Alzheimer's disease

2009 Psychology Press, an imprint of the Taylor & Francis Group, an Informa business

http://www.psypress.com/neurocase DOI: 10.1080/13554790902729465

NEUROCASE2009, 15 (4), 278293

NNCSThe emergence of cognitive discrepancies in preclinical Alzheimers disease: A six-year case study

Cognitive Discrepancy Case Study Mark W. Jacobson,1,2 Dean C. Delis,1,2 Guerry M. Peavy,3 Spencer R. Wetter,1 Erin D. Bigler,4,5 Tracy J. Abildskov,4 Mark W. Bondi,1,2 and David P. Salmon3

1Veterans Affairs San Diego Healthcare System, San Diego, CA, USA2Department of Psychiatry, University of California San Diego, La Jolla, CA, USA3Department of Neurosciences, University of California San Diego, La Jolla, CA, USA4Department of Psychology and Neuroscience, Brigham Young University, Provo, UT, USA5Department of Psychiatry, University of Utah School of Medicine and the Utah Brain Institute, Salt Lake City, UT, USA

We present neuropsychological data from an 81-year-old individual who was followed over a six-year period,initially as a healthy control participant. She performed above age-adjusted cutoff scores for impairment on mostneuropsychological tests, including learning and memory measures, until the final assessment when she received adiagnosis of probable Alzheimers disease (AD). Despite generally normal scores on individual cognitive tests, hercognitive profile revealed increasingly large cognitive discrepancies when contrasting verbal versus visuospatialtasks, and complex versus basic-level tasks. The present case provides intriguing evidence that cognitive-discrepancymeasures could improve our ability to detect subtle changes in cognition at the earliest, preclinical stages of AD.

Keywords: Alzheimers disease; Neuropsychology; Cognition; Discrepancy; Preclinical.

INTRODUCTION

Identifying cognitive markers of a preclinical phaseof Alzheimers disease (AD) has been a majorresearch focus in neuropsychology. This line ofstudy is important both scientifically for increasingour understanding of the earliest effects of thisneurodegenerative process, and clinically for indi-viduals who may benefit from pharmacological,environmental, behavioral, and long-term planning

interventions (Akaike, 2006; Allain, Bentue-Ferrer,Gandon, Le Doze, & Belliard, 1997; Giovannettiet al., 2007; Miguel-Hidalgo, Alvarez, Cacabelos, &Quack, 2002; Morrison et al., 2005; Vajda, 2002).Memory deficits are arguably the most frequentlyidentified cognitive markers for AD, and they alsoconstitute a primary diagnostic criterion foramnestic mild cognitive impairment (Petersen,2004). However, individuals in prodromal stagesprior to a dementia diagnosis often demonstrate

The authors wish to thank Dr Terry Jernigan, Sarah Archibald M.S., and Dr Christine Fennema-Notestine for their assistance withMRI image collection and presentation. We also thank the participant, as well as Jodessa Braga and the staff of the UCSD AlzheimersDisease Research Center for their contributions to this research. Finally, we would like to thank the anonymous reviewers for theirhelpful comments. Dr Delis is a co-author of the D-KEFS and CVLT-II and receives royalties from them. Preparation of this articlewas supported in by a Veterans Administration Career Development Award (M.W.J.), a Veterans Administration Merit ReviewAward (D.D.), NIMH award R01MH063782 (G.M.P.); Alzheimers Association Award IIRG-07-59343 (M.W.B.), National Instituteon Aging awards K24 AG026431 and R01 AG012674 (M.W.B.), and NIA Grant P50 AG 05131 (UCSD ADRC).

Address correspondence to Mark W. Jacobson, Ph.D., Veterans Affairs Medical Center San Diego, Department. 151B, 3350La Jolla Village Dr., San Diego, CA 92151, USA. (E-mail: [email protected]).

COGNITIVE DISCREPANCY CASE STUDY 279

heterogeneous cognitive deficits sometimes withoutsignificant memory deficits (Backman, Jones,Berger, Laukka, & Small, 2004), which poses adiagnostic challenge for clinicians. Recent studiesof preclinical AD (Pre-AD) have documented arange of subtle changes in cognitive domains inaddition to learning and memory, including executivefunctions, processing speed, working memory andattention, and visuospatial skills (Chen et al., 2001;Greenwood, Sunderland, Friz, & Parasuraman,2000; Jacobson, Delis, Bondi, & Salmon, 2005a;Jacobson et al., 2005b; Twamley, Ropacki, &Bondi, 2006). Other studies have failed to identify adistinct or consistent pattern of cognitive deficits inPre-AD (Goldman et al., 2001) resulting in a closerexamination of the characterization and classifica-tion of preclinical and prodromal periods of AD(Dubois et al., 2007). Even the hallmark signs ofinsidious onset and gradual cognitive decline havebeen questioned, as some studies describe a plateauperiod with largely preserved cognition, followedby a rapid decline just prior to an AD diagnosis(Small, Fratiglioni, & Backman, 2001).

Not surprisingly, no single test or cognitivedomain has proven sufficient for identifying themajority of individuals in the earliest, preclinicalstages of AD (Albert, Blacker, Moss, Tanzi, &McArdle, 2007; Backman et al., 2004; Chen et al.,2000; Lopez et al., 2006). Cognitive markers likedelayed-recall that have been investigated in large-scale studies do not always translate into accuratediagnosis at the individual-patient level (Ivnik et al.,2000), particularly when atypical cognitive profilesemerge. Consequently, case studies that characterizeless frequent phenomenon (Godbolt et al., 2005)promise to make important contributions to identi-fying clinical characteristics of Pre-AD.

In the current study, we present an individual(DH) who was medically and cognitively healthyand followed annually as an elderly control partici-pant over a six-year period before a rapid cognitivedecline resulted in an AD diagnosis. The results ofannual cognitive testing are inherently valuable,given the scarcity of comprehensive longitudinaldata in a documented preclinical phase. This casealso illustrates another potentially important fea-ture of Pre-AD: the possibility that significant cog-nitive discrepancies exist during an asymptomaticperiod when overall cognitive performance is abovea cutoff-level of impairment. In a recent series ofstudies, we have used cognitive-discrepancyanalyses to identify normal-functioning elderlysubgroups with subtle cognitive changes that may

herald the onset of AD (Houston et al., 2005;Jacobson et al., 2005a, 2005b; Wetter et al., 2005,2006). We have suggested that inconsistent findingsin the search for preclinical cognitive markers couldreflect the use of central-tendency analyses of singletest scores that were not sensitive to the presence ofsubgroups within the larger, at-risk elderly group.For example, both AD and Pre-AD subgroups havebeen identified with disproportionate impairment inverbal relative to spatial skills (or vice versa)(Albert, Duffy, & McAnulty, 1990; Demadura,Delis, Jacobson, & Salmon, 2001; Finton et al.,2003; Martin, 1990; Massman & Doody, 1996;Strite, Massman, Cooke, & Doody, 1997). It followsthat cognitive-discrepancy analyses contrasting twodomains may be superior to central-tendency analy-ses of each task individually for identifying asym-metric deficits, particularly in Pre-AD.

For example, Jacobson, Delis, Bondi, andSalmon (2002) investigated verbal/spatial discrep-ancies in cognitively intact, community-dwellingelderly participants, half of whom showed a subse-quent decline in cognition and conversion to an ADdiagnosis. When test scores were examined twoyears prior to the cognitive decline, performancediscrepancies between the Boston Naming Test(Kaplan, Goodglass, & Weintraub, 1983) andBlock Design subtest (Wechsler, 1981) revealed thatthe subgroup who would later convert to an ADdiagnosis had significantly greater cognitive-discrepancy scores. Importantly, these discrepanciesexisted despite the groups obtaining comparablescores on each measure when examined individu-ally. Similar results were also found in older adultsgenetically at-risk for AD because of the Apolipo-protein E4 allele (ApoE-e4) using discrepancy scoresderived from verbal versus spatial tests of attention(Jacobson et al., 2005a), global versus local memorytask (Jacobson et al., 2005b), and verbal versusdesign fluency (Houston et al., 2005). Discrepancyscores derived from higher-level versus more basictasks within the same modality (i.e., verbal memoryversus vocabulary) (Wetter et al., 2006) were foundto be effective for distinguishing elderly individualswith a higher genetic risk for AD.

We present a case study that examines neuropsy-chological tests data from an elderly individualwho had intact cognitive abilities initially, butwhose test performances also suggested emergingcognitive discrepancies over the six-year periodprior to a diagnosis of probable AD diagnosis. Ourgoal was to apply our previous methodologies tothis individuals annual cognitive assessments to

280 JACOBSON ET AL.

determine if cognitive-discrepancy scores would besensitive to subtle declines in DHs cognitive func-tioning during a plateau period in which most testscores were within normal limits. We highlightcognitive-discrepancy scores because they canpotentially identify intra-individual changes thatmight not be apparent in a single, norm-referencedtest score (Bondi et al., 2008).

CASE STUDY

DH is a right-handed, Caucasian woman with twoyears of college education who was one of approxi-mately 30 participants recruited as a healthy con-trol participant through the UCSD AlzheimersDisease Research Center (ADRC) for a study ofexecutive functions in older adults. In addition tocomprehensive neuropsychological assessments,she received annual neurologic and psychosocialevaluations to confirm stable physical and emo-tional health, as well as functional independence.We followed this participant for 6 years beginningat age 81. At the fifth assessment, DH reportedsymptoms consistent with mild depression, shereceived a working diagnosis of pseudodementia,and she was referred for neuroimaging studies. Atthe sixth assessment (age 86), DH complained ofsubjective changes in her memory ability and somefunctional impairment in activities of daily living.The assessment at this time revealed moderatelyimpaired performance on memory tests and inmultiple cognitive domains, and she subsequentlyreceived a consensus diagnosis of probable AD.Prior to this time, DH was living independently,had no obvious impairment in her daily activities,and met standard ADRC criteria for participationas a normal control subject, which excluded indi-viduals with a history of significant neurologic orpsychiatric disorder, alcohol or substance abuse ordependence, traumatic brain injury with loss ofconsciousness, stroke, learning disability, or use ofmedications that impair cognition. DH was ini-tially in an asymptomatic, preclinical stage of ADaccording to recent diagnostic criteria (Duboiset al., 2007). Subsequent MRI confirmation ofmedial temporal lobe atrophy was followed by abrief, prodromal period when cognitive and func-tional deficits became apparent, until she finallymet criteria for a probable AD diagnosis after thefinal, sixth assessment. Apolipoprotein E genotyp-ing had been completed using the methoddescribed in previous studies (Corder et al., 1993)

and indicated an Apolipoprotein E status of e3/3.She completed a UCSD IRB-approved informedconsent process prior to the assessments.

METHODS

Neuropsychological assessment

Trained psychometrists using standardized testprocedures supervised by a neuropsychologistadministered all cognitive tests. In addition to anannual neuropsychological test battery (seeSalmon et al., 2002), we administered several addi-tional tests to measure cognitive discrepanciesbased on our previous studies cited above. Table 1provides a description of DHs cognitive testresults, and Table 2 details specific test variablesand normative information used to calculatecognitive-discrepancy scores in the present study.Specifically, we examined three cognitive discrep-ancies that contrast differential declines in tasksthat are more verbally mediated relative to tasksthat are more non-verbally or spatially mediated(Bisiacchi, Borella, Bergamaschi, Carretti, &Mondini, 2008; Jacobson et al., 2002). We alsoexamined four discrepancies between a complexhigher-level cognitive skill and a correspondingbasic-ability skill in the same domain. Tests wereselected to contrast an ability that is likely to beresilient (e.g., vocabulary ability) with an abilityconsidered to be more vulnerable to a dementingprocess (e.g., verbal memory).

For most comparisons, cognitive-discrepancyscores were calculated by first converting rawscores to z-scores based on archival data from acohort of older, normal control participantsfrom the UCSD ADRC with a range of age andeducation similar to DH. This group met normalcontrol criteria, and had a minimum of 4 consec-utive annual neuropsychological and neurologicexaminations without a significant cognitivedecline or change in diagnosis. The resultinggroup (n = 28) had a mean age of 82.1 (SD =2.1), and education range from 12 to 18 years,mean = 15.4 (SD = 2.5). Because DH receivedexecutive function tests that were not adminis-tered to the ADRC group, we calculated somez-scores from the D-KEFS normative database(Delis & Kaplan, 2001), using the age-adjustedscaled scores for calculation of z-scores [cohortage range was 8089; n = 100; mean (SD) educa-tional level of 12.06 (2.54) years].


RESULTS

Overall cognitive functioning

Based on her ANART score, DH had an esti-mated premorbid Verbal IQ in the high averagerange, consistent with her previous level of educa-tion and occupation as a teacher. On a screeningmeasure for dementia, her initial performancewas in the above-average range (Mattis DementiaRating Scale total = 138 points; age-scaled score6081%ile) and declined to average, then low-average range (132 points; 2440%ile) at her finaltwo assessments. Her MMSE total scoresremained within normal limits (2927 points)during all six assessments.

Table 2 summarizes DHs longitudinal neuropsy-chological test scores (raw and standardizedz-scores). With a few exceptions (confrontationnaming at the fourth assessment), her performanceson individual tests in most domains (language, psy-chomotor speed, visuoconstruction, attention andworking memory, executive functioning) wereabove typical cutoff levels for impairment (i.e., >1.5SD below age-adjusted means) relative to the nor-mative group. Her performances on the most com-monly used indicators of Pre-AD such as the CVLTlong-delay recall scores did not decline into animpaired range (1.5 SD below the normative

group mean) until the final assessment prior to herAD diagnosis. For all six evaluations, her scoreswere at or above the average range relative to thenormative comparison group on all other memorymeasures including the Wechsler Memory Scale-Revised, Logical memory delayed recall, HeatonVisual Reproduction Test delayed recall, DRSmemory subscale, and CERAD verbal recall.

Cognitive discrepancies: Verbal versus Spatial tasks

Figure 1 shows cognitive-discrepancy scoresderived from an auditory attention test (WAIS-RDigit Span subtest) relative to a visual attentiontest (Visual Scanning condition of the DKEFSTrail Making Test). Her auditory attention spanremained within the average to low-average rangeswhile her visual attention ability was within nor-mal limits for her age. In contrast, there were cog-nitive discrepancies greater than 1 SD in four ofthe six assessments despite relatively unimpairedperformance on each task individually.

Figure 2 shows DHs performance on theBoston Naming Test (BNT) relative to the BlockDesign subtest of the WAIS-R (BD) across sixyears of testing. Although her naming abilityshowed a gradual decline, her performance did not

TABLE 1 Tests and variable descriptions used for cognitive discrepancy analyses

Verbal/spatial discrepancy Verbal test Non-verbal test

Verbal naming vs. visuoconstruction

Boston Naming Test1 30 item short form: total spontaneous correct

Block Design subtest1 (WAIS-R): total points including time bonus

Verbal vs. non-verbal attention Digit Span subtest1 (WAIS-R) total correct forward/backward

D-KEFS Trail-making test2: seconds to completion

Verbal vs. non-verbal memory California Verbal Learning1 Test: total words recalled on Long-delay free recall

Heaton Visual Reproduction1 Test: total points delayed recall of designs

Basic skill versus Complex task Discrepancy Basic skill task Complex skill task

Psychomotor speed vs. Sequencing/set-shifting (D-KEFS Trail-making test) 2

motor speed condition (secs. to completion) (a) average of letter & number sequencing conditions

(b) letter/number sequencing shifting condition

Color/Word naming vs. Inhibition & set-shifting (D-KEFS Color Word Interference Test)2

average of color naming & color word reading (secs. to completion)

(a) interference condition(b) color word interference/set shifting

conditionPhonemic vs. Semantic fluency

(D-KEFS Verbal Fluency Test)2letter fluency: total words (F, A, S) in 60 secs. (a) category fluency animals, boys names

(b) category shifting vegetables/furnitureVocabulary vs. verbal list learning Vocabulary subtest1 (WAIS-R) total pts. CVLT learning trials2 (total trials 15)

Notes: z-Scores based on normative information from these sources: 1UCSD ADRC age-corrected norms (ages 8087; n = 18); seeSalmon et al. (2002). 2D-KEFS age-corrected norms (8089 years, n = 100). (Delis & Kaplan, 2001).

282 JACOBSON ET AL.

reach a consistent level of impairment until thefinal two assessments. Her block constructionskills remained stable, at or above the averagerange throughout this period. However, cognitive-discrepancy score analyses showed a greater than 1SD discrepancy between BNT and BD z-scores infive of the six annual assessments prior to her diag-nosis change, with increasing discrepancies overtime.

Figure 3 shows cognitive-discrepancy scorescomparing the CVLT long-delay free recall condi-tion relative to the delayed recall condition of theVisual Reproduction subtest (Heaton version[HVRT]). Her non-verbal recall remained in theaverage to high average range for all six assess-ments relative to a gradual decline in her verbalrecall ability. However, her CVLT delayed-recallscores remained above a level of mild to moderate

TABLE 2 Diagnosis, age, and cognitive screening measures: Raw and z-scores for neuropsychological tests

Assessment #1

Assessment #2

Assessment #3

Assessment #4

Assessment #5

Assessment #6

Diagnosis NC NC NC NC Pseud-Dem Prob ADAge 81 82 83 84 85 86MMSE 29 n/a 29 n/a 27 26DRS total/(SS) 138 136 136 136 132 130

12 10 10 11 9 9ANART errors/est IQ 8/120+ 6/120+ 8/120+ 5/120+ 13/110120 13/117 110120

Raw score (z-score)

Raw score (z-score)

Raw score (z-score)

Raw score (z-score)

Raw score (z-score)

Raw score (z-score)

Vocabulary 58 62 53 54 50 510.44 1.18 0.48 0.31 1.04 0.85

Digit Symbol 47 49 47 43 43 400.34 0.54 0.34 0.07 0.07 0.38

Block Design 58 62 53 51 40 511.65 0.91 1.19 1.00 0.25 1.00

WCST Categories 6 5 6 6 5 60.56 0.19 0.56 0.56 0.19 0.56

WCST errors 9 7 8 0 8 50.30 0.06 0.11 1.3 0.11 0.41

Boston Naming total pts.

29 28 26 27 23 22

0.48 0.19 -1.55 0.87 -3.58 -4.20WMS LM LDR 22 14 24 20 17 12

0.35 0.50 0.70 0.0 0.25 1.0WMS LM % savings 100 100 100 83 100 67

2.25 2.25 2.25 0.92 2.25 0.34CVLT LD FR 11 9 9 6 5 3

0.35 0.60 0.60 0.4 0.7 -1.4CVLT Recog Discrim 0.91 0.91 0.89 0.80 0.84 0.93

0.15 0.15 0.6 -2.2 1.3 0.23DRS Memory (SS) 25 23 23 22 20 22

14 10 10 9 6 9Clock drawing*

/clock setting*3 of 3 12 of 12

3 of 3 12 of 12

3 of 3 12 of 12

3 of 312 of 12

3 of 3 12 of 12

3 of 312 of 12

CERAD list trial 3* 10/10 10/10 10/10 10/10 10/10 9/10CERAD recall* 10/10 10/10 10/10 10/10 10/10 10/10CERAD recall* 10/10 10/10 10/10 10/10 10/10 10/10

*Limited distribution of scores z-score not calculated. Bold type indicates a score below a cutoff for impairment of 1.5 SD belowcomparison group mean.Abbreviations: DRS, Dementia Rating Scale; ANART, American National Adult Reading Test; MMSE, Mini Mental State Exam;Pseu-Dem, pseudodementia; Prob AD, probable AD; WCST = Wisconsin Card Sorting Test; WMS-LR = Wechsler Memory Scale-Revised, Logical Memory subtest; CVLT, California Verbal Learning Test; CERAD, Consortium to Establish a Registry forAlzheimers Disease (Verbal Memory task).


impairment until one year prior to her AD diagno-sis. Her cognitive-discrepancy scores steadilyincreased over time, yielding a 1 SD differencebetween verbal/visual recall in four of six assess-ments.

Cognitive discrepancies: Executive function and Memory tasks

We examined verbal fluency discrepancies using threeconditions of the D-KEFS Verbal Fluency test:

Letter Fluency (average of F, A, and S), CategoryFluency (animals and boys names) and Category-Switching (alternating fruit and furniture words); seeFigure 4. Her test scores demonstrated consistentlyabove-average and stable performance on letter flu-ency relative to a gradual decline on the two types ofcategory-fluency tasks. The cognitive-discrepancyscores between letter and category-switching fluencywas greater than 1 SD in all six assessments.

We used the D-KEFS Trail Making Test toexamine discrepancies between complex visuomotorsequencing/shifting ability, and a basic-skill task

TABLE 3 Verbal vs. spatial discrepancies: Mean (SE), frequency and range over six years

Mean (SE) z-score discrepancy

# Years with >1 SD z-score discrepancy

Minimum and maximum z-score discrepancy

Cognitive Discrepancy: Verbal vs. Spatial Task ComparisonsNaming (BNT) vs. Visuoconstruction (BD) 2.56 (.72) 5/6 0.855.26Auditory Attention (DS) vs. Visuospatial Attention (VS) 1.38 (.13) 5/6 1.01.66Verbal Recall (CVLT) vs. Visuospatial Recall (HVRT) 1.47 (.41) 4/6 0.313.04

Basic-level Task vs. Complex Skill Comparisons Letter (LF) vs. Category Fluency (CF) 1.89 (.31) 5/6 0.672.66Letter vs. Category-Shifting Fluency (CSF) 2.71 (.23) 6/6 1.993.66Motor Speed (MS) vs. Number/Letter Sequencing &

Shifting (SS)1.05 (.36) 3/6 0.252.66

Color Naming & Word Reading (CNWR) vs. Color/Word Inhibition only (CWIT)

0.73 (.33) 2/6 01.66

Color Naming & Word Reading (CNWR) vs. Inhibition & Shifting (CWITS)

2.39 (.60) 5/6 03.33

Vocabulary (VOC) vs. Verbal Learning Trials (CVLT1-5) 1.40 (.42) 3/6 0.052.90

Abbreviations: BNT, Boston Naming Test; BD, WAIS-R Block Design subtest; DS, WAIS-R Digit Span subtest; VS, Visual Scanningcondition D-KEFS Trail Making; CVLT, California Verbal Learning Test Delayed Recall; HVRT, Heaton Visual Reproduction TestRussell version (Russell, 1975); LF / CF / CSF, D-KEFS Verbal Fluency: Letter / Category / Category Shifting; MS / SS, D-KEFSTrail Making Motor Speed / Sequencing and Shifting; CNWR, D-KEFS Color Word Interference Test combined Color and Wordreading; CWIT, D-KEFS Color Word Interference Test-Interference condition; CWITS / Interference Shifting condition; VOC,WAIS-R Vocabulary subtest; CVLT1-5, California Verbal Learning Test Learning Trials 15.

Figure 1. Visual Scanning versus Verbal Attention Span Discrepancies: D-KEFS Visual Scanning with Mildly Impaired Digit Span Scores.

VISUAL SCANNNG VERSUS VERBAL SPAN

1.5

1

0.5

0

0.5

1

1.5

YEARS TO AD DX

Z-SC

OR

E

D-KEFS Visual Scanning Digit Span Subtest

6 5 4 3 2 1

284 JACOBSON ET AL.

requiring visuomotor speed (tracing a line between aseries of circles with no letters or numbers; seeFigure 5). While DHs performance on thebasic-ability task remained stable (and withinnormal limits), her scores on the Number/LetterSwitching showed considerable variability,eventually declining into the impaired range. Hercognitive-discrepancy scores reflected a 1 SDdifference between complex and basic-skill tasks onthree of six assessments.

DHs performance on four conditions of the D-KEFS Color/Word Interference Test (CWIT) isshown in Figure 6. A basic-skill composite measurewas computed as the average of the Color Naming(CN) and the Word Reading (WR) conditions. Thecomplex skills consisted of (a) interference-onlycondition (traditional Stroop test), and (b) theinterference/shifting condition (switching between

naming the dissonant ink-color and reading thedissonant word). Cognitive-discrepancy scoresbetween basic-skill and interference-only measureswere greater than 1 SD in two of six assessments.However, the cognitive discrepancy score usingthe more difficult interference/shifting paradigmresulted in a greater than 1 SD discrepancy relativeto the basic-skill measure (mean of CN & WR) infive of six assessments, largely the result of declin-ing scores on the two complex CWIT conditions.

We examined the participants discrepancybetween basic word knowledge (WAIS-R Vocabu-lary subtest) and complex verbal learning ability(CVLT learning trials; total for trials 15) (seeFigure 7). DH showed a significant cognitive dis-crepancy between her word knowledge and hertotal verbal learning ability on three of the sixassessments prior to her diagnosis change.

Figure 2. Visuoconstruction versus Verbal Naming Discrepancies: Stable Block Design Scores with Declining Boston Naming Scores.

VISUOCONSTRUCTION VERSUS VERBAL NAMING

5

4

3

2

1

0

1

2

Years to AD Diagnosis

Z-SCORE

Block Design Test Boston Naming Test

6 5 4 3 2

Figure 3. Visual Recall versus Verbal Recall Discrepancies: Increasing Discrepancy between HVRT and CVLT Long-delay RecallScores.

VISUAL RECALL VERSUS VERBAL RECALL

21.5

10.5

00.5

11.5

22.5

Years to AD diagnosis

z-sc

ore

HVRT Delay Free Recall CVLT Long Delay free recall

6 5 4 3 2 1


She exhibited a gradual decline in verbal learningability, relative to stable and average to above-average performance on the Vocabulary task.

Figure 8 summarizes the pattern of the two typesof cognitive-discrepancy scores across six assess-ments. We plotted the average of the combined cog-nitive discrepancy scores (z-score differences) foreach assessment for all verbal/spatial discrepancies(Figure 8a), and for all complex/basic-skill discrep-ancies (Figure 8b). We fitted two linear trend lines tothese data, which indicated positive slopes over timefor verbal/spatial discrepancies (y = .391) and com-plex/basic-skill discrepancies (y = .387), suggesting

an overall increase in cognitive-discrepancy scoresas D.H. approached an AD diagnosis.

Neuroimaging

The participant underwent structural magneticresonance imaging (MRI) following the fifth assess-ment, one year prior to a change in diagnosis toprobable AD. High-resolution anatomical imageswere acquired using a T1-weighted fast-spin echo(SPGR) at a 1.5-Tesla GE MRI scanner (TR = 24ms, TE = 5 ms, flip angle = 45 degrees, FOV = 24

Figure 4. Verbal Fluency Discrepancies: Stable Letter Fluency Relative to Declining Category and Category Switching Scores.

VERBAL FLUENCY DISCREPANCIES

2

1.5

1

0.5

0

0.5

1

1.5

2

2.5

3


Z-SC

OR

E

DKEFS Letter Fluency DKEFS Category Fluency

DEKFS Category Switching

6 5 4 3 2 1

Figure 5. Trail Making Test Discrepancies: Stable Basic Motor, Sequencing Skills with Variable Performance and Deficits in ShiftingCondition.

TRAIL MAKING TEST DISCREPANCIES

3

2.5

2

1.5

1

0.5

0

0.5

1

1.5

2


Z-SC

OR

E

DKEFS Trails Motor SpeedDKEFS Trails Average of Num & Letter SequencingDEKFS Trails Number/Letter Sequencing + Shifting

6 5 4 3 2 1

286 JACOBSON ET AL.

cm, 1.2 mm contiguous sections). Selected axial andcoronal slices reveal asymmetric atrophy in medialtemporal regions (Figure 9a), particularly in the lefthemisphere. The neuroradiologists impressionsdescribed generalized cortical atrophy greater intemporal lobe regions, ventricular enlargement,and no indication of significant ischemic changes,tumor or mass. A 3-dimensional reconstruction(Bigler et al., 2002) of the T1-weighted image is alsodepicted (Figure 9b) to illustrate cortical atrophyand ventricular enlargement.

DISCUSSION

The present case illustrates the diagnostic chal-lenges inherent in identifying preclinical AD whenan individuals cognitive profile lacks many of the

typical hallmarks of this phase. Despite an even-tual diagnosis of probable AD after the sixth yearof participation, none of the first four assessmentswould justify an MCI diagnosis: her performanceson most individual cognitive tasks, including mem-ory tests, were within or above the average range,she was living independently, and she had no sub-jective memory complaints until the final assess-ment. With few exceptions (e.g., naming ability atthe fourth assessment), her performances on indi-vidual neuropsychological tests remained above alevel of impairment until the final year. Discrep-ancy scores have been used in many contexts as ameans of identifying subtle declines in cognitiveskills relative to those that are more resilient to aneurodegenerative process or dementia (Dori &Chelune, 2004; Finton et al., 2003; Ivnik et al.,2000; Wilde et al., 2001). Analysis of DHs

Figure 6. Stroop Discrepancies: Color/word Naming Scores and Traditional Stroop Scores with Impaired Interference/SwitchingCondition.

COLOR WORD INTERFERENCE DISCREPANCIES

3.5

32.5

2

1.51

0.5

0

0.51

1.5


Z-SC

OR

ES

DKEFS Color naming/Word Reading Baseline DKEFS Interference conditionDKEFS Interference + Switching condition

6 5 4 3 2 1

Figure 7. CVLT Learning versus Vocabulary Discrepancies: Stable Basic-skill Word Knowledge Ability with Declining VerbalLearning Scores.

WORD KNOWLEDGE VERSUS VERBAL LEARNING

3.5

3

2.5

2

1.5

1

0.5

0

0.5

1


Z-SCORE

Vocabulary CVLT Learning

6 5 4 3 2 1


intra-individual differences revealed a consistentpattern of cognitive discrepancies that may havesignaled an impending decline despite generallyintact performances on individual tasks. DHscomplex versus basic-level task discrepancy scoresreflected stable performance on more fundamentaltask components in the face of mild declines inexecutive function or memory abilities. Her verbalversus non-verbal discrepancies reflected an asym-metric decline in largely verbally mediated tasks.

Although the majority of DHs discrepancyscores reflected at least a 2 SD difference by thefinal assessment, the question arises as to whatconstitutes an atypical discrepancy? There is a con-siderable body of literature addressing the questionof clinical versus statistical significance of cognitivediscrepancies (Crawford & Garthwaite, 2005;Crawford, Garthwaite, & Gault, 2007; Lange &Chelune, 2006; Lange, Chelune, & Tulsky, 2006).A commonly used criteria considers a discrepancyunusual or significant if it occurs in less than 10%of the population (Hawkins & Tulsky, 2001;Kramer et al., 2006), although methods for dis-crepancy score calculations vary. Fortunately,

many tests now incorporate discrepancy measures(e.g., CVLT-II, WAIS-III, D-KEFS, WarringtonRecognition Memory Test) as standardized varia-bles, although base rate information is not alwaysavailable. Chelune, Holdnack, and Levy (2006)examined D-KEFS normative data and found thata letter-category fluency scaled-score discrepancyof +3 occurs in 10% of the sample (with 1315years of education). By comparison, DHs letter-category fluency discrepancy (at age 82) showed ascaled score difference of 5, with a contrast scaledscore of 15 (95th percentile rank) based on hersame age peers in the D-KEFS sample. In additionto the discrepancy magnitude, the frequency of sig-nificant discrepancies is also a consideration.A pattern of large, recurring discrepancies that areinternally consistent may provide more convincingevidence of impending cognitive decline than a sin-gle, isolated discrepancy. For example, three signi-ficant discrepancy measures occurred in 2% of thein the CVLT-II normative sample, while a singlesignificant discrepancy occurred in 22% of the sam-ple (Donders, 2006). Finally, discrepancy scoresvary by IQ and education with higher functioning

Figure 8. Scatterplot and Trend Lines: Increasing Longitudinal Discrepancies in (a) Verbal/Spatial Discrepancies and (b) Basic-ability/Complex Task Discrepancies.

Mean Discrepancy Scores: Verbal vs. Spatial Task Contrasts

y = 0.3872x + 0.8046R2 = 0.643

0

0.5

1

1.5

2

2.5

3

3.5

(a)

(b)

4

1 2 3 4 5 6years

z-sc

ore

diffe

renc

e

Mean discrepancy scores:Combined Complex vs. Basic Task Contrasts

y = 0.3917x + 0.4275

R2 = 0.6927

0

0.5

1

1.5

2

2.5

3

3.5

1 2 3 4 5 6years

z-sc

ore

diffe

renc

e

288 JACOBSON ET AL.

individuals exhibiting larger discrepancies, so theuse of stratified norms will prove important foraccurate interpretation (Hawkins & Tulsky, 2001).

A longitudinal review of DHs performanceoffers some evidence for a sub-clinical decline oncertain tests, although neuropsychologists in clini-cal practice rarely have the benefit of examining sixconsecutive annual assessments. The current caseunderscores some of the benefits afforded by serialassessments in individuals who have high premor-bid abilities. Cognitive impairment can be masked

in these individuals because declining scoresremain above norm-based cutoff scores(Mitrushina & Satz, 1991) as was seen in DHscase. Her test scores eventually indicated a veryrapid decline in the two years prior to her conver-sion to a probable AD diagnosis, and may be rep-resentative of a subgroup of patients whose diseaseonset begins with a plateau period followed by aprecipitous rate of cognitive decline (Small &Backman, 2007; Smith et al., 2007). The abruptdecline also was concurrent with subjective reports

Figure 9. Selected T1-weighted MRI scan: (a) images highlighting left temporal atrophy and (b) 3D cortical reconstruction withventricular system.

(a)

(b)


of emotional changes. The emergence of depres-sion and anxiety has been documented as a com-mon feature associated with symptoms ofdementia (Potter & Steffens, 2007), perhaps con-tributing to both subjective complaints and objec-tive deficits (van Norden et al., 2008).

DHs cognitive profile may be characteristic ofindividuals with greater cognitive reserve who areless likely to exhibit obvious cognitive deficits onindividual tests until neurodegenerative changesoverwhelm compensatory strategies (Riis et al.,2008; Tuokko, Garrett, McDowell, Silverberg, &Kristjansson, 2003). However, it is possible thather putative cognitive resilience was reinforced byrepeated testing. Confirmation of significantchange in neuropsychological test scores is a com-plex issue. Interpreting change in difference scoreshas additional considerations, as differential prac-tice effects of the two component tasks could serveto either mask or accentuate performance discrep-ancies. Detailed studies of practice effects in olderadults typically show either stable or improvedperformance on subsequent testings (Hickman,Howieson, Dame, Sexton, & Kaye, 2000; Paolo,Troester, & Ryan, 1997) with the most significantperformance gains occurring at the second testingoccasion (Ivnik et al., 1999). Thus, practice effectscould have contributed to the attenuation of cogni-tive discrepancy scores in the second year. Draw-ing firm conclusions about significant change isultimately enhanced with the use of reliable changeindices (Ivnik et al., 1999) and confidence intervalestimates which require information about meas-urement error, reliability estimates, standard errorof difference, and test intercorrelations (that arenot always readily available). Theoretically, verbal/spatial discrepancies would be more prone to prac-tice effects than would complex/basic-skill discrep-ancies because many performance based tasks(e.g., Block Design) show greater improvement onre-test relative to verbal tasks (Theisen, Rapport,Axelrod, & Brines, 1998).

Additional limitations of discrepancy scores canbe attributed to psychometric measurement issues,as the reliability of a contrast measure will typi-cally be lower than either of the contributing testvariables (Woods et al., 2005). Recently publisheddata on D-KEFS contrast measures showedreliability ranging from r = .43 (color/word namingvs. switching/interference) to r = .66 (letter vs.category fluency (Crawford, Sutherland, &Garthwaite, 2008). Discrepancy scores also can beinfluenced by relative differences in the sensitivity

and difficulty of the component tasks, differencesin score distributions, and floor or ceiling effects(Cherry, Buckwalter, & Henderson, 2002; Orsini,Trojano, Chiacchio, & Grossi, 1988). Finally, theuse of tests that quantify more discrete componentsof a cognitive domain (e.g., evaluating types ofsemantic memory rather than general semanticknowledge) (Ahmed, Arnold, Thompson, Graham,& Hodges, 2008) could improve sensitivity to pro-dromal changes compared to discrepancy scoresthat are based on variables from multi-factorialcognitive tests.

The presence of cognitive discrepancies in multi-ple domains could represent another early indica-tor of an incipient decline in functioning. Priorresearch has shown that cognitive discrepanciesgreater than 1 SD were more frequent in older indi-viduals who ultimately developed AD (Jacobsonet al., 2002), and in elderly groups at increased riskfor AD (Jacobson et al., 2005a,b). Several of ourprevious studies examining cognitive discrepanciesin older adults have shown that only 1020% ofolder adults (over age 65) who do not have anincreased genetic risk for AD (ApoE-e4 allele)show a greater than 1 SD discrepancy betweenverbal/spatial tasks or complex/basic-level tasks. Incontrast, 4055% of those with increased ApoE-e4genetic risk had large discrepancies (Houston et al.,2005; Jacobson et al., 2005a,b; Wetter et al., 2005),a ratio that is consistent with other studies of ADsubjects with asymmetric cognitive functioning(Massman & Doody, 1996; Strite et al., 1997).A significant caveat to the use of cognitive discrep-ancies is the possibility of premorbid asymmetriesin individual examinees. However, a recent studyof cognition in middle-aged twins who were atincreased genetic risk for AD due to ApoE-E4 gen-otype found larger verbal/non-verbal cognitive dis-crepancies relative to those without the E4 geneticrisk (Schultz et al., 2008). Cognitive discrepancieshave been documented in the early stages of AD aswell (Delis et al., 1992; Demadura et al., 2001;Finton et al., 2003; Massman & Doody, 1996).

Most of DHs discrepancy scores resulted fromdeclining scores on tasks with a substantial verbalcomponent, a profile often associated with greaterleft hemisphere compromise (Ott et al., 2000). Con-sistent with such profiles, DHs MRI findings afterher fifth assessment revealed atrophy in brainregions most likely to be impacted by AD-relatedneurodegeneration, particularly in medial temporalregions of the left hemisphere. Some studies suggestthat verbal/spatial discrepancies reflect asymmetric

290 JACOBSON ET AL.

changes in brain functions (Delis et al., 1992; Haxbyet al., 1990), brain structures (Finton et al., 2003;Soininen et al., 1994), or brain metabolism (France-schi et al., 1995). It should be noted, however, thatthe intermediate stages of progression to an ADdiagnosis are likely to be heterogeneous both interms of cognition and sites of neuropathology,particularly in individuals with multiple deficits(Mariani, Monastero, & Mecocci, 2007; Troster,2008). Rather than lateralized changes in brainstructure and function, cognitive discrepanciesparticularly complex versus basic-ability contrastscould reflect a more general disruption of neuralnetworks (Parasuraman & Martin, 1994). Recentstudies suggest that early AD is characterized by theloss of cortico-cortical projections that facilitateinteractions of multiple brain regions (Andrews-Hanna et al., 2007), which would likely result ingreater deficits in higher-level, executive functionsrelative to more fundamental cognitive abilities. Thebase-rates for structural brain asymmetry in Pre-AD require sample sizes that are only possible withlarge, multi-site neuroimaging studies.

In conclusion, some individuals without signific-ant memory impairment, and with generally nor-mal performance on individual cognitive tests, maynevertheless be in a preclinical phase of AD. Insome situations, cognitive-discrepancy scores maybe more sensitive than individual test scores foridentifying preclinical AD, because they highlightintra-individual changes in cognitive skills(Storandt, Grant, Miller, & Morris, 2006). Thepresent case suggests that serial assessments andthe unfolding of cognitive discrepancies over timemay improve detection of subtle cognitive changesin individuals with high premorbid ability who areat the earliest, preclinical stages of AD.

Original manuscript received 20 August 2008Revised manuscript accepted 19 December 2008

First published online 21 April 2009

REFERENCES

Ahmed, S., Arnold, R., Thompson, S. A., Graham,K. S., & Hodges, J. R. (2008). Naming of objects,faces and buildings in mild cognitive impairment.Cortex, 44(6), 746752.

Akaike, A. (2006). Preclinical evidence of neuroprotec-tion by cholinesterase inhibitors. Alzheimers Diseaseand Associated Disorders, 20(2 Suppl. 1), S811.

Albert, M. S., Duffy, F. H., & McAnulty, G. B. (1990).Electrophysiologic comparisons between two groups

of patients with Alzheimers disease. Archives of Neu-rology, 47(8), 857863.

Albert, M., Blacker, D., Moss, M. B., Tanzi, R., &McArdle, J. J. (2007). Longitudinal change in cogni-tive performance among individuals with mild cogni-tive impairment. Neuropsychology, 21(2), 158169.

Allain, H., Bentue-Ferrer, D., Gandon, J.-M., Le Doze, F.,& Belliard, S. (1997). Drugs used in Alzheimers dis-ease and neuroplasticity. Clinical Therapeutics: TheInternational Peer-Reviewed Journal of Drug Therapy,19(1), 415.

Andrews-Hanna, J. R., Snyder, A. Z., Vincent, J. L.,Lustig, C., Head, D., Raichle, M. E., & Buckner,R. L. (2007). Disruption of large-scale brain systemsin advanced aging. Neuron, 56(5), 924935.

Backman, L., Jones, S., Berger, A. K., Laukka, E. J., &Small, B. J. (2004). Multiple cognitive deficits duringthe transition to Alzheimers disease. Journal of Inter-nal Medicine, 256(3), 195204.

Bigler, E. D., Tate, D. F., Miller, M. J., Rice, S. A., Hessel,C. D., Earl, H. D., et al. (2002). Dementia, asymme-try of temporal lobe structures, and ApolipoproteinE genotype: Relationships to cerebral atrophyand neuropsychological impairment. Journal ofthe International Neuropsychological Society, 8(7),925933.

Bisiacchi, P. S., Borella, E., Bergamaschi, S., Carretti,B., & Mondini, S. (2008). Interplay between memoryand executive functions in normal and pathologicalaging. Journal of Clinical and Experimental Neuropsy-chology, 30(6), 723733.

Bondi, M. W., Jak, A. J., Delano-Wood, L., Jacobson,M. W., Delis, D. C., & Salmon, D. P. (2008). Neuro-psychological contributions to the early identificationof Alzheimers disease. Neuropsychology Review,18(1), 7390.

Chelune, G. J., Holdnack, J., & Levy, J. (2006). Educa-tion effects and base-rate information for the Delis-Kaplan Executive Function System (DKEFS)fluency measures. Archives of Clinical Neuropsychol-ogy, 21, B45.

Chen, P., Ratcliff, G., Belle, S. H., Cauley, J. A.,DeKosky, S. T., & Ganguli, M. (2000). Cognitivetests that best discriminate between presymptomaticAD and those who remain nondemented. Neurology,55(12), 18471853.

Chen, P., Ratcliff, G., Belle, S. H., Cauley, J. A.,DeKosky, S. T., & Ganguli, M. (2001). Patterns ofcognitive decline in presymptomatic Alzheimer dis-ease: A prospective community study. Archives ofGeneral Psychiatry, 58(9), 853858.

Cherry, B. J., Buckwalter, J. G., & Henderson, V. W.(2002). Better preservation of memory span relativeto supraspan immediate recall in Alzheimers disease.Neuropsychologia, 40(7), 846852.

Corder, E. H., Saunders, A. M., Strittmatter, W. J.,Schmechel, D. E., Gaskell, P. C., Small, G. W., et al.(1993). Gene dose of apolipoprotein E type 4 alleleand the risk of Alzheimers disease in late onset famil-ies. Science, 261(5123), 921923.

Crawford, J. R., & Garthwaite, P. H. (2005). Testingfor suspected impairments and dissociations in sin-gle-case studies in neuropsychology: Evaluation of


alternatives using Monte Carlo simulations andrevised tests for dissociations. Neuropsychology,19(3), 318331.

Crawford, J. R., Garthwaite, P. H., & Gault, C. B.(2007). Estimating the percentage of the populationwith abnormally low scores (or abnormally large scoredifferences) on standardized neuropsychological testbatteries: A generic method with applications. Neuro-psychology, 21(4), 419430.

Crawford, J. R., Sutherland, D., & Garthwaite, P. H.(2008). On the reliability and standard errors ofmeasurement of contrast measures from theD-KEFS. Journal of the International Neuropsycho-logical Society, 14(6), 10691073.

Delis, D., & Kaplan, E. (2001). Manual for the Delis-Kaplan Executive Function System. San Antonio, TX:Psychological Corporation.

Delis, D. C., Massman, P. J., Butters, N., Salmon, D. P.,et al. (1992). Spatial cognition in Alzheimers disease:Subtypes of global-local impairment. Journal of Clini-cal & Experimental Neuropsychology, 14(4), 463477.

Demadura, T., Delis, D. C., Jacobson, M., & Salmon, D.(2001). Do subgroups of patients with Alzheimersdisease exhibit asymmetric deficits on memory tests?Journal of Clinical & Experimental Neuropsychology,23(2), 164171.

Donders, J. (2006). Performance discrepancies on theCalifornia Verbal Learning Test Second Edition(CVLT-II) in the standardization sample. Psychologi-cal Assessment, 18(4), 458463.

Dori, G. A., & Chelune, G. J. (2004). Education-stratifiedbase-rate information on discrepancy scores withinand between the Wechsler Adult Intelligence Scale Third Edition and the Wechsler Memory Scale ThirdEdition. Psychological Assessment, 16(2), 146154.

Dubois, B., Feldman, H. H., Jacova, C., Dekosky, S. T.,Barberger-Gateau, P., Cummings, J., et al. (2007).Research criteria for the diagnosis of Alzheimers dis-ease: Revising the NINCDS-ADRDA criteria. Lan-cet Neurology, 6(8), 734746.

Finton, M. J., Lucas, J. A., Rippeth, J. D., Bohac, D. L.,Smith, G. E., Ivnik, R. J., Petersen, R. C., &Graff-Radford, N. R. (2003). Cognitive asymmetriesassociated with apolipoprotein E genotype in patientswith Alzheimers disease. Journal of the InternationalNeuropsychological Society, 9(5), 751759.

Franceschi, M., Alberoni, M., Bressi, S., Canal, N.,Comi, G., Fazio, F., Grassi, F., Perani, D., &Volonte, M. A. (1995). Correlations between cogni-tive impairment, middle cerebral artery flow velocityand cortical glucose metabolism in the early phase ofAlzheimers disease. Dementia, 6(1), 3238.

Giovannetti, T., Bettcher, B. M., Libon, D. J., Brennan,L., Sestito, N., & Kessler, R. K. (2007). Environmen-tal adaptations improve everyday action performancein Alzheimers disease: Empirical support from per-formance-based assessment. Neuropsychology, 21(4),448457.

Godbolt, A. K., Cipolotti, L., Anderson, V. M., Archer,H., Janssen, J. C., Price, S., et al. (2005). A decade ofpre-diagnostic assessment in a case of familial Alzhe-imers disease: Tracking progression from asympto-matic to MCI and dementia. Neurocase, 11(1), 5664.

Goldman, W. P., Price, J. L., Storandt, M., Grant, E. A.,McKeelJr, D. W., Rubin, E. H., & Morris, J. C.(2001). Absence of cognitive impairment or declinein preclinical Alzheimers disease. Neurology, 56(3),361367.

Greenwood, P. M., Sunderland, T., Friz, J. L., &Parasuraman, R. (2000). Genetics and visual atten-tion: Selective deficits in healthy adult carriers of theepsilon 4 allele of the apolipoprotein E gene. Proceed-ings of the National Academy of Sciences of the UnitedStates of America, 97(21), 1166111666.

Hawkins, K. A., & Tulsky, D. S. (2001). The influence ofIQ stratification on WAIS-III/WMS-III FSIQ-general memory index discrepancy base-rates in thestandardization sample. Journal of the InternationalNeuropsychological Society, 7(7), 875880.

Haxby, J. V., Grady, C. L., Koss, E., Horwitz, B.,Heston, L., Schapiro, M., et al. (1990). Longitudinalstudy of cerebral metabolic asymmetries and associ-ated neuropsychological patterns in early dementia ofthe Alzheimer type. Archives of Neurology, 47(7),753760.

Hickman, S. E., Howieson, D. B., Dame, A., Sexton, G.,& Kaye, J. (2000). Longitudinal analysis of the effectsof the aging process on neuropsychological test per-formance in the healthy young-old and oldest-old.Developmental Neuropsychology, 17(3), 323337.

Houston, W. S., Delis, D. C., Lansing, A., Jacobson,M. W., Cobell, K. R., Salmon, D. P., & Bondi, M. W.(2005). Executive function asymmetry in older adultsgenetically at-risk for Alzheimers disease: Verbal ver-sus design fluency. Journal of the International Neuro-psychological Society, 11(7), 863870.

Ivnik, R. J., Smith, G. E., Lucas, J. A., Petersen, R. C.,Boeve, B. F., Kokmen, E., & Tangalos, E. G. (1999).Testing normal older people three or four times at1- to 2-year intervals: Defining normal variance. Neu-ropsychology, 13(1), 121127.

Ivnik, R. J., Smith, G. E., Petersen, R. C., Boeve, B. F.,Kokmen, E., & Tangalos, E. G. (2000). Diagnosticaccuracy of four approaches to interpreting neuropsy-chological test data. Neuropsychology, 14(2), 163177.

Jacobson, M. W., Delis, D. C., Bondi, M. W., &Salmon, D. P. (2002). Do neuropsychological testsdetect preclinical Alzheimers disease: Individual-testversus cognitive-discrepancy score analyses. Neuro-psychology, 16(2), 132139.

Jacobson, M. W., Delis, D. C., Bondi, M. W., &Salmon, D. P. (2005a). Asymmetry in auditoryand spatial attention span in normal elderly genet-ically at risk for Alzheimers disease. Journal ofClinical and Experimental Neuropsychology, 27(2),240253.

Jacobson, M. W., Delis, D. C., Lansing, A., Houston, W.,Olsen, R., Wetter, S., et al. (2005b). Asymmetries inglobal-local processing ability in elderly people withthe apolipoprotein e-epsilon4 allele. Neuropsychol-ogy, 19(6), 822829.

Kaplan, E., Goodglass, H., & Weintraub, S. (1983). TheBoston Naming Test. Philadelphia, PA: Lea andFebiger.

Kramer, J. H., Nelson, A., Johnson, J. K., Yaffe,K., Glenn, S., Rosen, H. J., & Miller, B. L. (2006).

292 JACOBSON ET AL.

Multiple cognitive deficits in amnestic mild cognitiveimpairment. Dementia and Geriatric Cognitive Disor-ders, 22(4), 306311.

Lange, R. T., & Chelune, G. J. (2006). Application ofnew WAIS-III/WMS-III discrepancy scores for eval-uating memory functioning: Relationship betweenintellectual and memory ability. Journal of Clinicaland Experimental Neuropsychology, 28(4), 592604.

Lange, R. T., Chelune, G. J., & Tulsky, D. S. (2006).Development of WAIS-III General Ability IndexMinus WMS-III memory discrepancy scores. ClinicalNeuropsychology, 20(3), 382395.

Lopez, O. L., Becker, J. T., Jagust, W. J., Fitzpatrick, A.,Carlson, M. C., DeKosky, S. T., et al. (2006). Neu-ropsychological characteristics of mild cognitiveimpairment subgroups. Journal of Neurology, Neu-rosurgery and Psychiatry, 77(2), 159165.

Mariani, E., Monastero, R., & Mecocci, P. (2007). Mildcognitive impairment: A systematic review. Journal ofAlzheimers Disease, 12(1), 2335.

Martin, A. (1990). Neuropsychology of Alzheimers dis-ease: The case for subgroups. In M. F. Schwartz(Ed.), Modular deficits in Alzheimer-type dementia.Issues in the biology of language and cognition (pp.143175). Cambridge, MA: The MIT Press.

Massman, P. J., & Doody, R. S. (1996). Hemisphericasymmetry in Alzheimers disease is apparent inmotor functioning. Journal of Clinical and Experi-mental Neuropsychology, 18(1), 110121.

Miguel-Hidalgo, J. J., Alvarez, X. A., Cacabelos, R., &Quack, G. (2002). Neuroprotection by memantineagainst neurodegeneration induced by beta-amyloid(140). Brain Research, 958(1), 210221.

Mitrushina, M., & Satz, P. (1991). Effect of repeatedadministration of a neuropsychological battery in theelderly. Journal of Clinical Psychology, 47(6), 790801.

Morrison, R. S., Chichin, E., Carter, J., Burack, O.,Lantz, M., & Meier, D. E. (2005). The effect of asocial work intervention to enhance advance careplanning documentation in the nursing home. Journalof American Geriatric Society, 53(2), 290294.

Orsini, A., Trojano, L., Chiacchio, L., & Grossi, D.(1988). Immediate memory spans in dementia. Per-ceptual & Motor Skills, 67(1), 267272.

Paolo, A. M., Troester, A. I., & Ryan, J. J. (1997).California Verbal Learning Test: Normative data forthe elderly. Journal of Clinical & Experimental Neu-ropsychology, 19(2), 220234.

Parasuraman, R., & Martin, A. (1994). Cognition inAlzheimers disease: Disorders of attention andsemantic knowledge. Current Opinion in Neurobiol-ogy, 4(2), 237244.

Petersen, R. C. (2004). Mild cognitive impairment as adiagnostic entity. Journal of Internal Medicine,256(3), 183194.

Potter, G. G., & Steffens, D. C. (2007). Contribution ofdepression to cognitive impairment and dementia inolder adults. Neurologist, 13(3), 105117.

Riis, J. L., Chong, H., Ryan, K. K., Wolk, D. A., Rentz,D. M., Holcomb, P. J., & Daffner, K. R. (2008).Compensatory neural activity distinguishes differentpatterns of normal cognitive aging. Neuroimage,39(1), 441454.

Russell, E. W. (1975). A multiple scoring method for theassessment of complex memory functions. Journal ofConsulting and Clinical Psychology, 43, 800809.

Salmon, D. P., Thomas, R. G., Pay, M. M., Booth, A.,Hofstetter, C. R., Thal, L. J., et al. (2002). AlzheimersDisease can be accurately diagnosed in very mildlyimpaired individuals. Neurology, 59(7), 10221028.

Schultz, M. R., Lyons, M. J., Franz, C. E., Grant, M. D.,Boake, C., Jacobson, K. C., et al. (2008). Apolipopro-tein E genotype and memory in the sixth decade oflife. Neurology, 70(19 Pt 2), 17711777.

Small, B. J., & Backman, L. (2007). Longitudinal trajec-tories of cognitive change in preclinical Alzheimersdisease: A growth mixture modeling analysis. Cortex,43(7), 826834.

Small, B. J., Fratiglioni, L., & Backman, L. (2001).Canaries in a coal mine: Cognitive markers of pre-clinical Alzheimer disease. Archives of General Psy-chiatry, 58(9), 859860.

Smith, G. E., Pankratz, V. S., Negash, S., Machulda,M. M., Petersen, R. C., Boeve, B. F., et al. (2007). Aplateau in pre-Alzheimer memory decline: Evidencefor compensatory mechanisms? Neurology, 69(2),133139.

Soininen, H. S., Partanen, K., Pitkaenen, A., Vainio, P.,et al. (1994). Volumetric MRI analysis of the amy-gdala and the hippocampus in subjects with age-asso-ciated memory impairment: Correlation to visual andverbal memory. Neurology, 44(9), 16601668.

Storandt, M., Grant, E. A., Miller, J. P., & Morris, J. C.(2006). Longitudinal course and neuropathologicoutcomes in original vs revised MCI and in pre-MCI.Neurology, 67(3), 467473.

Strite, D., Massman, P. J., Cooke, N., & Doody, R. S.(1997). Neuropsychological asymmetry inAlzheimers disease: Verbal versus visuoconstruc-tional deficits across stages of dementia. Journal ofthe International Neuropsychological Society, 3(5),420427.

Theisen, M. E., Rapport, L. J., Axelrod, B. N., & Brines,D. B. (1998). Effects of practice in repeated adminis-trations of the Wechsler Memory Scale Revised innormal adults. Assessment, 5(1), 8592.

Troster, A. I. (2008). Neuropsychological characteristicsof dementia with Lewy bodies and Parkinsons dis-ease with dementia: Differentiation, early detection,and implications for mild cognitive impairment andbiomarkers. Neuropsychology Review, 18(1), 103119.

Tuokko, H., Garrett, D. D., McDowell, I., Silverberg, N.,& Kristjansson, B. (2003). Cognitive decline in high-functioning older adults: Reserve or ascertainmentbias? Aging and Mental Health, 7(4), 259270.

Twamley, E. W., Ropacki, S. A., & Bondi, M. W.(2006). Neuropsychological and neuroimagingchanges in preclinical Alzheimers disease. Journal ofthe International Neuropsychological Society, 12(5),707735.

Vajda, F. J. (2002). Neuroprotection and neurodegener-ative disease. Journal of Clinical Neurosciences, 9(1),48.

van Norden, A. G., Fick, W. F., de Laat, K. F., vanUden, I. W., van Oudheusden, L. J., Tendolkar, I.,et al. (2008). Subjective cognitive failures and


hippocampal volume in elderly with white matterlesions. Neurology, 71(15), 11521159.

Wechsler, D. (1981). Wechsler Adult Intelligence Scale-Revised. San Antonio, TX: The PsychologicalCorporation.

Wetter, S. R., Delis, D. C., Houston, W. S., Jacobson,M. W., Lansing, A., Cobell, K., et al. (2005). Deficitsin inhibition and flexibility are associated with theAPOE-E4 allele in nondemented older adults. Journalof Clinical and Experimental Neuropsychology, 27(8),943952.

Wetter, S. R., Delis, D. C., Houston, W. S., Jacobson,M. W., Lansing, A., Cobell, K., et al. (2006). Hetero-geneity in verbal memory: A marker of preclinical

Alzheimers disease? Neuropsychology, Developmentand Cognition. Section B, Aging Neuropsychology andCognition, 13(34), 503515.

Wilde, N., Strauss, E., Chelune, G. J., Loring, D. W.,Martin, R. C., Hermann, B. P., et al. (2001). WMS-III performance in patients with temporal lobe epi-lepsy: Group differences and individual classifica-tion. Journal of the InternationalNeuropsychological Society, 7(7), 881891.

Woods, S. P., Scott, J. C., Conover, E., Marcotte, T. D.,Heaton, R. K., & Grant, I. (2005). Test-retest reliabil-ity of component process variables within the Hop-kins Verbal Learning Test-Revised. Assessment,12(1), 96100.

Documents

Alzheimer's disease