at-risk individuals. As noted, these include multiple measures of brain structure and function such as volumetric measures (e.g., Fox et al., 1996, 2001; Killiany et al., 2000; van der Flier, van den Heuvel, Weverling-Rijnsburger, Bollen, et al., 2002), glucose metabolism (e.g., Arnaíz et al., 2001; Silverman et al., 2001), blood flow (e.g., Kogure et al., 2000; S. A. Small et al., 1999), amyloid deposits (Morris & Price, 2001; Yamaguchi et al., 2001), and white-matter hyperintensities (Wolf et al., 2000). They also include markers for genetic predisposition such as the presence of the apolipoprotein ␧4 allele (for an overview, see Farrer et al., 1997), subjective memory complaints (Geerlings, Jonker, Bouter, Ade`r, & Schmand, 1999; Palmer, Ba¨ckman, Winblad, & Fratiglioni, 2003), family reports of cognitive impairment (Daly et al., 2000), and depressive symptoms (Berger et al., 1999; R. S. Wilson, Schneider, et al., 2003). The identification of specific risk factors acting as precipitating factors may provide further help in differentiating between preclinical AD and nonprogressive cognitive impairment. Possible precipitating factors include medical events such as unrecognized hypertension (Launer et al., 2000), head trauma (Plassman et al., 2000), and stroke (Snowdon et al., 1997) as well as factors such as engagement in intellectual activities (e.g., Scarmeas, Levy, Tang, Manly, & Stern, 2001; Wang, Karp, Winblad, & Fratiglioni, 2002), proneness to distress (R. S. Wilson, Evans, et al., 2003), and social isolation (Fratiglioni, Wang, Ericsson, Maytan, & Winblad, 2000). From a clinical perspective, then, the point is that although a variety of different factors may show relatively good predictive accuracy in identifying persons in a preclinical phase of AD, these factors are rarely used in the same prediction model. An important avenue for future research is to combine predictors from different behavioral and biological domains. To the extent that the factors included account for unique variance in group classification, such a multivariate approach will increase overall prediction accuracy. Further, it enables the examination of possible interactive effects among various preclinical markers. Although the present analysis of cognitive markers in preclinical AD provides a comprehensive account of the current state of knowledge in this domain of research, there are limitations to note. The first concern has to do with the precision with which the diagnosis of AD could be made among the studies included in the current analysis. We included only studies in which standardized diagnostic instruments were used (e.g., DSM–III–R, DSM–IV, ICD-10, NINCDS–ADRDA criteria). Recent evidence suggests that in research settings comparable to those represented in the

528


meta-analysis, accuracy of the clinical diagnosis is high. Specifically, a diagnostic accuracy around .90 has been reported in both clinical (e.g., Salmon et al., 2002) and community-based (e.g., Massoud et al., 1999) samples when verified against autopsy data. Nonetheless, occasional failures to identify mildly impaired persons as early AD cases at baseline assessment may have resulted in an overestimation of the cognitive differences between preclinical AD cases and controls. Second, among the studies included in the analysis few had information on all, or even most, of the ability domains targeted. The fact that the effect-size estimates for some cognitive domains (i.e., global cognitive ability and episodic memory) were based on more than 7,000 participants, whereas those for others (i.e., executive functioning) were based on fewer than 1,000 participants has to be taken into account when comparing effect sizes across domains. As a result of these differences, the analyses of moderator variables were restricted to global cognitive ability and episodic memory, and it was not possible to examine systematically whether effect sizes varied across cognitive domains depending on the time between baseline assessment and diagnosis. For example, proponents of the view that episodic memory is the earliest cognitive marker of incipient AD could still argue their case; the present analysis is not informative regarding the onset of decline for different cognitive domains. However, it may be noted that the effect size was larger for global cognitive ability (d ⫽ 1.12) than for episodic memory (d ⫽ 0.76) for follow-up intervals over 3 years. Note also that those studies that used assessment periods spanning several decades found clear preclinical impairment in domains other than episodic memory, such as linguistic skill (Snowdon et al., 1996) and general intelligence (Whalley et al., 2000). Another consequence of the limited database is that we were unable to assess the influence of the moderator variables simultaneously. In some cases, moderator variables may be confounded. For example, studies in which participants are preselected on the basis of their cognitive performance may primarily use convenience sampling rather than population-based sampling. Disentangling the effects of these and other variables would require considerably more studies than are available in the current literature. Further, the existing database does not allow for partitioning the length of the follow-up period into shorter time intervals. Future longitudinal studies should address this issue in order to provide more definite information about the specific time at which precipitous cognitive decline normally occurs in preclinical AD. A final concern has to do with the way in which different outcome measures were classified into cognitive domains. Obviously, for certain tasks (e.g., MMSE, word recall) classification is unproblematic, whereas for others (e.g., verbal fluency, Digit Symbol) it is more ambiguous. Given the present findings, it is interesting to note that there is also emerging evidence of a preclinical period with cognitive deficits in other dementing disorders, such as vascular dementia (e.g., Ingles, Wentzel, Fisk, & Rockwood, 2002; Jonsson Laukka, Jones, Small, Fratiglioni, & Ba¨ckman, 2004), frontotemporal dementia (Geschwind et al., 2001), and Huntington’s disease (Lawrence et al., 1998; Siemers et al., 1996). Thus, preclinical cognitive impairment is not isolated to AD. Measures of cognitive performance may prove useful in the early identification of at-risk

individuals over a wide spectrum of dementia diseases, although the specific patterns of impairment may vary across etiologies. In summary, the current meta-analysis has provided compelling evidence that impairment in multiple cognitive domains several years before clinical diagnosis is characteristic of AD. The generality of the cognitive impairment observed is highly consistent with recent observations that numerous brain structures and functions are affected prior to the AD diagnosis. Finally, we would like to highlight the finding that the magnitude of the preclinical cognitive impairment was affected relatively little by various study and task characteristics.

References Albert, M. S., Moss, M. B., Tanzi, R., & Jones, K. (2001). Preclinical prediction of AD using neuropsychological tests. Journal of the International Neuropsychological Society, 7, 631– 639. American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd ed., rev.). Washington, DC: Author. American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author. Andreasen, N., Minthon, L., Vanmechelen, E., Vanderstichele, H., Davidsson, P., Winblad, B., & Blennow, K. (1999). Cerebrospinal fluid tau and A␤ 42 as predictors of development of Alzheimer’s disease in patients with mild cognitive impairment. Neuroscience Letters, 273, 5– 8. Anstey, K. J., & Christensen, H. (2000). Education, activity, health, blood pressure, and apolipoprotein E as predictors of cognitive change in old age: A review. Gerontology, 46, 163–177. Arnaíz, E., Jelic, V., Almkvist, O., Wahlund, L. O., Winblad, B., Valind, S., & Nordberg, A. (2001). Impaired cerebral glucose metabolism and cognitive functioning predict deterioration in mild cognitive impairment. NeuroReport, 12, 851– 855. Ba¨ckman, L., & Forsell, Y. (1994). Episodic memory functioning in a community-based sample of old adults with major depression: Utilization of cognitive support. Journal of Abnormal Psychology, 103, 361– 370. Ba¨ckman, L., Jones, S., Small, B. J., Aguëro-Torrres, H., & Fratiglioni, L. (2003). Rate of cognitive decline in preclinical Alzheimer⬘s disease: The role of comorbidity. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 58, P228 –P236. Ba¨ckman, L., & Small, B. J. (1998). Influences of cognitive support on episodic remembering: Tracing the process of loss from normal aging to Alzheimer’s disease. Psychology and Aging, 13, 267–276. Ba¨ckman, L., Small, B. J., & Fratiglioni, L. (2001). Stability of the preclinical episodic memory deficit in Alzheimer’s disease. Brain, 124, 96 –102. Ba¨ckman, L., Small, B. J., & Fratiglioni, L. (2004). Cognitive deficits in preclinical Alzheimer’s disease: Current knowledge and future directions. In R. A. Dixon, L. Ba¨ckman, & L.-G. Nilsson (Eds.), New frontiers in cognitive aging (pp. 161–177). New York: Oxford University Press. Ba¨ckman, L., Small, B. J., & Wahlin, Å. (2001). Aging and memory: Cognitive and biological perspectives. In J. E. Birren & K. W. Schaie (Eds.), Handbook of the psychology of aging (5th ed., pp. 349 –377). San Diego: Academic Press. Ba¨ckman, L., Small, B. J., Wahlin, Å., & Larsson, M. (1999). Cognitive functioning in very old age. In F. I. M. Craik & T. A. Salthouse (Eds.), Handbook of aging and cognition (2nd ed., pp. 499 –558). Mahwah, NJ: Erlbaum. Benton, A. L. (1974). Revised Visual Retention Test (4th ed.). New York: Psychological Corporation. Benton, A. L., & Hamsher, K. (1983). Multilingual Aphasia Examination. Iowa City, IA: AJA Associates.

COGNITIVE IMPAIRMENT IN PRECLINICAL AD Berger, A-K., Fratiglioni, L., Forsell, Y., Winblad, B., & Ba¨ckman, L. (1999). The occurrence of depressive symptoms in the preclinical phase of AD: A population-based study. Neurology, 53, 1998 –2002. Bondi, M. W., Salmon, D. P., Galasko, D., Thomas, R-G., & Thal, L. J. (1999). Neuropsychological function and apolipoprotein E genotype in the preclinical detection of Alzheimer’s disease. Psychology and Aging, 14, 295–303. Bozoki, A., Giordani, B., Heidebrink, J. L., Berent, S., & Foster, N. L. (2001). Mild cognitive impairments predict dementia in nondemented elderly patients with memory loss. Archives of Neurology, 58, 411– 416. Braak, H., & Braak, E. (1995). Staging of Alzheimer disease-related neurofibrillary tangles. Neurobiology of Aging, 16, 271–284. Calvaresi, E., & Bryan, J. (2001). B vitamins, cognition, and aging: A review. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 56, P327–P339. Chen, P., Ratcliff, G., Belle, S. H., Cauley, J. A., DeKosky, S. T., & Ganguli, M. (2001). Patterns of cognitive decline in pre-symptomatic Alzheimer⬘s disease: A prospective community study. Archives of General Psychiatry, 58, 853– 858. Christensen, A. L. (1984). Luria’s neuropsychological investigation. Copenhagen, Denmark: Munksgaard. Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Erlbaum. Cohn, L. D., & Becker, B. (2003). How meta-analysis increases statistical power. Psychological Methods, 8, 243–253. Craik, F. I. M. (1983). On the transfer of information from temporary to permanent memory. Philosophical Transactions of the Royal Society of London, 302, 341–359. Craik, F. I. M. (1986). A functional account of age differences in memory. In F. Klix & H. Hagendorf (Eds.), Human memory and cognitive capabilities (pp. 409 – 422). Amsterdam: Elsevier. Craik, F. I. M., & McDowd, J. M. (1987). Age differences in recall and recognition. Journal of Experimental Psychology: Learning, Memory, and Cognition, 13, 474 – 479. Daly, E., Zaitchik, D. Copeland, M., Schmahmann, J., Gunther, J., & Albert, M. (2000). Predicting conversion to Alzheimer’s disease using standardized clinical information. Archives of Neurology, 57, 675– 680. Delis, D. C., Kramer, J. H., Kaplan, E., & Ober, B. A. (1987). California Verbal Learning Test. San Antonio, TX: Psychological Corporation. Devanand, D. P., Michaels-Marston, K. S., Liu, X., Pelton, G. H., Padilla, M., Marder, K., et al. (2000). Olfactory deficits in patients with mild cognitive impairment predict Alzheimer’s disease at follow-up. American Journal of Psychiatry, 15, 1399 –1405. Easterbrook, P. J., Berlin, J. A., Gopalan, R., & Matthews, D. R. (1991). Publication bias in clinical research. Lancet, 337, 867– 872. Egger, M., Davey Smith, G., Schneider, M., & Minder, C. (1997). Bias in meta-analysis detected by a simple, graphical test. British Medical Journal, 315, 629 – 634. Elias, M. F., Beiser, A., Wolf, P. A., Au, R., White, R. F., & D’Agostino, R. B. (2000). The preclinical phase of Alzheimer’s disease: A 22-year prospective study of the Framingham cohort. Archives of Neurology, 57, 808 – 813. Fabrigoule, C., Rouch, I., Taberly, A., Letenneur, L., Commenges, D., Mazaux, J. M., et al. (1998). Cognitive process in preclinical phase of dementia. Brain, 121, 135–141. Fahlander, K., Wahlin, Å., Fastbom, J., Grut, M., Forsell, Y., Hill, R. D., et al. (2000). The relationship between signs of cardiovascular deficiency and cognitive performance in old age: A population-based study. Journal of Gerontology, Series B: Psychological Sciences and Social Sciences, 55, P259 –P265. Farrer, L. A., Cupples, L. A., Haines, J. L., Hyman, B., Kukull, W. A., Mayeux, R., et al. (1997). Effects of age, sex, and ethnicity on the

529

association between apolipoprotein E genotype and Alzheimer disease. A meta-analysis. Journal of the American Medical Association, 278, 1349 –1356. Flicker, L. (1999). Acetylcholinesterase inhibitors for Alzheimer’s disease. British Medical Journal, 318, 615– 616. Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). “Mini-Mental State”: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12, 189 –198. Fox, N. C., Crum, W. R., Scahill, R. I., Stevens, J. M., Janssen, J. C., & Rossor, M. N. (2001). Imaging of onset and progression of Alzheimer’s disease with voxel-compression mapping of serial magnetic images. Lancet, 358, 201–205. Fox, N. C., Warrington, E. K., Freeborough, P. A., Hartikainen, P., Kennedy, A. M., Stevens, J. M., & Rossor, M. N. (1996). Presymptomatic hippocampal atrophy in Alzheimer’s disease. Brain, 119, 2001– 2007. Fratiglioni, L., Wang, H. X., Ericsson, K., Maytan, M., & Winblad, B. (2000). Influence of social network on occurrence of dementia: A community-based longitudinal study. Lancet, 355, 1315–1319. Geerlings, M. I., Jonker, C., Bouter, L. M., Ade`r, H. J., & Schmand, B. (1999). Association between memory complaints and incident Alzheimer’s disease in elderly people with normal baseline cognition. American Journal of Psychiatry, 156, 531–537. Geschwind, D. H., Robidoux, J., Alarcoń, M., Miller, B. L., Wilhelmsen, K. C., Cummings, J. L., & Nasreddine, Z. S. (2001). Dementia and neurodevelopmental predisposition: Cognitive dysfunction in presymptomatic subjects precedes dementia by decades in frontotemporal dementia. Annals of Neurology, 50, 741–746. Giambra, L. M., Arenberg, D., Kaws, C., Zonderman, A. B., & Costa, P. T. (1995). Adult life span changes in immediate visual memory and verbal intelligence. Psychology and Aging, 10, 123–139. Grober, E., & Kawas, C. (1997). Learning and retention in preclinial and early Alzheimer’s disease. Psychology and Aging, 12, 183–188. Haist, F., Gore, J. B., & Mao, H. (2001). Consolidation of human memory over decades revealed by functional magnetic resonance imaging. Nature Neuroscience, 4, 1139 –1145. Hansen, L. A., De Teresa, R., Davies, P., & Terry, R. D. (1988). Neocortical morphometry, lesion counts, and choline acetyltransferase levels in the age spectrum of Alzheimer’s disease. Neurology, 38, 48 –54. Hedges, L. V., & Olkin, I. (1985). Statistical methods for meta-analysis. New York: Academic Press. Heinrichs, R. W., & Zakzanis, K. K. (1998). Neurocognitive deficits in schizophrenia: A quantitative review of the evidence. Neuropsychology, 12, 426 – 445. Helkala, E. L., Koivisto, K., Hanninen, T., Vanhanen, M., Kuusisto, J., Mykkanen, L., et al. (1997). Stability of age-associated memory impairment during a longitudinal population-based study. Journal of the American Geriatrics Society, 45, 120 –122. Howieson, D. B., Dame, A., Camicioli, R., Sexton, G., Payami, H., & Kaye, J. A. (1997). Cognitive markers preceding Alzheimer’s dementia in the healthy oldest old. Journal of the American Geriatrics Society, 45, 584 –589. Hultsch, D. F., Hertzog, C., Dixon, R. A., & Small, B. J. (1998). Memory change in the aged. Cambridge, England: Cambridge University Press. Ingles, J. L., Wentzel, C., Fisk, J. D., & Rockwood, K. (2002). Neuropsychological predictors of incident dementia in patients with vascular cognitive impairment. Stroke, 33, 1999 –2002. Jacobs, D. M., Sano, M., Dooneief, G., Marder, K., Bell, K. L., & Stern, Y. (1995). Neuropsychological detection and characterization of preclinical Alzheimer’s disease. Neurology, 45, 317–324. Jacoby, L. L., Toth, J. P., & Yonelinas, A. P. (1993). Separating conscious and unconscious influences on memory: Measuring recollection. Journal of Experimental Psychology: General, 122, 139 –154.

530


Jonsson Laukka, E., Jones, S., Small, B. J., Fratiglioni, L., & Ba¨ckman, L. (2004). Similar patterns of cognitive deficits in the preclinical phases of vascular dementia and Alzheimer’s disease. Journal of the International Neuropsychological Society, 10, 382–391. Kaplan, E., Goodglass, H., & Weintraub, S. (1983). Boston Naming Test. Philadelphia: Lea & Fabiger. Killiany, R. J., Gomez-Isla, T., Moss, M., Kikinis, R., Sandor, T., Jolesz, F., et al. (2000). Use of structural magnetic resonance imaging to predict who will get Alzheimer’s disease. Annals of Neurology, 47, 430 – 439. Kogure, D., Matsuda, H., Ohnishi, T., Asada, T., Uno, M., Kunihiro, T., et al. (2000). Longitudinal evaluation of early Alzheimer’s disease using brain perfusion SPECT. Journal of Nuclear Medicine, 41, 1155–1162. Korten, A. E., Henderson, A. S., Christensen, H., Jorm, A. F., Ridgers, B., Jacomb, P., & Mackinnon, A. J. (1997). A prospective study of cognitive function in the elderly. Psychological Medicine, 27, 919 –930. Lange, K. L., Bondi, M. W., Salmon, D. P., Galasko, D., Delis, D. C., Thomas, R. G., & Thal, L. J. (2002). Decline in verbal memory during preclinical Alzheimer’s disease: Examination of the effect of APOE genotype. Journal of the International Neuropsychological Society, 8, 943–955. Launer, L. J., Ross, G. W., Petrovitch, H., Masaki, K. H., Foley, D. J., White, L. R., & Havlik, R. J. (2000). Midlife blood pressure and dementia: The Honolulu-Asia aging study. Neurobiology of Aging, 21, 49 –55. Lawrence, A. D., Hodges, J. R., Rosser, A. E., Kershaw, A., FrenchConstant, C., Rubinsztein, D. C., et al. (1998). Evidence for specific cognitive deficits in preclinical Huntington’s disease. Brain, 121, 1329 – 1341. Lezak, M. D. (1995). Neuropsychological assessment (3rd ed.). London: Oxford University Press. Light, R. J., & Pillemer, D. B. (1984). Summing up: The science of reviewing research. Cambridge, MA: Harvard University Press. Lindenberger, U., & Baltes, P. B. (1997). Intellectual functioning in old and very old age: Cross-sectional results from the Berlin Aging Study. Psychology and Aging, 12, 410 – 432. Linn, R. T., Wolf, P. A., Bachman, D. L., & Knoefel, J. (1995). The “preclinical phase” of probable Alzheimer’s disease: A 13-year prospective study of the Framingham cohort.. Archives of Neurology, 52, 485– 490. Mann, D. M. A. (1994). Pathological correlates of dementia in Alzheimer’s disease. Neurobiology of Aging, 15, 357–360. Martin, A., Brouwers, P., Cox, C., & Fedio, P. (1985). On the nature of the verbal memory deficit in Alzheimer⬘s disease. Brain and Language, 25, 323–341. Massoud, F., Devi, G., Stern, Y., Lawton, A., Goldman, J. E., Liu, Y., et al. (1999). A clinicopathological comparison of community-based and clinic-based cohorts of patients with dementia. Archives of Neurology, 56, 1368 –1373. McKhann, G., Drachman, D., Folstein, M., Katzman, R., Price, D.,& Stadlan, E. M. (1984). Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA work group under the auspices of Department of Health and Human Services Task force on Alzheimer’s disease. Neurology, 34, 939 –944. Morris, J. C., & Price, J. L. (2001). Pathologic correlates of nondemented aging, mild cognitive impairment, and early-stage Alzheimer’s disease. Journal of Molecular Neuroscience, 17, 101–118. Nyberg, L., McIntosh, A. R., Houle, S., Nilsson, L-G., & Tulving, E. (1996). Activation of medial temporal structures during episodic memory retrieval. Nature, 380, 715–717. Olichney, J. M., Morris, S. K., Ochoa, C., Salmon, D. P., Thal, L. J., Kutas, M., & Iragui, V. J. (2002). Abnormal verbal event-related potentials in mild cognitive impairment and incipient Alzheimer’s disease. Journal of Neurology, Neurosurgery, and Psychiatry, 73, 377–384.

Palmer, K., Ba¨ckman, L., Winblad, B., & Fratiglioni, L. (2003). Detection of Alzheimer’s disease and dementia in the preclinical phase: Predictivity of a three-step procedure in a population-based study. British Medical Journal, 326, 245–247. Palmer, K., Wang, H. X., Ba¨ckman, L., Winblad, B., & Fratiglioni, L. (2002). Differential evolution of cognitive impairment in nondemented older persons: Results from the Kungsholmen Project. American Journal of Psychiatry, 159, 436 – 442. Park, D. C., Lautenschalger, G., Hedden, T., Davidson, N. S., Smith, A. D., & Smith, P. K. (2002). Models of visuospatial and verbal memory across the adult life span. Psychology and Aging, 17, 299 –320. Petersen, R. C., Smith, G. E., Waring, S. C., Ivnik, R. J., Tangalos, E. G., & Kokmen, E. (1999). Mild cognitive impairment: Clinical characterization and outcome. Archives of Neurology, 56, 303–308. Plassman, B. L., Havlik, R. J., Steffens, D. C., Helms, M. J., Newman, T. N., Drosdick, D., et al. (2000). Documented head injury in early adulthood and risk of Alzheimer’s disease and other dementias. Neurology, 55, 1158 –1166. Post, S. G. (1999). Future scenarios for the prevention and delay of Alzheimer disease onset in high-risk groups: An ethical perspective. American Journal of Preventive Medicine, 16, 105–110. Reitan, R. (1958). Validity of the Trailmaking Test as an indicator of organic brain damage. Perceptual and Motor Skills, 8, 271–276. Rey, A. (1941). L’examen psychologique dans les cas d⬘encephalopathie traumatique[Psychological examination of cases with traumatic brain injury]. Archives of Psychology, 28, 286 –340. Roth, M., Huppert, F. A., Mountjoy, C. Q., & Tym, E. (1988). CAMDEX: The Cambridge Examination for Mental Disorders of the Elderly. Cambridge, England: Cambridge University Press. Rubin, E. H., Storandt, M., Miller, J. P., Kinscherf, D. A., Grant, E. A., Morris, J. C., & Berg, L. (1998). A prospective study of cognitive function and onset of dementia in cognitively healthy elders. Archives of Neurology, 55, 395– 401. Salmon, D. P., Thomas, R. G., Pay, M. M., Booth, A., Hofstetter, C. R., Thal, L. J.,& Katzman, R. (2002). Alzheimer’s disease can be accurately diagnosed in very mildly impaired individuals. Neurology, 59, 1022– 1028. Salthouse, T. A. (1996). The processing-speed theory of adult age differences in cognition. Psychological Review, 103, 403– 428. Scarmeas, N., Levy, G., Tang, M. X., Manly, J., & Stern, Y. (2001). Influence of leisure activity on the incidence of Alzheimer’s disease. Neurology, 57, 2236 –2242. Siemers, E., Foroud, T., Bill, D. J., Sorbel, J., Norton, J. A., Jr., Hodes, M. E., et al. (1996). Motor changes in presymptomatic Huntington disease gene carriers. Archives of Neurology, 53, 487– 492. Silverman, D. H. S., Small, G. W., Chang, C. Y., Lu, C. S., Kung de Aburto, M. A., Chen, W., et al. (2001). Positron emission tomography in evaluation of dementia: Regional brain metabolism and long-term outcome. Journal of the American Medical Association, 286, 2120 –2127. Simon, E., Leach, L., Winocur, G., & Moscovitch, M. (1995). Intact primary memory in mild to moderate Alzheimer’s disease: Indices from the California Verbal Learning Test. Journal of Clinical and Experimental Neuropsychology, 16, 414 – 422. Small, B. J., Fratiglioni, L., Viitanen, M., Winblad, B., & Ba¨ckman, L. (2000). The course of cognitive impairment in preclinical Alzheimer’s disease: 3- and 6-year follow-up of a population-based sample. Archives of Neurology, 57, 839 – 844. Small, B. J., Herlitz, A., Fratiglioni, L., Almkvist, O., & Ba¨ckman, L. (1997). Cognitive predictors of incident Alzheimer’s disease: A prospective longitudinal study. Neuropsychology, 11, 413– 420. Small, B. J., Viitanen, M., & Ba¨ckman, L. (1997). Mini-Mental State Examination scores as predictors of Alzheimer’s disease: Incidence data from the Kungsholmen Project, Stockholm. Journals of Gerontology, Series A: Biological Sciences and Medical Sciences, 52, M299 –M304.

COGNITIVE IMPAIRMENT IN PRECLINICAL AD Small, S. A., Perara, G., DeLaPaz, R., Mayeux, R., & Stern, Y. (1999). Differential regional dysfunction of the hippocampal formation among elderly with memory decline and Alzheimer’s disease. Annals of Neurology, 45, 466 – 472. Snowdon, D. A., Greiner, L. H., Mortimer, J. A., Riley, K. P., Greiner, P. A., & Markesbery, W. R. (1997). Brain infarction and the clinical expression of Alzheimer’s disease: The Nun Study. Journal of the American Medical Association, 277, 813– 817. Snowdon, D. A., Kemper, S. J., Mortimer, J. A., Greiner, L. H., Wekstein, D. R., & Markesbery, W. R. (1996). Linguistic ability in early life and cognitive function and Alzheimer’s disease in late life: Findings from the Nun Study. Journal of the American Medical Association, 275, 28 –532. Spreen, O., & Strauss, E. (1998). A compendium of neuropsychological tests (2nd ed.). London: Oxford University Press. Squire, L. R. (1986). Mechanisms of memory. Science, 232, 1612–1619. Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18, 643– 662. Tierney, M. C., Szalai, J. P., Dunn, E., Geslani, D., & McDowell, I. (2000). Prediction of probable Alzheimer disease in patients with symptoms suggestive of memory impairment. Archives of Family Medicine, 9, 527–532. Tierney, M. C., Szalai, J. P., Snow, W. G., & Fisher, R. H. (1996). The prediction of Alzheimer disease: The role of patient and informant perceptions of cognitive deficits. Archives of Neurology, 53, 423– 427. Tierney, M. C., Szalai, J. P., Snow, W. G., Fisher, R. H., Nores, A., Nadon, G., et al. (1996). Prediction of probable Alzheimer’s disease in memoryimpaired patients: A prospective longitudinal study. Neurology, 46, 661– 665. Tierney, M. C., Szalai, J. P., Snow, W. G., Fisher, R. H., Tsuda, T., Chi, H., et al. (1996). A prospective study of the clinical utility of ApoE genotype in the prediction of outcome in patients with memory impairment. Neurology, 46, 149 –154. Tulving, E., & Colotla, V. (1970). Free recall of trilingual words. Cognitive Psychology, 1, 86 –98. van der Flier, W. M., van den Heuvel, D. M. J., Weverling-Rijnsburger, A. W. E., Bollen, E. L. E. M., Westendorp, G. J., van Buchem, M. A., & Middlekoop, H. A. M. (2002). Magnetization transfer imaging in normal aging, mild cognitive impairment, and Alzheimer’s disease. Annals of Neurology, 52, 62– 67. van der Flier, W. M., van den Heuvel, D. M. J., Weverling-Rijnsburger, A. W. E., Spilt, A., Bollen, E. L. E. M., Westendorp, R. G. J., et al. (2002). Cognitive impairment in AD and mild cognitive impairment is associated with global brain damage. Neurology, 59, 874 – 879. Vargha-Khadem, F., Gadian, D. G., Watkins, K. E., Connelly, A., Van Paesschen, W., & Mishkin, M. (1997). Differential effects of early hippocampal lesions on episodic and semantic memory. Science, 277, 376 –380.

531

Wahlin, Å., Robins Wahlin, T-B., Small, B. J., & Ba¨ckman, L. (1998). Influences of thyroid stimulating hormone on cognitive functioning in very old age. Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 53, P234 –P239. Wang, H. X., Karp, A., Winblad, B., & Fratiglioni, L. (2002). Late-life engagement in social and leisure activities is associated with a decreased risk of dementia: A longitudinal study from the Kungsholmen Project. American Journal of Epidemiology, 155, 1081–1087. Wechsler, D. (1945). A standardized memory scale for clinical use. Journal of Psychology, 19, 87–95. Wechsler, D. (1981). Wechsler Adult Intelligence Scale—Revised. New York: Psychological Corporation. Whalley, L. J., Starr, J. M., Athawes, R., Hunter, D., Pattie, A., & Deary, I. J. (2000). Childhood mental ability and dementia. Neurology, 55, 1455–1459. Wilson, C. J., Finch, C. E., & Cohen, H. J. (2002). Cytokines and cognition—The case for a head-to-toe inflammatory paradigm. Journal of the American Geriatrics Society, 50, 2041–2056. Wilson, R. S., Evans, D. A., Bienias, J. L., de Leon, C. F. M., Schneider, J. A., & Bennett, D. A. (2003). Proneness to psychological distress is associated with risk of Alzheimer’s disease. Neurology, 61, 1479 –1485. Wilson, R. S., Gilley, D. W., Bennett, D. A., Beckett, L. A., & Evans, D. A. (2000). Person-specific paths of cognitive decline in Alzheimer’s disease and their relation to age. Psychology and Aging, 15, 18 –28. Wilson, R. S., Schneider, J. A., Bienias, J. L., Arnold, S. E., Evans, D. A., & Bennett, D. A. (2003). Depressive symptoms, clinical AD, and cortical plaques and tangles in older persons. Neurology, 61, 1102–1107. Wolf, H., Ecke, G. M., Bettin, S., Dietrich, J., & Gertz, H. J. (2000). Do white matter changes contribute to the subsequent development of dementia in patients with mild cognitive impairment? A longitudinal study. International Journal of Geriatric Psychiatry, 15, 803– 812. World Health Organization. (1992). International statistical classification of diseases and related health problems (10th ed., rev.). Geneva: World Health Organization. Yamaguchi, H., Sugihara, S., Ogawa, A., Oshima, N., & Ihara, Y. (2001). Alzheimer beta amyloid deposition enhanced by ApoE epsilon 4 gene precedes neurofibrillary pathology in the frontal association cortex of nondemented senior subjects. Journal of Neuropathology and Experimental Neurology, 60, 731–739. Zazzo, R. (1974). Test des deux barrages. Actualite´s pe´dagogiques et psychologiques [Test of two impairments. Current issues in pedagogy and psychology](Vol. 7). Neuchatel, Switzerland: Delachaux et Nestle.

Received May 3, 2004 Revision received October 6, 2004 Accepted October 15, 2004 䡲