BRAIN AND COGNITION 38, 220–233 (1998)ARTICLE NO. BR981029

Ideomotor Apraxia in Early Alzheimer’s Disease:Time and Accuracy Measures

Lee Willis,* Maurine Behrens,† Wendy Mack,* and Helena Chui*

*University of Southern California School of Medicine; and †Whittier College, California

Twenty-six Alzheimer’s disease (AD) and 42 healthy control (NC) subjects wereevaluated with neuropsychological and apraxia batteries. ADs produced a greaterrange of error types, but did not differ from NCs in their most frequent error types.Hand sequencing ability contributed significantly to AD praxis with no predictorsfor NCs. Although groups did not differ in gesture time, the AD group had signifi-cantly longer response latencies for periods prior to gesture execution and the effectwas prominent for transitive tasks and nondominant hand use. Results illustrate thesensitivity of timing measures in identifying abnormal praxis in early stages of AD. 1998 Academic Press

Key Words: Apraxia; Alzheimer’s disease; accuracy; time.

INTRODUCTION

Apraxia is traditionally defined by exclusion as a disorder of skilled ges-tures not resulting from absence or poverty of movement, sensory loss, weak-ness, abnormal tone, extrapyramidal or cerebellar involvement or resultingsolely from intellectual deterioration or defective comprehension (Heil-man & Gonzalez Rothi, 1993; Roy & Square, 1985; Poeck, 1986). An exten-sive research literature supports the existence and uniqueness of the disorderwith general agreement about its varied presentations, which may be elicitedthrough manipulation of task demands. For example, ideomotor apraxia re-fers to difficulty with accurate and smooth execution of simple or complexmovements by imitation or by pantomime. Usually elicited during clinicaltesting, ideomotor apraxia can predict the later emergence of problems withactivities of daily living, having significance for the patient’s future treatment(Edwards, Deuel, Baum, & Morris, 1991; Foundas, Macauley, Raymer,Maher, Heilman, & Gonzalez Rothi, 1995).

Although apraxia has been studied extensively in a variety of patient

Address correspondence and reprint requests to Lee Willis, University of Southern Califor-nia School of Medicine, Rancho Los Amigos Medical Center, 7601 East Imperial Highway,800-West Annex, Downey, CA 90242.

2200278-2626/98 $25.00Copyright 1998 by Academic PressAll rights of reproduction in any form reserved.

TIME AND ACCURACY MEASURES IN EARLY AD 221

groups, formal investigations have been confined primarily to stroke. Rela-tively few studies of apraxia in Alzheimer’s disease (AD) have been reported(Della Sala, Lucchelli, & Spinnler, 1987; Edwards et al., 1991; Foster,Thomas, Chase, Patronas, Gillepsie, & Fedio, 1986; Lucchelli, Lopez, Fagli-oni, & Boller, 1993; Ochipa, Gonzalez Rothi, & Heilman, 1992; Rapcsak,Croswell, & Rubens, 1989; Taylor, 1994; Travniczek-Marterer, Danielczyk,Simanyi, & Fisher, 1993; Yesavage, Brooks, Taylor, & Tinklenberg, 1993).Even fewer studies have investigated apraxia in the early stages of AD de-spite the fact that information about initial characteristics of a dementia canhave substantial diagnostic value (Della Sala et al., 1987; Edwards et al.,1991; Travniczek-Marterer et al., 1993). Timing characteristics also have notbeen directly investigated with AD samples, although the importance of thisfeature of apraxic performance has been demonstrated (Gonzalez Rothi,Mack, Verfaellie, Brown, & Heilman, 1988; Raade, Gonzalez Rothi, & Heil-man, 1991; Shelton & Knopman, 1991).

The purpose of the present investigation was to develop a further under-standing of the nature of ideomotor apraxia in early AD by contrasting theperformances of a well-characterized AD patient group with those of an age-equivalent healthy control group. Goals of the comparison were to obtaina detailed description of error patterns and gesture timing and to providequantitative information about other cognitive and motor deficits in the sam-ple to place apraxia within an overall performance profile in early stages ofAD dementia.

METHOD

Participants

The study sample included 68 right-handed, English-speaking subjects; 26 patients diag-nosed with AD and 42 older nondemented volunteers serving as normal control (NC) subjects.AD subjects were drawn from the Southern California Alzheimer’s Disease Diagnostic andTreatment Center at Rancho Los Amigos Medical Center in Downey, California. Normalcontrols were relatives of patients or were recruited through community organizations. AllAD subjects showed evidence of mild to moderate dementia according to scores on the Mini-Mental State examination (MMSE, Folstein, Folstein, & McHugh, 1975). Their scores rangedfrom 17 to 28 (M 5 22.6, SD 5 3.4). The NC subjects showed no evidence of cognitiveimpairment during enrollment screening, later confirmed by neuropsychological test results.To qualify for participation, both AD and NC subjects’ histories had to be free of psychiatricillness, substance abuse, head injury with loss of consciousness, secondary neurological disor-ders, or current use of medications that alter cognition. All AD subjects fulfilled NINCDS-ADRDA criteria for the diagnosis of probable AD both initially and after at least 1 year ofclinical follow-up (McKhann, Drachman, Folstein, Katzman, Price, & Stadlan, 1984). Diagno-ses were made after a comprehensive evaluation of cognition, review of medical histories,physical and neurological examinations, laboratory tests, and neuroimaging studies. Each ofthese had to be free of any evidence for ischemic vascular disease according to publishedcriteria (Chui, Victoroff, Margolin, Jagust, Shankle, & Katzman, 1992). Table 1 summarizesdemographic characteristics of the study samples.

222 WILLIS ET AL.

TABLE 1Demographic Characteristics of Study Samples

Performance accuracy sampleNC AD

(n 5 42) (n 5 26)

M (SD) M (SD)

Age (years) 72.5 (6.2) 75.0 (9.2)Education (years) 14.0 (2.3) 12.6 (3.9)Gender

Females 25 (60%) 21 (81%)Males 17 (40%) 5 (19%)

Performance time subsampleNC AD

(n 5 15) (n 5 13)

M (SD) M (SD)

Age (years) 72.1 (9.9) 74.8 (11.8)Education (years) 13.6 (1.8) 14.0 (2.1)Gender

Females 13 (87%) 13 (100%)Males 2 (3%) 0

Procedures

All subjects completed a comprehensive neuropsychological battery followed by a video-taped apraxia battery, both administered by three trained psychologists. Pilot testing estab-lished consistent test administration and scoring procedures for the three examiners and peri-odic consistency checks were carried out by comparing videotaped records against scoredprotocols from each. Results of the apraxia battery from 68 subjects were analyzed for perfor-mance accuracy and videotaped protocols from a subsample of 28 subjects were analyzed forperformance time.

Measures

Neruopsychological Battery

A neuropsychological battery covered a range of cognitive functions. Tests of attentionincluded Digit Span, Visual Span, and Mental Control, which contribute to the Attention Quo-tient of the Wechsler Memory Scale—Revised (Wechsler, 1987) and simple and choice reac-tion time subtests (DCAR 1 and 2) of the Denver Computer Assisted Retraining program(Pendleton, Meyerinck, & Hughes, 1985). A 10-item version of the Visual Form Discrimina-tion test (Benton, Hamsher, Varney, & Spreen, 1983) and copying accuracy on 6 items fromthe Biber Figure Learning Test (Glosser, Goodglass, & Biber, 1989) were used to evaluatevisual–spatial processing. Auditory discrimination was assessed with Word Discrimination,Body-Part Identification, and Commands from the Boston Diagnostic Aphasia Examination(Goodglass and Kaplan, 1983). Visual learning and memory were evaluated with a 6-itemversion of the Biber Figure Learning test, which includes five learning trials and 30-min de-layed recall and recognition trials (Glosser et al., 1989). Verbal learning and memory were

TIME AND ACCURACY MEASURES IN EARLY AD 223

assessed with a 12-item uncategorized list learning test, including five learning trials and 30-min delayed recall and recognition trials (Drachman, Glosser, Fleming, & Longenecker, 1982).

Upper extremity motor sequencing was assessed with five tasks similar to those developedby Luria (Christensen, Jensen, & Risberg, 1989). These included right- and left-hand rapidthumb-to-finger touching sequences, rapid alternation of right and left hands in fist down–palm down and in palm up–palm down sequences, and rapid serial positioning of the dominanthand. Standard test administration and scoring procedures were developed. Examiners firstmodeled the movement then elicited an imitation from the subject while still modeling it andfinally asked the subject to perform the movement independently for at least one completemovement cycle. This procedure was repeated a second time if the subject was unable to carryout the movement alone. If correct solo performance was not accomplished during the secondattempt, a score of 0 was assigned for that sequence. If the subject succeeded in one correctmovement cycle, they were instructed to repeat the movement as accurately and rapidly aspossible until told to stop. The score for each movement sequence was the number of correctcycles completed in 10 s.

Apraxia Battery

The apraxia battery contained 40 different tasks assessing upper extremity limb praxis. Thisincluded transitive and intransitive tasks (gestures requiring use of a tool or not), symbolicand nonsymbolic references (gestures conveying meaning or not), and trials using both domi-nant and nondominant hands. Items included in the battery can be found in the Appendix. Onall apraxia trials, subjects were asked to pantomime the gestures following a command embed-ded in the sentence frame: ‘‘Show me how you would. . . .’’ A correct gesture earned a scoreof 3. If a subject failed to respond within 10 s, after being prompted once, they were askedto imitate the gesture modeled by the examiner. Correct imitation gained a score of 2. Apraxiascores for task type and hand dominance were obtained for each subject by summing theirscores on all trials for that condition. The total apraxia score was the sum across all 40 trials.

Six preliminary error types were selected during pilot testing based on errors operationalizedin the literature (Haaland & Flaherty, 1984; Gonzalez et al., 1988; Shelton & Knopman, 1991;McDonald, Tate, & Rigby, 1994). Three additional error types were added, which could beconsistently identified. Two raters independently viewed taped sessions of pilot subjects todetermine performance on the battery and establish consistency in scoring errors. Seventy-nine percent of all error ratings for the two raters were identical. A nonparametric test ofdifferences found that the two raters’ scores of error types were not significantly different:Wilcoxon signed rank, t 5 .31, p 5 .76. There were nine performance errors, which couldbe reliably scored by two independent raters: No response (NR), an incomplete gesture (INC),a unique intrusion into a gesture (INTRU), an intruded fragment into an otherwise accurategesture (INTFR), a perseverated gesture (PER), a mislocated target of a gesture (MT), a handposition error (HP), a spatial orientation error (SO), and use of a body part as an object of agesture (BPO).

Timing

Videotaped records of 13 AD and 15 NC subjects were selected based on the ability tomatch subjects by age and education. Videotapes of these subjects’ performances on the batterywere analyzed at the Media Center of Whittier College using a JVC super-VHS editing deck,model BRS810U, and a Paltex Abner video editor controller. Each tape was viewed in 1/30-sframes by two psychologists (LW and MB) and from this process the criteria for consistentidentification of boundaries of movement segments were established. Based on these criteria,a final analysis was carried out by one rater (MB). In order to obtain apraxia timing measures,

224 WILLIS ET AL.

TABLE 2Composite and Total Scores on Neuropsychological and Apraxia Batteries

NC AD(n 5 42) (n 5 26)

M (SD) M (SD) SD delta

Index scoresAttention 1.83 (2.9) 236.04 (3.6)*** 13.0Auditory discrimination 2.52 (1.2) 25.64 (1.5)* 4.3Visual–spatial processing 2.33 (0.5) 23.3 (0.6)*** 5.9Motor sequencing 2.09 (0.2) 21.6 (0.2)*** 7.6Verbal learning and memory 20.26 (4.28) 212.31 (4.29)*** 2.8Visual learning and memory 20.14 (4.45) 216.45 (4.48)*** 3.7

Total scoresNeuropsychological battery 0.48 (3.9) 269.7 (0.2)*** 17.7Apraxia battery 0.41 (10.40) 221.99 (10.47)*** 2.1

Note. SD delta scores are the differences between the mean scores of the AD and NC groupsin terms of the NC standard deviation score units.

***p , .001; **p , .01; *p , .05.

which were independent of time associated with the commission of errors, only error-freetrials were analyzed. Three distinct periods could be reliably identified by this procedure.

Premovement period. The time between the end of the command and the beginning of anydetectable hand or arm movements.

Pregesture period. The time between the beginning of any hand or arm movement and thebeginning of movements clearly identified as the beginning of the requested gesture.

Gesture period. The time between the beginning and end of the correct gesture. Repetitivemovements, such as ‘‘stirring a cup of coffee with a spoon’’ were defined as one completemovement cycle, e.g., a stirring cycle.

Data Analysis

All individual test scores from the neuropsychological battery were standardized using themean and standard deviation of the NC group. Composite scores for specific cognitive andmotor functions were calculated adding the individual standardized scores on tests associatedwith that function. Composite scores were obtained in this manner for the total battery scoreand six functional indexes: Attention, visual–spatial processing, auditory discrimination, vi-sual learning and memory, verbal learning and memory, and motor sequencing. Tests contrib-uting to functional indexes are described above in the Measures section.

Initial analyses indicated similar group differences when pantomime scores or pantomimeplus imitation scores were analyzed in two separate ANCOVAs. In other words, adding imita-tion scores did not contribute to information about group or task contrasts. For this reason,pantomime scores alone were used in further analyses, a procedure followed in a prior studyof apraxia in AD (Rapcsak et al., 1989). The potentially confounding effects of age and educa-tion were examined by t tests. In order to compare groups on performance accuracy, scoreson the apraxia battery were evaluated by separate ANCOVAs, one for each task conditionwith group as a main factor, level of task (e.g., transitive vs intransitive) as a repeated measure,and education and auditory discrimination as covariates.

Error profiles for the AD and NC groups were constructed by first computing for each

TIME AND ACCURACY MEASURES IN EARLY AD 225

subject the percentage of total errors represented by each error type. Two NC subjects wereeliminated from this analysis because they made no errors. Group differences in error catego-ries were compared using a nonparametric procedure. To explore the relationships betweenapraxia and other performance domains, separate multiple linear regression analyses werecarried for each group using the total apraxia score as the dependent measure. Potential pre-dictor variables allowed to enter in a forward stepwise selection procedure were the six func-tional index scores, first forcing in education.

To evaluate timing differences between the two subsamples of AD and NC subjects, timingdata were analyzed by ANOVA. One analysis combined tasks and three others included levelof task (e.g., transitive vs intransitive) as a repeated measure. Gesture time in the nonsymboliccondition could not be reliably measured and therefore was not analyzed statistically. Sincegroups did not differ in age and education, these factors were not controlled in the timinganalyses.

RESULTS

There were significant group differences on neuropsychological andapraxia battery scores with consistently higher performance for the NCgroup. Relative differences, in terms of standard deviation units, are pre-sented in Table 2. Although differences were greater on the neuropsychologi-cal battery, all contrasts were significant. There was a marginal associationbetween total scores on the two batteries for the AD group after controllingfor education and auditory discrimination; Pearson Correlations, NC, r 5.26, p 5 .38; AD, r 5 .33, p 5 .10.

Analysis of both the total accuracy sample and the performance time sub-sample did not show significant group differences in age or education (Table1). However, since there was a trend for the NC group to have more educa-tion in the total sample (t 5 1.93, p 5 .058), subsequent analyses of accuracywere adjusted for this factor. Measures of verbal comprehension have beenassociated with pantomime performance (Foster et al., 1986; Alexander,Baker, Naeser, Kaplan, & Palumbo, 1992; Wang and Goodglass, 1992). Tocontrol for the possibility that pantomime results would be thus confounded,analyses also adjusted for comprehension of spoken language using the audi-tory discrimination composite score as a covariate.

Performance Accuracy

There was a significant difference between AD and NC groups in perfor-mance accuracy on the apraxia battery for all three task conditions with con-sistently worse performance by the AD group (Table 3). Only symbolic andnonsymbolic tasks differed with both groups obtaining lower scores on sym-bolic tasks. Results of the multiple linear regression analysis for the ADgroup identified motor sequencing as the only significant predictor of accu-racy on the apraxia battery, contributing 38% (p , .01) to the variability inperformance. In the regression analysis for the NC group, no variables suc-ceeded in entering into the model, indicating that none significantly predictedaccuracy of performance on the battery.

226 WILLIS ET AL.

TABLE 3Group Comparisons for Apraxia Battery: Accuracy and Time Measures

Performance accuracy

AD NC(n 5 26) (n 5 42)

M (SEM) M (SEM ) Significance level

Transitive/intransitive Group**Transitive 40.6 [1.1] 45.3 [0.8]Intransitive 40.2 [1.3] 46.0 [0.9]

Dominant/nondominant Group**Dominant 40.9 [1.1] 45.9 [0.8]Nondominant 39.9 [1.2] 45.4 [0.8]

Symbolic/nonsymbolic Group**Symbolic 19.7 [0.8] 22.9 [0.6] Task*Nonsymbolic 21.6 [0.4] 23.4 [0.3]

Performance time by movement periodAD NC

(n 5 26) (n 5 42)

M (SD) M (SD) Significance level

Transitive/intransitivePremovement Group**

Transitive 8.8 (4.3) 3.3 (2.3)Intransitive 11.9 (2.1) 3.4 (1.9)

Pre-gesture Group**Transitive 39.6 (31.6) 14.4 (6.9) Group by task**Intransitive 16.1 (7.8) 8.6 (4.1)

GestureTransitive 12.7 (3.6) 11.7 (5.3)Intransitive 9.4 (4.1) 8.4 (2.8)

Dominant/nondominantPremovement Group***

Dominant 10.8 (6.3) 4.4 (2.6)Nondominant 10.0 (10.1) 2.3 (2.2)

Pregesture Group**Dominant 18.3 (9.1) 12.7 (7.3) Group by task*Nondominant 37.4 (34.7) 10.2 (3.2)

Gesture Group by task*Dominant 10.1 (2.9) 10.8 (5.5)Nondominant 11.9 (3.5) 9.2 (2.3)

Symbolic/nonsymbolicPremovement Group***

Symbolic 6.5 (4.9) 2.4 (1.5)Nonsymbolic 4.3 (3.1) 1.4 (1.6)

Pregesture Group*Symbolic 7.6 (4.9) 4.8 (3.8)Nonsymbolic 6.5 (5.3) 4.3 (1.0)

Note. Values in brackets are the standard error of measurement.***p , .001; **p , .01; *p , .05.

TIME AND ACCURACY MEASURES IN EARLY AD 227

FIG. 1. Percentage of error types for AD and NC groups. Error types: INTRU 5 intrusion;BPO 5 body part as object; INC 5 incomplete gesture; INTFR 5 intruded fragment intootherwise correct gesture; MT 5 mislocated target; HP 5 hand position error; SO 5 spatialorientation error; NR 5 no response; PER 5 perseveration.

Errors

The error patterns of AD and NC groups can be seen in Fig. 1. All nineerror types were committed by the AD group, contrasting with only fourtypes of errors made by the NC group: Incomplete gestures (INC), use ofbody part as object (BPO), intrusions (INTRU), and hand position errors(HP). Significant differences were found between AD and NC groups forthese four types: INC, p , .05; BPO p , .0001; INTRU p , .0001; HPp , .0001. AD and NC groups showed similar patterns of most frequenterror types: INC (AD 5 20%, NC 5 18.5%), BPO (AD 5 31%, NC 564%), and INTRU (AD 5 29%, NC 5 16.5%).

Timing

Results of the apraxia timing analysis combing tasks showed significantmain effects for both group and time: group, F(1, 26) 5 19.24, p , .001;time, F(2, 52) 5 20.76, p , .0001. There was also a significant group-by-time interaction, F(2, 52) 5 6.71, p , .01. Post hoc comparisons revealedsignificantly longer times for the AD group on both premovement and pre-gesture periods (p , .05), but no group difference in gesture time. Longertimes for the AD group compared with the NC group were particularly prom-inent for the pregesture period, as shown in Figure 2a.

Each task also was analyzed separately with task level as a repeated mea-

228 WILLIS ET AL.

FIG. 2. Significant differences in response times for group and task conditions. Period 1represents premovement time. Period 2 represents time after movement occurs, but beforegesture movement. Period 3 represents gesture execution time.

sure. Results are shown in Table 3 and in Figures 2b, 2c, and 2d. Groupdifferences were significant for all task conditions at both premovement andpregesture periods, but not for the gesture period, with AD times consistentlylonger than NC times. There were significant group-by-transitivity andgroup-by-dominance interactions for the pregesture period, reflecting the rel-atively longer time taken by AD subjects prior to execution of a gesturefor transitive tasks and when using their nondominant hand. A group-by-dominance interaction for the gesture period resulted from a combination oflonger times using the right (dominant) hand by the NC group and longertimes using the left hand for the AD group. Since the apraxia battery re-quested use of the right hand before the left, shorter gesture times using theleft hand by the NC group probably reflected a practice effect. By contrast,longer gesture times using the left hand for AD subjects must indicate greaterdifficulty employing the nondominant hand, since they also had receivedpractice with the right hand.

DISCUSSION

Although apraxia is traditionally described as a relatively late feature ofAD (Cummings & Benson, 1983), the results of this investigation demon-strate its presence in patients with mild to moderate levels of dementia at a

TIME AND ACCURACY MEASURES IN EARLY AD 229

time when the disorder is not seen as a prominent symptom in the clinic orreported by family members. Differences between the AD and NC groupsin this study were highly significant in terms of both gesture accuracy andtiming. However, standardized scores on the apraxia battery did not differen-tiate the two groups as much as the total score or domain specific compositescores derived from a neuropsychological battery. This finding is consistentwith prior reports, which place apraxia in early AD at a lower severity levelcompared with other areas of cognitive functioning. Unlike results from priorstudies (Edwards et al., 1991; Rapcsak et al, 1989), dementia level was notsignificantly correlated with apraxia severity. However, in the previous in-vestigations cognition and apraxia were not studied exclusively in early de-mentia when distributions of scores may be narrow for both variables, fa-voring low intercorrelations.

In this study, AD subjects made a larger number of error types than NCsubjects, but the pattern of most frequent errors did not differ greatly betweengroups. Patient subjects made more intrusion errors, consistent with the ap-pearance of intrusions in other performance areas in AD (Fuld, 1983;Salmon, Granholm, McCullough, Butters, & Grant, 1989). BPO errors werethe most common type for both AD and NC groups, a result that is in keepingwith previous findings (Goodglass & Kaplan, 1963; Haaland & Flaherty,1984; Gonzalez Rothi et al., 1988; Rapcsak et al., 1989; Shelton & Knopman,1991; Alexander et al., 1992; McDonald, 1994).

Transitive tasks are reported to be more difficult than intransitive onesfor apraxic patients (Goodglass & Kaplan, 1963; Kertesz & Hooper, 1982;Haaland & Flaherty, 1984; Rapcsak et al., 1989; Edwards et al., 1991; Raadeet al., 1991). Although there was no difference between transitive and intran-sitive tasks in terms of response accuracy, the AD group had significantlylonger response latencies than the NC group prior to execution of transitivegestures. We propose that added time was needed in transitive tasks to acti-vate mnemonic representations of both the required movement sequence andthe required target object. This is related to the interpretation of a findingby Ochipa and associates, i.e., that AD patients have difficulty evoking repre-sentations of appropriate actions as well as of appropriate objects of actionsduring their use of tools (Ochipa et al., 1992).

An extensive research literature on apraxia supplies few systematic in-vestigations of differences between dominant and nondominant hand use(De Renzi, Motti, & Nichelli, 1980; Lehmkuhl, Poeck, & Willmes, 1983;Rapcsak et al., 1989; Alexander, et al., 1992). In the present study perfor-mance accuracy of the two hands did not differ, but pregesture latencies weresignificantly longer for the AD group when the left (nondominant) hand wasused. In addition, several patient subjects showed marked nonfluidity ofmovement when attempting to execute gestures with their left hands. Lesspractice in performing skilled movements with the nondominant hand over

230 WILLIS ET AL.

the lifetime may result in greater vulnerability of skilled gestures usingthat hand as overall cerebral functioning declines with the progress of de-mentia.

Symbolic gestures were less accurate than nonsymbolic ones for both theAD and NC groups. Wang and Goodglass (1992) have proposed that mecha-nisms underlying the execution of symbolic and nonsymbolic tasks may bequite different. Meaningful (e.g., symbolic) gestures may be interpreted andstored semantically, whereas meaningless ones may require only temporaryvisuokinesthetic storage for accurate reproduction. AD patients in particularmay have greater difficulty executing symbolic gestures since their disorderis characterized by a breakdown in semantic processing (Ober, Dronkers,Koss, Delis, & Friedland, 1986; Binetti, Magni, Cappa, Padovani, Bian-chetti, & Trabucchi, 1995; Chan, Butters, Salmon, Johnson, Paulsen, &Swenson, 1995).

The results of this study furnish needed quantitative information aboutperformance characteristics of apraxia in early-stage AD. The results alsoindicate that the timing of gestures may provide a more sensitive indicationof problems with praxis than accuracy of gestures alone for mild or moder-ately impaired AD patients.

APPENDIX

Tasks Included in the Apraxia Battery According to Task Condition

Transitive SymbolicBlow out a matchPeal a potato with a knifeBrush your teethUse a hammer to hammer a nailDial a telephoneWrite with a penUse a key to open a door

Intransitive SymbolicWave goodbyeSnap your fingersHitchhikeSalute like a soldierSignal stop with your handBlow a kissSignal quiet with your finger to your lipsMake a ‘‘V’’ for victory

Intransitive NonsymbolicPut your hand on opposite shoulderPut your fist on your chestPut your hand over your ear

TIME AND ACCURACY MEASURES IN EARLY AD 231

Put your palm on your foreheadPut your fingertips on your chinPoint to your foreheadPoint to your noseMake a circle in the air

REFERENCES

Alexander, M. P., Baker, E., Naeser, M. A., Kaplan, E., & Palumbo, C. 1992. Neuropsycholog-ical and Neuroanatomical dimensions of ideomotor apraxia. Brain, 115, 87–107.

Benton, A. L., Hamsher, K. de S., Varney, N., & Spreen, O. 1983. Contributions to Neuropsy-chological Assessment. New York: Oxford University Press.

Binetti, G., Magni, E., Cappa, S. F., Padovani, A., Bianchetti, A., & Trabucchi, M. 1995.Semantic memory in Alzheimer’s disease: An analysis of category fluency. Journal ofClinical and Experimental Neuropsychology, 17, 82–89.

Chan, A. S., Butters, N., Salmon, D. P., Johnson, S. A., Paulsen, J. S., & Swensen, M. R.1995. Comparison of the semantic networks in patients with dementia and amnesia. Neu-ropsychology, 9, 177–186.

Christensen, A. L., Jensen, L. R., & Risberg, J. 1989. Luria’s neuropsychological and neurolin-guistic testing. Journal of Neurolinguistics, 4, 137–154.

Chui, H. C., Victoroff, J. I., Margolin, R., Jagust, W., Shankle, R., & Katzman, R. 1992.Criteria for the diagnosis of ischemic vascular dementia proposed by the State of Califor-nia Alzheimer’s Disease Diagnostic and Treatment Centers. Neurology, 42, 473–480.

Cummings, J. L., & Benson, D. F. 1983. Dementia: A clinical approach. Boston, MA: Butter-worths.

Della Sala, S., Lucchelli, F., & Spinnler, H. 1987. Ideomotor apraxia in patients with dementiaof Alzheimer type. Journal of Neurology, 234, 91–93.

De Renzi, E., Motti, F., & Nichelli, P. 1980. Imitating gestures: A quantitative approach toideomotor apraxia. Archives of Neurology, 37, 6–10.

Drachman, D. A., Glosser, G., Fleming, P., & Longenecker, G. 1982. Memory decline in theaged: Treatment with lecithin and physostigmine. Neurology, 32, 944–950.

Edwards, D. F., Deuel, R. K., Baum, C. M., & Morris, J. C. 1991. A quantitative analysis ofapraxia in senile dementia of the Alzheimer type: Stage-related differences in prevalenceand type. Dementia, 2, 142–149.

Folstein, M. F., Folstein, S. E., & McHugh, P. R. 1975. Mini-mental state: A practical methodfor grading the cognitive state of patients for the clinician. Journal of Psychiatric Re-search, 12, 189–198.

Foster, N. L., Thomas, N., Chase, T. N., Patronas, H. J., Gillespie, M. M., & Fedio, P. 1986.Cerebral mapping of apraxia in Alzheimer’s disease by positron emission tomography.Annals of Neurology, 19, 139–143.

Foundas, A. L., Macauley, B. L., Raymer, A. M., Maher, L. M., Heilman, K. M., & GonzalezRothi, L. J. 1995. Ecological implications of limb apraxia: Evidence from mealtime be-havior. Journal of the International Neuropsychological Society, 1, 62–66.

Fuld, P. A. 1983. Word intrusion as a diagnostic sign in Alzheimer’s disease. Geriatric Medi-cine Today, 2, 33–41.

Glosser, G., Goodglass, H., & Biber, C. 1989. Assessing visual memory disorders. Journalof Consulting and Clinical Psychology, 1, 82–91.

232 WILLIS ET AL.

Goodglass, H., & Kaplan, E. 1963. Disturbance of gesture and pantomime in aphasia. Brain,86, 703–720.

Goodglass, H., & Kaplan, E. 1983. Boston Diagnostic Aphasia Examination. Philadelphia,PA: Lea & Febiger.

Haaland, K. Y., & Flaherty, D. 1984. The different types of limb apraxia errors made bypatients with left vs right hemisphere damage. Brain and Cognition, 3, 370–384.

Heilman, K. M., & Gonzalez Rothi, L. J. 1985. Apraxia. In K. M. Heilman and E. Valenstein(Eds.), Clinical neuropsychology. New York: Oxford University Press. Pp. 141–150.

Kertesz, A., & Hooper, P. 1982. Praxis and language: The extent and variety of apraxia inaphasia. Neuropsychologia, 20, 275–286.

Lehmkuhl, G., Poeck, K., & Willmes, K. 1983. Ideomotor apraxia and aphasia: An examina-tion of types and manifestations of apraxic symptoms. Neuropsychologia, 21(3), 199–212.

Lucchelli, F., Lopez, O. L., Faglioni, P., Boller, F. 1993. Ideomotor and ideational apraxia inAlzheimer’s disease. International Journal of Geriatric Psychiatry, 8, 413–417.

McDonald, S., Tate, R. L., & Rigby, J. 1994. Error types in ideomotor apraxia: A qualitativeanalysis. Brain and Cognition, 25, 250–270.

McKhann, G., Drachman, D., Folstein, M. F., Katzman, R., Price, D., & Stadlan, E. M. 1984.Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA Work Groupunder auspices of the Department of Health and Human Services Task Force on Alzhei-mer’s Disease. Neurology, 34, 939–944.

Ober, B. A., Dronkers, N. F., Koss, E., Delis, D. C., & Friedland, R. P. 1986. Retrieval fromsemantic memory in Alzheimer-type dementia. Journal of Clinical and ExperimentalNeuropsychology, 8, 75–92.

Ochipa, C., Gonzalez Rothi, L. J., & Heilman, K. M. 1992. Conceptual Apraxia in Alzheimer’sdisease. Brain, 115, 1061–1071.

Pendleton, M., Meyerinck, L. H., & Hughes, R. B. 1985. Changes in neuropsychological testperformances following computerized assisted cognitive retraining. Journal of Clinicaland Experimental Neuropsychology, 7, 155.

Poeck, K. 1986. The clinical examination of motor apraxia. Neuropsychologia, 24, 129–134.

Raade, A. S., Gonzalez Rothi, L. J., & Heilman, K. M. 1991. The relationship betweenbuccofacial and limb apraxia. Brain and Cognition, 16, 130–146.

Rapcsak, S. Z., Croswell, S. C., & Rubens, A. B. 1989. Apraxia in Alzheimer’s disease. Neu-rology, 39, 664–668.

Rothi, L. J. G., Mack, L., Verfaellie, M., Brown, P., & Heilman, K. M. 1988. Aphasiology,2, 381–388.

Roy, E. A., & Square, P. A. 1985. Common considerations in the study of limb, verbal, and oralapraxia. In E. A. Roy (Ed.), Neuropsychological studies of apraxia and related disorders.Amsterdam: North-Holland. Pp. 111–161.

Salmon, D. P., Granholm, E., McCullough, D., Butters, N., & Grant, I. 1989. Recognitionmemory span in mildly and moderately demented patients with Alzheimer’s disease.Journal of Clinical and Experimental Neuropsychology, 11(4), 429–443.

Shelton, P. A., & Knopman, D. S. 1991. Ideomotor apraxia in Huntington’s disease. Archivesof Neurology, 48, 35–41.

Taylor, R. 1994. Motor apraxia in dementia. Perceptual and Motor Skills, 79, 523–528.

Travniczek-Marterer, A., Danielczyk, W., Simanyi, M., & Fischer, P. 1993. Ideomotor apraxiain Alzheimer’s disease. Acta Scandinavica, 88, 1–4.

TIME AND ACCURACY MEASURES IN EARLY AD 233

Wang, L., & Goodglass, H. 1992. Pantomime, praxis, and aphasia. Brain and Language, 42,402–418.

Wechsler, D. 1987. The Wechsler Memory Scale—Revised. New York: Harcourt Brace Jova-novich.

Yesavage, J. A., Brooks, J. O., Taylor, J., & Tinklenberg, J. 1993. Development of aphasia,apraxia, and agnosia and decline in Alzheimer’s disease. American Journal of Psychiatry,150(5), 742–747.