Neuropsychology

2000, Vol. 14, No. 1,29-40Copyright 2000 by the American Psychological Association, Inc.

0894-4105/00/S5.00 DOI: 10.1037//0894-4105.14.1.29

Structural Brain Correlates of Verbal and Nonverbal Fluency Measuresin Alzheimer's Disease

Rosemary Fama, Edith V. Sullivan,Paula K. Shear, Deborah A. Cahn-Weiner,

and Laura MarshStanford University School of Medicine

Kelvin O. Lim, Jerome A. Yesavage,and Jared R. Tinklenberg

Stanford University School of Medicineand VA Palo Alto Health Care System

Adolf PfefferbaumSRI International

This study examined the relationships between regional brain volumes and semantic,phonological, and nonverbal fluency in 32 participants with Alzheimer's disease (AD). Objectbut not animal semantic fluency correlated with frontal and temporal gray matter volumes.Phonological fluency was not significantly associated with any brain volume examined.Nonverbal fluency was selectively associated with bilateral frontal gray matter volumes.Hippocampal volumes, although markedly reduced in these patients, were not related to any ofthe fluency measures. Results lend evidence to the importance of the frontal lobes in thedirected generation of nonverbal and verbal exemplars by AD patients. Furthermore, both left-and right-hemisphere regions contribute to the generation of verbal and nonverbal exemplars.

The ability to generate verbal or nonverbal material

rapidly according to specified rules, as is demanded in

Rosemary Fama, Edith V. Sullivan, Paula K. Shear, Deborah A.Cahn-Weiner, and Laura Marsh, Department of Psychiatry andBehavioral Sciences, Stanford University School of Medicine;Kelvin O. Lim, Jerome A. Yesavage, and Jared R. Tinklenberg,Department of Psychiatry and Behavioral Sciences, StanfordUniversity School of Medicine and Psychiatry Service, VA PaloAlto Health Care System; Adolf Pfefferbaum, NeuropsychiatryProgram, SRI International, Menlo Park, California.

Rosemary Fama is now at the Neuropsychiatry Program, SRIInternational, Menlo Park, California. Paula K. Shear is now at theDepartments of Psychology and Psychiatry, University of Cincin-nati. Deborah A. Cahn-Weiner is now at the Memorial Hospital ofRhode Island, Brown University. Laura Marsh is now at theDepartment of Psychiatry, Johns Hopkins Hospital, Baltimore,Maryland. Kelvin O. Lim is now at the Nathan S. Kline Institute forPsychiatric Research, Orangeburg, New York.

Portions of this research were presented at the 26th AnnualMeeting of the International Neuropsychological Society, Hono-lulu, Hawaii, February 1998. This research was supported byNational Institute of Mental Health Grants MH18905, MH40041,and MH30854, National Institute on Aging Grant AG11427,National Institute on Alcohol Abuse and Alcoholism GrantsAA05965 and AA10723, and grants from the Department ofVeterans Affairs and the State of California.

We thank Kenneth Chow for his statistical assistance, BrianMatsumoto for assistance in image analysis, and Jennifer Johnsonand our other dedicated research assistants for their invaluablework in participant recruitment, data collection, scoring, andcomputer entry.

Correspondence concerning this article should be addressed toEdith V. Sullivan, Department of Psychiatry and BehavioralSciences, Stanford University School of Medicine (5717), Stan-ford, California 94305-5717. Electronic mail may be sent toedie@leland.stanford.edu.

fluency tasks, is commonly impaired in patients with frontal

lobe lesions (Baldo & Shimamura, 1998; Benton, 1968;

Jones-Gotman & Milner, 1977; Milner, 1964; Ferret, 1974;

Ramier & Hecaen, 1970). More recent brain imaging work

using functional magnetic resonance imaging (MRI) and

positron emission tomography protocols suggests that verbal

fluency produces activation of the frontal lobes, particularly

the prefrontal cortex in the language dominant hemisphere

(Frith, Friston, Liddle, & Frackowiak, 1991; Parks et al.,

1988; Phelps, Hyder, Blamire, & Shulman, 1997; Pujol et

al., 1996). The active search and retrieval of information

necessary for adequate performance on fluency tasks are

believed to be primarily mediated by the frontal lobes

(Baldo & Shimamura, 1998).

The majority of studies investigating brain correlates of

fluency performance have involved patients with focal

lesions; fewer studies have investigated the brain—behavior

correlates of fluency performance in patients with wide-

spread or multifocal brain dysfunction. Although fluency

performance is generally deficient in degenerative disorders

such as Alzheimer's disease (AD), it is not yet known

whether these deficits are related to specific brain regions or

instead are reflective of more diffuse dysfunction. Fluency

ability can be assessed with verbal—phonological (i.e.,

letters of the alphabet) and semantic (i.e., categories such as

animals or objects)—and nonverbal (i.e., design) tasks. Both

verbal and nonverbal fluency abilities are impaired in AD

(Butters, Granholm, Salmon, Grant, & Wolfe, 1987; Fama et

al., 1998; A. Martin & Fedio, 1983; Mickanin, Grossman,

Onishi, Auriacombe, & Clark, 1994; Monsch et al., 1992;

Rosser & Hodges, 1994). Within the verbal domain, seman-

tic fluency has been reported to be even more impaired than

phonological fluency on average (Butters et al., 1987). The

relatively greater deficit in semantic fluency has been

29

30 FAMA ET AL.

interpreted as evidence of a breakdown in semantic structure

in AD, which is consistent with the associated neuropathol-

ogy of the neocortical temporal area. Nonverbal fluency can

be as impaired as semantic fluency in AD patients (Fama et

al., 1998). Differences in the severity and pattern of impair-

ment among different fluency tasks have provided informa-

tion about the underlying neural and processing mechanisms

involved in dementing conditions (Butters et al., 1987;

Monsch et al., 1994; Rosser & Hodges, 1994).

Studies on individuals with focal lesions reveal that

fluency deficits are not restricted to patients with frontal lobe

impairment and can accompany more posterior lesions

(Bolter, Long, & Wagner, 1983; Jones-Gotman & Milner,

1977; Miceli, Caltagirone, Gainotti, Masullo, & Silveri,

1981; Newcombe, 1969; Pendleton, Heaton, Lehman, &

Hulihan, 1982; Stuss et al., 1998). Pendleton et al. (1982)

reported that participants with documented frontal lesions

and those with nonfrontal and diffuse lesions performed

significantly worse than control participants on a verbal

fluency task, with no difference in performance between

patient groups with the two different lesion sites. On a

nonverbal fluency task, Jones-Gotman and Milner (1977)

reported that although the greatest impairment occurred in

participants with right frontal lesions, participants with right

nonfrontal lesions were also impaired on this task. Thus,

fluency tasks are sensitive to frontal lobe dysfunction, yet

their specificity and utility in the localization of brain

dysfunction remains unresolved (cf. Monsch et al., 1992).

Laterality of a lesion can influence fluency performance.

Deficits on verbal fluency tasks are more strongly associated

with left- than right-hemisphere lesions, whereas deficits on

nonverbal or design fluency tasks are more strongly associ-

ated with right- than left-hemisphere lesions (Benton, 1968;

Jones-Gotman & Milner, 1977; Lee, Strauss, McCloskey,

Loring, & Drane, 1996; Miceli et al., 1981; Milner, 1964;

Ramier & Hecaen, 1970; Ruff, 1988; Ruff, Allen, Farrow,

Niemann, & Wylie, 1994). These findings were based on

patients with surgical resections or relatively large space-

occupying lesions. Despite evidence for material-specific

processing and hemispheric laterality, right-hemisphere le-

sions can influence efficiency in verbal fluency performance

(Joanette & Goulet, 1986; R. C. Martin, Loring, Meador, &

Lee, 1990), and left-hemisphere lesions can influence effi-

ciency in nonverbal fluency performance (Jones-Gotman &

Milner, 1977). Joanette and Goulet (1986) found an associa-tion between right-hemisphere dysfunction and semantic but

not phonological fluency deficits in 35 individuals who

suffered unilateral vascular based lesions to the right hemi-

sphere, and R. C. Martin et al. (1990) found an association

between right-hemisphere dysfunction and both semanticand phonological fluency performance in 32 patients with

unilateral temporal lobe epilepsy, 15 patients with left-

hemisphere focus, and 17 patients with right-hemisphere

focus. More recently, Baldo and Shimamura (1998) ob-

served that patients with right as well as those with left

MRI-documented unilateral lesions performed significantly

worse than control participants on semantic and phonologi-

cal fluency tasks. Additionally, Parks et al. (1988) reportedactivation of the left and right frontal and temporal lobes in

normal participants when performing verbal fluency tasks,

providing further evidence for right-hemisphere influence

on verbal fluency performance.

Category-specific deficits occur where, within the seman-

tic domain for example, an individual may be impaired in

naming exemplars from particular semantic categories (ani-

mate objects, such as animals) but unimpaired for other

semantic categories (inanimate objects, such as tools; see

Devlin, Gonnerman, Andersen, & Seidenberg, 1998; Farah

& McClelland, 1991; Silveri, Daniele, Giustolisi, & Gain-

otti, 1991; Warrington & McCarthy, 1987; Warrington &

Shallice, 1984, for reviews). Whether differences in perfor-

mance between semantic categories reflect the workings of

relatively independent brain systems is not yet clear. It has

been hypothesized that the ability to generate names of

animals may be more reliant on the ability to associate from

one animal to another on the basis of perceptual features

(e.g., what the animal looks or feels like), whereas the ability

to generate names of inanimate objects may be more reliant

on the ability to associate from functional features (e.g.,

what an object is used for; see Devlin et al., 1998, for a

review). Additionally, the perceptual information used to

generate exemplars to animate categories may be mediated

by the temporal-limbic regions, whereas functional informa-

tion used to generate exemplars to inanimate categories may

be mediated by the frontal-parietal regions (Damasio,

Grabowski, Tranel, Hichwa, & Damasio, 1996; A. Martin,

Haxby, Lalonde, Wiggs, & Ungerleider, 1995; A. Martin,

Wiggs, Ungerleider, & Haxby, 1996; Mummery, Patterson,

Hodges, & Wise, 1996; Perani et al., 1995; Silveri et al., 1991).Whether category-specific deficits noted in AD are due to

impairment in specific anatomical regions (e.g., knowledge

of animals associated with temporal regions; knowledge of

nonanimate objects associated with frontal and parietal

regions) or rather a function of the nature of the representa-

tion of semantic knowledge (e.g., exemplars associated by

perceptual features being more interrelated than exemplars

associated by functional features) has been debated (cf.

Garrard, Patterson, Watson, & Hodges, 1998; Gonnerman,

Andersen, Devlin, Kempler, & Seidenberg, 1997). Thus,

different semantic category tasks may be reflective of the

integrity of different neural pathways.

In this study, we investigated whether the verbal and

nonverbal fluency abilities of patients with AD were selec-

tively associated with the volumes of regional cortical and

hippocampal volumes. On the basis of previous research, we

tested the following hypotheses: (a) that scores on semantic,

phonological, and nonverbal fluency tests would be related

to frontal lobe gray matter volumes because of the produc-

tion and retrieval processes required for these tasks; (b) that

semantic fluency would be related to gray matter volumes of

the temporal lobe of both hemispheres; (c) that nonverbal

fluency performance would be related to parietal lobe gray

matter volumes because of the visuospatial component of

the task; (d) that the verbal fluency tasks would be more

strongly associated with left-hemisphere volumes and that

the nonverbal fluency tasks would show a stronger associa-

tion with right-hemisphere volumes; and (e) that although

hippocampal abnormalities are a hallmark feature of AD and

MRI AND FLUENCY IN AD 31

are related to disease severity and explicit memory perfor-

mance in AD (see Cahn et al., 1998; Fama et al., 1997),

fluency performance would not be related to hippocampalvolumes.

Method

Participants

Participants in this study included 32 AD patients (23 men, 9women) recruited from the Geriatric Psychiatry Rehabilitation Unitand the National Institute of Mental Health Aging Clinical Re-search Center, both housed at the VA Palo Alto Health Care Systemin California. All AD patients met the National Institute ofNeurological and Communicative Diseases and Stroke and Alzhei-mer's Disease and Related Disorders Association criteria forpossible or probable AD (Khachaturian, 1985; McKhann, Drach-man, Folstein, & Katzman, 1984). Normal control participantsspanning the adult age range (20-84 years of age) were used as areference group for the MRI volumetric measures. These partici-pants, or a subset of them, have formed the norms for other studiesfrom our laboratory (Cahn et al., 1998; Fama et al., 1997;Pfefferbaum et al., 1994; Sullivan, Marsh, Mathalon, Lim, &Pfefferbaum, 1995a). The normal control group was composed of95 men and 41 women for the axial MRI protocol and 84 men and28 women for the coronal protocol. The reference group for thefluency measures consisted of 51 normal control participants,spanning the ages of the AD participants (age = 66.7 ±7.4 years;education = 16.4 ± 2.3 years). The fluency performance of thesecontrol participants was used to calculate standardized Z scores forthe AD participants that corrected the raw scores for age. Thesefluency data and group results were reported in a previous article byFama et al. (1998), which reported a comparison of the pattern ofverbal and nonverbal fluency deficits in patients with AD andpatients with Parkinson's disease.

Screening for all participants included a psychiatric interviewand medical examination. Potential participants were excluded ifthey had significant history of psychiatric or neurological disordernot related to their diagnosis (e.g., stroke, closed head injury), pastor present alcohol or drug abuse or dependence, or other seriousmedical condition. Informed consent was obtained from all partici-pants. Demographic information for the AD group is summarizedin Table 1.

Neuropsychological Measures

Verbal Fluency Tests

The semantic fluency test consisted of two 1-min trials: Partici-pants first generated names of animals and then generated names ofinanimate objects. The semantic fluency score was the meannumber of correct unique responses given across the two trials. The

phonological fluency test required participants to generate wordsthat began with the letter F, then A, and finally S (excluding propernouns and the same word repeated with different suffixes) in three1-min trials (Borkowski, Benton, & Spreen, 1967). The phonologi-cal fluency score was the total number of different correct wordsproduced across the three trials.

Nonverbal Fluency Test

Nonverbal fluency was assessed with the Ruff Figural FluencyTest (RFFT; Ruff, 1988; Ruff, Light, & Evans, 1987), whichrequires participants to generate as many different designs aspossible by using straight lines to connect arrays of five dots. Thetest consists of five trials, each with a 1-min time limit. Thenonverbal fluency score was the total number of correct uniquedesigns summed across the five trials.

Not all participants received all tests. A given participant had tohave phonological fluency and at least figural or semantic fluencydata along with an MRI scan within 3 months of neuropsychologi-cal testing to be included in the study. Of the 32 AD patients withMRI data included in this study, 23 received figural fluency, 32phonological fluency, and 31 semantic fluency.

MRI Scanning and Quantification

Axial Protocol

Acquisition parameters. Participants were scanned with 1.5TGeneral Electric Signa MRI scanners (Milwaukee, WT). Imageacquisition procedures and parameters have been previously de-scribed in detail in other studies (see Lim & Pfefferbaum, 1989;Pfefferbaum et al., 1994; Zipursky, Lim, Sullivan, Brown, &Pfefferbaum, 1992). Axial MRI images were 5 mm thick (2.5 mmskip) and were acquired in an oblique plane using a spin-echosequence (20- and 80-ms echoes), with a 24-cm field of view and a256 X 256 matrix. Acquisition was gated to every other cardiaccycle for an effective repetition time (TR) of >2,400 ms, with oneexcitation for each of 256 phase encodes.

All images were stored on magnetic tape and transferred tooptical disks for analysis. For each participant, the index slice wasidentified as the most inferior slice above the level of the orbits,where the anterior horns of the lateral ventricles could be seenbilaterally. Seven consecutive slices, beginning with the sectioninferior to the index slice and proceeding superiorly, were analyzedfor each participant.

Regional divisions and segmentation of images. Each MRIslice was segmented into cerebrospinal fluid (CSF), gray matter,and white matter compartments, using a semiautomated imageanalysis technique (Lim & Pfefferbaum, 1989; Lim, Zipursky,Wans, & Pfefferbaum, 1992). To separate the cerebral hemispheres,we drew a midline manually on each slice. Additionally, each slicewas divided into an inner 55% region (to facilitate quantification of

Table 1

Demographics for Alzheimer's Disease Participants

Measure

MSD

Age(years)

70.96.8

Age ofsymptom

onset (years)

65.97.7

Duration ofdiagnosis

5.03.9

Education(years)

15.03.7

MMSE

18.84.1

DRS

108.916.7

NARTIQ

105.79.1

Note. There were 23 male and 9 female participants with Alzheimer's disease. MMSE = Mini-Mental State Exam (Folstein, Folstein, &. McHugh, 1975); DRS = Dementia Rating Scale (Mattis,1976, 1988); NART = National Adult Reading Test (Nelson, 1982).

32 FAMA ET AL.

central CSF, which arose primarily but not exclusively from thelateral ventricles) and an outer 45% (to facilitate quantification ofthe cortical tissue volumes and sulcal CSF; Pfefferbaum et al.,1992).

The images were divided according to anatomical landmarks anda priori geometric rules in an effort to achieve standardized regionaldivisions of the brain images. The cortical tissue measure (the outer45% of each image) was divided into six anatomically andgeometrically defined regions of interest, which roughly corre-sponded to lobar anatomy. The regions did not fully volume thecortical lobes after which they were named but provided a reliablebasis for dividing cortical sections (Zipursky et ah, 1992). To formthese divisions, we divided each MRJ slice into four regions bythree coronal planes, which passed through the most anteriorextreme of the genu of the corpus callosum, the most posteriorextreme of the splenium of the corpus callosum, and midwaybetween them. The first plane established a boundary for theprefrontal region. From these quadrants and slices, we devised sixcortical regions, defined as follows: prefrontal—the most anteriorquadrant of all seven slices; frontal—the anterior middle quadrantof Slices 3-7; anterior superior temporal—the anterior middlequadrant of Slices 1 and 2, which included the anterior superiortemporal gyrus and the most posterior extents of the frontal lobes atthe level of the superior temporal gyrus; posterior superiortemporal—the posterior middle quadrant of Slices 1 and 2, whichincluded the posterior superior temporal gyrus and the anteriorextents of the parietal lobes just above the superior temporal gyrus;anterior parietal—the posterior middle quadrant of Slices 3-7; andposterior parietal-occipital—the most posterior quadrant of Slices3-7, which also included much of the occipital lobes (see Figure I).

Coronal Protocol

A detailed description of the acquisition methods, anatomicalborders, and reliability of measurement technique used in ottr

laboratory to examine the temporal lobes and hippocampus hasbeen reported previously (Sullivan et al., 1995a; Sullivan, Marsh,Mathalon, Lim, & Pfefferbaum, 1995b). A summary of theacquisition parameters and regional divisions and segmentation ofimages is presented below.

Acquisition parameters. With this protocol, 22 contiguous3-mm thick coronal images were acquired using a multiecho,flow-compensated cardiac-gated pulse sequence (echo time[TE] = 40, 80 ms; effective TR = 2,800 ms) with field of viewequal to 24 cm, number of excitations (NEX) = 1, and a 256 X 256matrix. The plane of image acquisition was oriented perpendicularto the anterior commissure—posterior commissure line. Imageacquisition was specifically designed to encompass temporal-limbic structures, beginning 6 mm anterior to the anterior commis-sure and extending 66 mm posteriorly. Hippocampal tissue couldnot be adequately segmented into gray matter and white mattercompartments, and, therefore, its volume reflected total tissue (i.e.,pixels represented gray matter plus white matter). The hippocam-pus was outlined on each consecutive slice on each side of thebrain, and the volumes were derived by adding the areas of eachmeasured slice.

Regional divisions and segmentation of images. The anteriorlimit of the hippocampal measurement began where hippocampaltissue was clearly distinguished from amygdala, coinciding withthe appearance of the digitations of Ammon's horn and consistentwith the location of the pes hippocampus. This initial slice usuallycoincided with the visualization of ventricular temporal horn CSFover the digitations. Because the boundary between the subiculumand the amygdala cannot be distinguished at the medial aspect ofthe hippocampal formation at this slice level, we drew a line fromthe most medial point of the subiculum to die most medial aspect ofthe temporal horn, thereby excluding amygdala from the measure.The posterior limit corresponded to the posterior temporal lobeimage, as detailed below. This measure predominantly includedportions of the head and body of the hippocampus, with exclusion

b ~ H\'«+™ ••

Figure 1. Each measured magnetic resonance imaging slice was divided into an outer 45%, whichrepresented cortical regions, and an inner 55%, which included ventricular regions. The cortical areawas then divided into regions that roughly correspond to lobar anatomy: a indicates prefrontal, Slices1-7; b indicates frontal, Slices 3-7; c indicates anterior superior temporal, Slices 1 and 2; d indicatesposterior superior temporal. Slices 1 and 2; e indicates parietal, Slices 3-7; and f indicates posteriorparietal-occipital, Slices 3-7. The volume of each cortical region was divided into gray matter(shown in dark gray), white matter (light gray), and cerebrospinal fluid (black).

MRI AND FLUENCY IN AD 33

of the most anterior portion of the pes (where it could not bedistinguished from the amygdala) and the most posterior portion ofthe hippocampus (the ascending tail). The media] hippocampalboundary included the regional outline at the choroidal fissure. Theinferior boundary was formed by, but did not include, the whitematter of the parahippocampal gyrus. The fornix was not includedin posterior slices unless it was inextricably embedded within thehippocampal structure. In the absence of histological confirmation,this measure presumably included hippocampal fields CA1 to CA4,the dentate gyrus, and portions of the subiculum (Amaral &Insausti, 1990; see Figure 2).

Statistical Analysis

The volumes of each brain region of interest in the AD patientswere corrected for variation attributable to intracranial volume (i.e.,head size) and age through regression analyses described in detailin previous articles (Mathalon, Sullivan, Rawles, & Pfefferbaum,1993; Pfefferbaum et al., 1994). Briefly, a correction using atwo-step regression analysis to adjust for normal variation in headsize and age that was based on normal control participants yieldedhead size and age-corrected Z scores. All regional brain measureswere expressed as Z scores, where the expected mean of the controlparticipants at any age was 0 ± 1. This correction method allowsfor analysis of disease-related changes that are independent of theknown effects of normal aging and head size. For the ADparticipants, these head size and age-corrected Z scores providedvolume estimates relative to that which would be expected fornormal control participants of a particular head size and age. LowerZ scores for tissue measures reflect greater abnormality. Fluencyscores were also standardized to allow direct comparisons acrosstasks to be made. Raw scores were converted to age-correctedstandardized Z scores on the basis of data from 51 controlparticipants because of significant correlations found between age

and semantic (r = -.40) and nonverbal (r = -.37) fluency in asimilar aged control group (see Fama et al., 1998).

Relationships between the age-corrected standardized fluencymeasures and the head size and age-corrected standardized re-gional, bilateral and lateralized (left and right), brain volumemeasures were examined with Pearson product-moment correla-tions. We used a familywise Bonferroni correction (6 correlationsfor each fluency measure and bilateral brain volumes and 12correlations for each fluency measure and lateralized brain vol-umes) when determining whether a correlation was statisticallysignificant. We interpreted a correlation as being significant if itreached the p ^ .008 level for bilateral brain volume correlationsand p £ .004 for lateralized brain volume correlations. Statisticaltrends are reported for correlations at the p ^ .05 level. Multipleregression models were used to examine whether the brain volumesof specific brain regions contributed independently to the predic-tion of fluency scores after the contributions of other relevant brainregions were taken into account. Finally, to ensure that therelationships observed between regional brain volumes and fluencymeasures were not due to the influence of gender on thesevariables, we used partial correlations as follow-up analyses.

Results

Fluency Performance

When compared with age-appropriate normal controlparticipants, AD participants were significantly impaired onall fluency measures (group differences, p < .01). Controlgroup raw scores were as follows: phonological (F, A, S) =41.2 ± 12.9, semantic = 24.1 ± 5.1 (animals = 22.1 ± 4.4,objects = 26.1 ± 7.3), and nonverbal (figural) = 76.1 ±21.1 (see Fama et al., 1998). Mean scores on the fluency

Figure 2. Coronal magnetic resonance image of a 42-year-old normal healthy man. The left sideshows an outline of regions of interest—temporal lobes and hippocampi. The right side showssegmentation of the image into gray matter (dark gray), white matter (light gray), and cerebrospinalfluid (black).

34 FAMA ET AL.

N

Cortical Gray Matter Volumes

• AD(n=32)

Hippocampal Volumes

I AD(n=19)

If

Figure 3. Z scores depicting cortical and hippocampal volume deficits of the Alzheimer's disease(AD) participants included in this study compared with a control group of similarly aged normalparticipants. Sup. Temp. = superior temporal; Par.-Occ. = parietal-occipital.

tasks for the AD group were as follows: phonological (F, A,

S) = 19.9 ± 10.9, semantic = 8.3 ± 4.4 (animals =

8.2 ± 4.4, objects = 8.4 ± 4.9), and nonverbal (figural) =

25.3 ± 12.3. All three fluency scores were significantly

correlated with the Mini-Mental State Examination (MMSE;

Folstein, Folstein, & McHugh, 1975), a measure of dementia

severity (phonological r = .40, p < .01; semantic r = .53,

p<.0l; nonverbal r = .44, p < .05). None of the raw

fluency scores was significantly correlated with age, years of

education, or years since diagnosis. Age-corrected Z scores

for the AD group indicated greater than 1.5 to 3 SD deficits

in fluency scores: phonological (F, A, S) = -1.7 ± 0.85,

semantic = -2.8 ± 0.86 (animals = -3.2 ± 1.1, objects =

-2.4 ± 0.78), and nonverbal (figural) = -2.4 ± 0.82.

MRI Volume Abnormalities

Head size and age-corrected Z scores indicated that theAD group had almost 0.5 to 1 5D deficits in gray matter

volume present throughout the cortex (see Figure 3; Z

scores: prefrontal = -0.53 ± 1.2, frontal = -0.79 ± 0.93,

anterior superior temporal = —0.74 ± 0.96, posterior supe-

rior temporal = -0.97 ± 0.88, anterior parietal = -0.76 ±

0.94, and posterior parietal-occipital = —0.38 ± 1.1; total

cortical = -0.87 ± 1.2 SD from the expected mean). In

addition, the hippocampus was abnormally small in volume

bilaterally (Z scores for left = -0.77 ± 1.1, right = -0.70 ±

1.3; also see Cahn et al., 1998; Fama et al., 1997).

Regional Brain Volume Correlates of Verbaland Nonverbal Fluency Measures

Correlations With Bilateral Brain Volumes

No brain-behavior correlation reached statistical signifi-

cance when overall semantic fluency scores were examined

(see Table 2). When animal and object scores were analyzed

separately, there was a trend toward a relationship between

object fluency Z scores and bilateral gray matter volume Z

scores of the frontal (r = .43, p = .015) and the posterior

superior temporal (r = A0,p = .026) regions (see Figure 4).

As the scatterplots indicate, there was 1 AD patient who did

not show a deficit on object fluency (Z score = 0). Post hoc

analyses with this participant removed indicated that al-

though the nature of the relationship between these brain

volumes and object fluency scores remained positive, the

significance level of these correlations dropped (frontal

r = .33, p < .08, posterior superior temporal r = .33,

p < .08). Animal fluency scores did not significantly corre-

late with any of the bilateral brain volumes measured.

To test whether the correlations between animal fluency

and object fluency were significantly different from one

Table 2

Correlations Between Fluency and Bilateral Gray Matter Volumes

Brain region

FrontalPrefrontalAnterior superior temporalPosterior superior temporalAnterior parietalPosterior parietal-occipitalTotal cortical

Semantic"(« = 31)

.31

.27

.27

.26

.15

.12

.26

Animals(1 = 31)

.19

.17

.18

.13

.09

.07

.15

Objects(n = 31)

.43*

.35

.35

.40*

.20

.17

.37*

Phonological(n = 32)

-.07.04

-.10.03

-.10-.02

.02

Nonverbal(n = 23)

.57**

.2849*

.41

.42*

.18

.38

aSemantic refers to animals plus objects.*p < .05. **p < .008.

MRI AND FLUENCY IN AD 35

r̂ i -1<D

1-1.5

1 -2

•2.5

-3

-3.5'-3.5 -3 -2.5 -2 -1.5 -1 -.5

Object Fluency Z-score

Posterior Superior Temporal Gray Matter

.5 •

0- .

-5N .1

1-1.51§ -2.

-25

-3

-3.5

-4 -3.5 -3 -2.5 -2 -1.5 -1 -.5Object Fluency Z-score

Figure 4. Object fluency and bilateral volume. A trend toward a relation was found between object fluency scores and gray matter volumesof the frontal and posterior superior temporal regions.

another, we calculated t values using Fisher z transforma-

tions. The correlation between posterior superior temporal

volume and object fluency was significantly greater than the

correlation between the same brain region and animal flu-

ency, f(30) = 2.2, p < .05. The difference between the

correlations for total cortical volume and object and animal

fluency scores approached significance, t(30) = 1.7, p < . 10.

Phonological fluency scores were not significantly corre-

lated with any bilateral brain volume measured.

Nonverbal fluency scores were significantly correlated

with frontal lobe volumes (r = .51, p = .005), and statistical

trends were noted with the bilateral volume of the anterior

superior temporal (r = .49, p = .018) and anterior parietal

(r = .42, p = .048) regions (see Figure 5). A multiple

regression analysis indicated that the incremental proportion

of variance in nonverbal fluency accounted for by the frontal

gray matter volume, after accounting for the contributions ofthe anterior superior temporal and anterior parietal volumes

of gray matter, approached significance (p < .06). Neither

of the two nonfrontal regions made significant unique

contributions to nonverbal fluency performance after the

other regions were accounted for (see Table 3). To establish

that the nonverbal fluency scores were not simply reflective

of generalized gray matter volume loss, we performed a

multiple regression that predicted nonverbal fluency perfor-

mance from both frontal gray matter and total cortical gray

matter. Frontal (p < .02) but not total cortical gray matter

(p = .33) volume made a significant independent contribu-

tion to the prediction of nonverbal fluency scores.

Correlations With Interallied Brain Volumes

A statistical trend was present between overall semantic

fluency scores and volume of the right anterior superior

temporal region (r = .37, p = .04; see Table 4). When the

animal and object trials were analyzed separately, animal

fluency scores were not significantly correlated with any of

the lateralized brain volumes measured, but object fluency

showed modest correlations with a number of brain regions:

gray matter volumes of the left frontal (r = .44, p =

.014), right prefrontal (r = .40, p = .025), right frontal

(r = 39,p = .032), right anterior superior temporal (r = .42,

p = .018), and right posterior superior temporal (r = .42,

p = .018) regions (see Figure 6). Post hoc analyses with the

AD participant who showed no deficit on this measure

removed indicated that the nature of these brain-behavior

relationships remained the same; however, significance

levels dropped throughout, with only the correlation be-

tween object fluency and right posterior superior temporal

volumes reaching a statistical trend (r = .36, p < .05): left

frontal r = .33,p < .08; right frontal r = 30,p = .11; right

prefrontal r = .29, p = .13; right anterior superior temporal

r = .29, p = .13.

-1.5

-2

-2.5

Frontal Gray Matter Volume

-3.5 -3 -2.5 -2 -1.5Total Unique Designs

8 0

3 -1

-1.5

-2

-2.5

-3

Anterior Superior TemporalGray Matter Volume

-3.5 -3 -2.5 -2 -1.5 -1 -.5Total Unique Designs

1.5

1

.5'

S 0

<a -.5-

N -I-

-1.5'

-2-

•2.5

-3'

Anterior Parietal Qray Matter Volume

-3.5 -3 -2.5 -2 -1.5 -1 --5Total Unique Designs

Figure 5. Nonverbal fluency and bilateral volume. Nonverbal fluency scores were significantlycorrelated with frontal gray matter volumes. A trend toward a relation between nonverbal fluencyscores and anterior superior temporal and anterior parietal regions was noted; poorer performancewas related to smaller gray matter volumes.

36 FAMA ET AL.

Table 3

Multiple Regressions Predicting Nonverbal Fluency Scores

From Brain Volumes

Dependent measure Predictors

Nonverbal fluency Frontal .708t .133Anterior superior temporal .173 .020Anterior parietal -.406 .048

Nonverbal fluency FrontalOverall cortical

tp < .06. *p < .05.

.712*-.225

.212

.031

Phonological fluency scores were not significantly corre-

lated with any lateralized MRI volume measures.

Nonverbal fluency scores were significantly correlated

with gray matter volumes of the right anterior superior

temporal region (r = .61, p = .002; see Figure 7). Strong

statistical trends were found between nonverbal fluency

scores and left frontal (r = .57, p = .005) and right frontal

(r = .53, p = .009) volumes. Statistical trends were also

noted with right posterior superior temporal (r = .43,

p = .043) and left anterior parietal regions (r = .42,

p = .045). Amultiple regression model predicting nonverbal

fluency performance from left frontal and right frontal gray

matter volumes indicated that neither region independently

contributed to nonverbal fluency scores once the contribu-

tion of the other region was taken into account, left frontal

t(22) = 1.18, p = .25, right frontal r(22) = 0.42, p = .68.

Fisher z calculations indicated that the correlation for

nonverbal fluency scores and right posterior superior tempo-

ral volumes was significantly greater than the correlation for

nonverbal fluency scores and left posterior superior temporal

volumes, t(22) = 2.3, p < .05. No differences were found

between any other corresponding left and right correlation.

Nonverbal, phonological, and semantic fluency scores were

not significantly correlated with hippocampal volumes of

either hemisphere.

Partial Correlations Examining Gender Influence on

Brain-Behavior Relationships

All reported significant correlations between regional

brain volumes and fluency scores remained when partial

correlations were calculated to control the influence of

gender.

Discussion

These results demonstrate that in AD, certain fluency

measures are related to specific cortical regions, which have

volume deficits. By contrast, hippocampal volume, although

associated with explicit memory performance in AD (Fama

et al., 1997; Killiany et al., 1993; Laakso et al., 1995; Stout,

Jernigan, Archibald, & Salmon, 1996), was not related to

fluency performance. As predicted, nonverbal fluency scores

showed selective associations with frontal lobe gray matter

volumes (cf. Jones-Gotman & Milner, 1977; Lee et al.,

1997; Ruff et al., 1994). Despite widespread cortical and

hippocampal volume deficits in these AD patients, nonver-

bal fluency showed predicted relationships with regional

cortical and not hippocampal volumes. These results demon-

strate that nonverbal fluency performance is not simply

related to volume loss in indiscriminate areas of the brain or

to total cortical gray matter volume, also demonstrably

affected in AD; rather, nonverbal fluency is associated with

selective and predicted brain volumes. However, whereas

Jones-Gotman and Milner (1977) reported that in patients

with surgical lesions, the right frontal lobe was more

strongly associated with nonverbal fluency performance

than the left frontal lobe, our data did not show lateralized

effects in the frontal lobes. Differences in results between

our study and that of Jones-Gotman and Milner may be due

to the different populations studied; the present study used

individuals with a degenerative disorder affecting various

cortical and subcortical structures, whereas Jones-Gotman

and Milner studied individuals with focal lesions. Also,

different nonverbal fluency tasks were used in each study.

Design fluency (used by Jones-Gotman & Milner, 1977)

requires generation of novel and different nonnameable

figures without external guidelines, whereas the RFFT (used

in the present study) provides additional constraints in the

instructions of using only straight lines within the predeter-

mined array of five dots presented. It may be that the RFFT

more strongly relies on the bilateral integrity of the frontal

lobes than does design fluency. Alternatively, the association

between the frontal lobes and nonverbal fluency may, in

part, signify the importance of posterior cortical regions for

this task, by way of frontal-posterior neural connections. It is

also possible that the nonverbal fluency task showed associa-

tion with the frontal lobes, whereas the verbal fluency tasks

Table 4Fluency — Lateralized Gray Matter Volumes

Semantic"

Brain region

FrontalPrefrontalAnterior superior temporalPosterior superior temporalAnterior parietalPosterior parietal-occipitalTotal corticalHippocampus

Left

.34

.16

.14

.11

.08

.17

.22

.03

Right

.25

.34

.37*

.32

.18

.07

.28

.06

Animals

Left

.23

.06

.06

.01

.03

.13

.13

.04

Right

.11

.26

.29

.21

.13

.01

.17

.04

Objects

Left

.44*

.26

.22

.26

.15

.19

.32

.12

Right

.39*

.40*

.42*

.42*

.22

.14

.39*

.08

Phonological

Left

.01

.14

.15

.09

.18

.03

.06

.21

Right

.13

.20

.02

.14

.03

.06

.02

.24

Nonverbal

Left

.57***

.22

.31

.28

.42*

.22

.37

.19

Right

.53***

.31

.61****

.43*

.39

.12

.37

.13

"Semantic refers to animals plus objects.*p<.05. ***p<.01. ****p<.004.

MRI AND FLUENCY IN AD 37

Left Frontal Gray Matter

-4 -35 -3 -2.5 -2 -1.5 -1 -.5Object Fluency Z-score

Right Frontal Gray Matter

-3.5 -3 -2.5 -2 -1.5 -1 -.5Object Fluency Z-score

Right Prefrontal Gray MatterRight Anterior Superior Temporal Gray Matter

-4 -3.5 -3 -2.5 -2 -1.5 -1 -.5Object Fluency Z-score

-4 -3.5 -3 -2.5 -2 -1.5 -1 -.5 0Object Fluency Z-score

3

2o>

8 1V

1°

r-3

Right Posterior Superior Temporal Gray Matter

-4 -3.5 -3 -2.5 -2 -1.5 -1 -.5 0Object Fluency Z-score

Figure 6. Object fluency and bilateral brain volume. Object fluency scores showed modestcorrelations with a number of lateralized regional brain volumes: right prefrontal, left frontal, rightfrontal, right anterior superior temporal, and right posterior superior temporal.

did not because of the differences in complexity between the

nonverbal and verbal fluency tasks.

Although the total semantic score was not associated with

frontal lobe volume, there was evidence that object fluency

was related to frontal lobe volume. Object fluency, but not

animal fluency, showed modest associations with a number

of regional brain volumes, namely, left and right frontal,

right prefrontal, right anterior superior temporal, and right

Left Frontal Gray Matter

-2

-2.5

~-4 -3.5 -3 -2.5 -2 -1.5 -1 -.5Total Unique Designs (Z-score}

1.5

1

£ .5

^ -.5

I i-2.5

-3

Left Anterior Parietal Gray Matter

-3.5 -3 -2.5 -2 -1.5 -1Total Unique Designs (Z-score)

-4 -3.5...-3

Right Anterior Superior Temporal Gray Matter

-3.5 -3 -2.5 -2 -1.5 -1Total Unique Designs (Z-score)

•3.5 -3 -2.5 -2 -1.5 -1Total Unique Designs (Z-score)

1.51-

I *« 0'

I ,5

5-1.5-

-2

-2.5

Right Posterior Superior Temporal Gray Matter

-4 -3.5 -3 -2.5 -2 -1.5 -1 -.5Total Unique Designs (Z-score)

Figure 7. Nonverbal fluency and lateralized brain volume. Nonverbal fluency scores weresignificantly correlated with gray matter volumes of the right anterior superior temporal region.Statistical trends were noted between nonverbal fluency scores and gray matter volumes of the leftfrontal, right frontal, right posterior superior temporal, and left anterior parietal regions.

38 FAMA ET AL.

posterior superior temporal. This study lends evidence to the

notion that generation of animal names is dissociable from

generation of object names and that these abilities are

subserved, at least in part, by different brain systems (cf.

Damasio et al., 1996; A. Martin et al., 1995; Perani et al.,

1995; Silveri et al., 1991). It may also be that object fluency

denotes a more general category of exemplars than does

animal fluency. Thus, the associations between object flu-

ency and frontal regions may reflect the greater need of

organization of search for this more general category.

Dissociations between semantic fluency categories have

been shown with functional imaging studies. Using positron

emission tomography, A. Martin et al. (1996) showed that

although there was bilateral activation of the ventral tempo-

ral lobes and Broca's area when right-handed participants

named animals and tools from pictures; selective activation

of the left medial occipital lobe occurred when animals were

named, whereas activation of the left premotor and left

middle temporal gyrus occurred when tools were named.

We, like others (Pasquier, Lebert, Grymonprez, & Petit,

1995; Pendleton et al., 1982; Ruff et al., 1994), did not find

significant associations between phonological fluency and

integrity of the frontal regions. It may be that phonological

fluency is sensitive to generalized brain dysfunction yet not

be particularly useful in localization of dysfunction. Failure

to find selective relationships between fluency performance

and brain volumes may reflect the likelihood that successful

fluency performance requires several processing systems

subserved by various anterior and posterior brain regions

(Bornstein, 1986; R. C. Martin et al., 1990). Alternatively,

the specific brain regions contributing to the phonological

deficit may not have been included in our analyses. Use of

imaging protocols using thinner slices or covering the full

extent of the brain would provide further opportunity to

identify specific regions contributing to performance. Even

in instances in which selective brain-behavior relationships

are present, we are not proposing that the particular fluency

ability is localizable to a specific region. Instead, we believe

that these selective relationships indicate the importance of

the particular brain region in the successful completion of

the specific task.

In conclusion, performance on fluency tests, whether

verbal or nonverbal, relies on the efficient generation of

exemplars within specific constraints, a process degraded in

AD. This study provides evidence that the degeneration of

nonverbal fluency performance is attributable, at least in

part, to structural volume loss in the gray matter of selective

cortical areas but not to hippocampal loss. Furthermore,

these brain-behavior relationships indicate involvement of

both the left and the right hemispheres in the generation of

nonverbal and verbal exemplars.

References

Amaral, D. G., & Insausti, R. (1990). Hippocampal formation. InG. Paxinos (Ed.), The human nervous system (pp. 711-755).New York: Academic Press.

Baldo, J. V., & Shimamura, A. P. (1998). Letter and categoryfluency in patients with frontal lobe lesions. Neuropsychology,12, 259-267.

Benton, A. L. (1968). Differential behavioral effects in frontal lobedisease. Neuropsychologia, 6, 53-60.

Bolter, J. E, Long, C. J., & Wagner, M. (1983). The utility of theThurstone Word Fluency Test in identifying cortical damage.Clinical Neuropsychology, 5, 77—82.

Borkowski, J. G., Benton, A. L., & Spreen, O. (1967). Wordfluency and brain damage. Neuropsychologia, 5, 135-140.

Bornstein, R. A. (1986). Contribution of various neuropsychologi-cal measures to detection of frontal lobe impairment. TheInternational Journal of Clinical Neuropsychology, VIII, 18-22.

Butters, N., Granholm, E., Salmon, D. P., Grant, A., & Wolfe, J.(1987). Episodic and semantic memory: A comparison ofamnesic and demented patients. Journal of Clinical and Experi-mental Neuropsychology, 9, 479-497.

Cahn, D. A., Sullivan, E. V., Shear, P. K., Marsh, L., Fama, R., Lim,K. O., Yesavage, J. A., Tinklenberg, J. R., & Prefferbaum, A.(1998). Structural MRI correlates of recognition memory inAlzheimer's disease. Journal of the International Neuropsycho-logical Society, 4, 106-114.

Damasio, H., Grabowski, T. J., Tranel, D., Hichwa, R. D., &Damasio, A. R. (1996, April 11). A neural basis for lexicalretrieval. Nature, 380, 499-505.

Devlin, J. T., Gonnerman, L. M., Andersen, E. S., & Seidenberg,M. S. (1998). Category-specific semantic deficits in focal andwidespread brain damage: A computational account. Journal ofCognitive Neuroscience, 10, 77-94.

Fama, R., Sullivan, E. V., Shear, P. K., Cahn-Weiner, D. A.,Yesavage, J. A., Tinklenberg, J. R., & Pfefferbaum, A. (1998).Fluency performance patterns in Alzheimer's disease and Parkin-son's disease. The Clinical Neuropsychologist, 12, 487-499.

Fama, R., Sullivan, E. V., Shear, P. K., Marsh, L., Yesavage, J. A.,Tinklenberg, J. R., Lim, K. O., & Pfefferbaum, A. (1997).Selective cortical and hippocampal volume correlates of MattisDementia Rating Scale in Alzheimer disease. Archives ofNeurology, 54, 719-728.

Farah, M., & McClelland, J. (1991). A computational model ofsemantic memory impairment: Modality specificity and emer-gent category specificity. Journal of Experimental Psychology:General, 120, 339-357.

Folstein, M. E, Folstein, S. E., & McHugh, P. R. (1975).Mini-mental state: A practical method for grading the cognitivestate of patients for the clinician. Journal of Psychiatric Re-search, 12, 189-198.

Frith, C. D., Friston, K. J., Liddle, P. P., & Frackowiak, R. S. J.(1991). A PET study of word finding. Neuropsychologia, 29,1137-1148.

Garrard, P., Patterson, K., Watson, P. C., & Hodges, J. R. (1998).Category specific semantic loss in dementia of Alzheimer's type.Brain, 121, 633-646.

Gonnerman, L. M., Andersen, E. S., Devlin, J. T, Kempler, D., &Seidenberg, M. S. (1997). Double dissociation of semanticcategories in Alzheimer's disease. Brain and Language, 57,254-279.

Joanette, Y., & Goulet, P. (1986). Criterion-specific reduction ofverbal fluency in right brain-damaged right-handers. Neuropsy-chologia, 24, 875-879.

Jones-Gotman, M., & Milner, B. (1977). Design fluency: Theinvention of nonsense drawings after focal cortical lesions.Neuropsychologia, 15, 653-674.

Khachaturian, Z. S. (1985). Diagnosis of Alzheimer's disease.Archives of Neurology, 42, 1097-1105.

MRI AND FLUENCY IN AD 39

Killiany, R. J., Moss, M. B., Albert, M. S., Sander, T, Tieman, J., &Jolesz, F. (1993). Temporal lobe regions on magnetic resonanceimaging identify patients with early Alzheimer's disease. Ar-chives of Neurology, 50, 949-954.

Laakso, M. P., Soininen, H., Partanen, K., Helkala, E.-L., Harti-kainen, P., Vainio, P., Hallikainen, M., Hanninen, T., & Riekki-nen, P. J. (1995). Volumes of hippocampus, amygdala and frontallobes in the MRI-based diagnosis of early Alzheimer's disease:Correlation with memory functions. Journal of Neural Transmis-sion, 9, 73-86.

Lee, G. P., Strauss, E., Loring, D. W., McCloskey, L., Haworth,J. M., & Lehman, R. A. W. (1997). Sensitivity of figural fluencyon the Five-Point Test to focal neurological dysfunction. TheClinical Neuropsychologist, 11, 59-68.

Lee, G. P., Strauss, E., McCloskey, L., Loring, D., & Drane, D. L.(1996, February). Localization of frontal lobe lesions usingverbal and nonverbal fluency measures. Poster session presentedat the 24th annual meeting of the International Neuropsychologi-cal Society, Chicago.

Lim, K. O., & Pfefferbaum, A. (1989). Segmentation of MR brainimages into cerebrospinal fluid spaces, white and gray matter.Journal of Computer Assisted Tomography, 13, 588-593.

Lim, K. O., Zipursky, R. B., Watts, M. C., & Pfefferbaum, A.(1992). Decreased grey matter in normal aging: An in vivomagnetic resonance study. Journal of Gerontology, 47, B26-B30.

Martin, A., & Fedio, P. (1983). Word production and comprehen-sion in Alzheimer's disease: The breakdown of semantic knowl-edge. Brain and Language, 19, 124-141.

Martin, A., Haxby, J. V., Lalonde, F. M., Wiggs, C. L., &Ungerleider, L. G. (1995, October 6). Discrete cortical regionsassociated with knowledge of color and knowledge of action.Science, 270, 102-104.

Martin, A., Wiggs, C. L., Ungerleider, L. G., & Haxby, J. V. (1996,February 15). Neural correlates of category-specific knowledge.Nature, 379, 649-655.

Martin, R. C., Loring, D. W., Meador, K. J., & Lee, G. P. (1990).The effects of lateralized temporal lobe dysfunction on formaland semantic word fluency. Neuropsychologia, 28, 823-829.

Mathalon, D. H., Sullivan, E. V., Rawles, J. M., & Pfefferbaum, A.(1993). Correction for head size in brain-imaging measurements.Psychiatry Research: Neuroimaging, 50, 121-139.

Mattis, S. (1976). Mental status examination for organic mentalsyndrome in the elderly patient. In L. Bellack & T. B. Karasu(Eds.), Geriatric psychiatry (pp. 77-120). New York, Grune &Stratum.

Mattis, S. (1988). Dementia Rating Scale (DRS) professionalmanual. Odessa, FL: Psychological Assessment Resources.

McKhann, G., Drachman, D., Folstein, M., & Katzman, R. (1984).Clinical diagnosis of Alzheimer's disease: Report of the NINCDS-ADRDA Work Group under the auspices of Department ofHealth and Human Services Task Force on Alzheimer's disease.Neurology, 37, 939-944.

Miceli, G., Caltagirone, C., Gainotti, G., Masullo, C., & Silveri, M.(1981). Neuropsychological correlates of localized cerebrallesions in non-aphasic brain-damaged patients. Journal of Clini-cal Neuropsychology, 3, 53-63.

Mickanin, J., Grossman, M., Onishi, K., Auriacombe, S., & Clark,C. (1994). Verbal and nonverbal fluency in patients withprobable Alzheimer's disease. Neuropsychology, 8, 385-394.

Milner, B. (1964). Some effects of frontal lobectomy in man. InJ. M. Warren & D. Akert (Eds.), Frontal granular cortex andbehavior (pp. 313-334). New York: McGraw-Hill.

Monsch, A. U., Bondi, M., Butters, N., Paulsen, J. S., Salmon,D. P., Brugger, P., & Swenson, M. R. (1994). A comparison of

category and letter fluency in Alzheimer's and Huntington'sdisease. Neuropsychology, 8, 25—30.

Monsch, A. U., Bondi, M. W., Butters, N., Salmon, D. P., Katzman,R., & Thai, L. J. (1992). Comparisons of verbal fluency tasks inthe detection of dementia of the Alzheimer type. Archives ofNeurology, 49, 1253-1258.

Mummery, C. J., Patterson, K., Hodges, J. R., & Wise, R. J. S.(1996). Generating "tiger" as an animal name or a wordbeginning with R: Differences in brain activation. Proceedingsof the Royal Society of London, Series B. (Biological Sciences),263, 989-995.

Nelson, H. E. (1982). The National Adult Reading Scale (NART).Windsor, England: Nelson.

Newcombe, F. (1969). Missile wounds of the. brain: A study ofpsychological deficits. London: Oxford University Press.

Parks, R. W., Loewenstein, D. A., Dodrill, K. L., Barker, W. W.,Hopshii, F., Chang, J. Y., Emran, A., Apicella, A., Sheramata,W. A., & Duara, R. (1988). Cerebral metabolic effects of a verbalfluency test: A PET scan study. Journal of Clinical andExperimental Neuropsychology, 10, 565-575.

Pasquier, F., Lebert, F., Grymonprez, L., & Petit, H. (1995). Verbalfluency in dementia of frontal lobe type and dementia ofAlzheimer type. Journal of Neurology, Neurosurgery, and Psy-chiatry, 58, 81-84.

Pendleton, M. G., Heaton, R. K., Lehman, R. A. W., & Hulihan, D.(1982). Diagnostic utility of the Thurstone Word Fluency Test inneuropsychological evaluations. Journal of Clinical Neuropsy-chology, 4, 307-317.

Perani, D., Cappa, S. F, Bettinardi, V., Bressi, S., Gorao-Tempini,M., Matarrese, M., & Fazio, F. (1995). Different neural systemsfor the recognition of animals and man-made tools. NeuroRe-port,6, 1637-1641.

Ferret, E. (1974). The left frontal lobe of man and the suppressionof habitual responses in verbal categorical behaviour. Neuropsy-chologia, 12, 323-330.

Pfefferbaum, A., Lim, K. O., Zipursky, R. B., Mathalon, D. H.,Lane, B., Ha, C. N., Rosenbloom, M. J., & Sullivan, E. V. (1992).Brain gray and white matter volume loss accelerates with agingin chronic alcoholics: A quantitative MRI study. Alcoholism:Clinical and Experimental Research, 16, 1078-1089.

Pfefferbaum, A., Mathalon, D. H., Sullivan, E. V, Rawles, J. M.,Zipursky, R. B., & Lim, K. O. (1994). A quantitative magneticresonance imaging study of changes in brain morphology frominfancy to late adulthood. Archives of Neurology, 51, 874—887.

Phelps, E. A., Hyder, P., Blamire, A. M., & Shulman, R. G. (1997).fMRl of the prefrontal cortex during overt verbal fluency.NeuroReport, 8, 561-565.

Pujol, J., Vendrell, P., Kulisevsky, J., Marti-Vilalta, J. L., Garcia,C., Junque, C., & Capdeveila, A. (1996). Frontal lobe activationduring word generation studied by functional MRI. Acta Neuro-logica Scandinavica, 93, 403-410.

Ramier, A., & Hecaen, H. (1970). Role respectif des atteintesfrontales et de la lateralisation lesionnelle dans les deficits de lafluence verbale [Respective roles of frontal lesions and lesionlateralization in "verbal fluency" deficiencies]. Revue Neu-rologique, 123, 17—22.

Rosser, A. E., & Hodges, J. R. (1994). The dementia rating scale inAlzheimer's disease, Huntington's disease and progressive supra-nuclear palsy. Journal of Neurology, 241, 531-536.

Ruff, R. (1988). Ruff Figural Fluency Test: Administration manual.San Diego, CA: Neuropsychological Resources.

Ruff, R. M., Allen, C. C., Farrow, C. E., Niemann, H., & Wylie, T.(1994). Figural fluency: Differential impairment in patients withleft versus right frontal lobe lesions. Archives of ClinicalNeuropsychology, 9, 41—55.

40 FAMA ET AL.

Ruff, R. M., Light, R. H., & Evans, R. W. (1987). The Ruff FiguralFluency Test: A normative study with adults. DevelopmentalNeuropsychology, 3, 37-51.

Silveri, M. C., Daniele, A., Giustolisi, L., & Gainotti, G. (1991).Dissociation between knowledge of living and nonliving thingsin dementia of the Alzheimer type. Neurology, 41, 545-546.

Stout, J., Jernigan, T, Archibald, S. L., & Salmon, D. P. (1996).Association of dementia severity with cortical gray matter andabnormal white matter volumes in dementia of the Alzheimertype. Archives of Neurology, S3, 742-749.

Stuss, D. T, Alexander, M. P., Hamer, L., Palumbo, C., Dempster,

R., Binns, M., Levine, B., & Izukawa, D. (1998). The effects offocal anterior and posterior brain lesions on verbal fluency.Journal of the International Neuropsychological Society, 4,

265-278.Sullivan, E. V., Marsh, L., Mathalon, D. H., Lim, K. O., &

Pfefferbaum, A. (1995a). Age-related decline in MRI volumes of

temporal lobe gray matter but not hippocampus. Neurobiology ofAging, 16, 591-606.

Sullivan, E. V, Marsh, L., Mathalon, D. H., Lim, K. O., &Pfefferbaum, A. (1995b). Anterior hippocampal volume deficitsin nonamnesic, aging chronic alcoholics. Alcoholism: Clinicaland Experimental Research, 19, 110-122.

Warrington, E. K., & McCarthy, R. A. (1987). Categories ofknowledge. Brain, 110, 1273-1296.

Warrington, E. K., & Shallice, T. (1984). Category specific

semantic impairments. Brain, 107, 829-854.Zipursky, R. B., Lim, K. O., Sullivan, E. V., Brown, B. W., &

Pfefferbaum, A. (1992). Widespread cerebral gray matter vol-ume deficits in schizophrenia. Archives of General Psychiatry,49, 195-205.

Received July 10, 1998

Revision received January 20,1999

Accepted May 13, 1999

Editor Transition for Neuropsychology

Dr. Laird S. Cermak, Editor for Neuropsychology since June 1995, died on November 4,

1999, of complications from a bone marrow transplant. Dr. Cermak prized his work on the

journal and continued to devote his time to it throughout his illness.

Associate Editor Patricia B. Sutker, PhD (Departments of Anesthesiology and Neuro-

psychiatry and Behavioral Science, Texas Tech University School of Medicine), has

graciously volunteered to serve as Acting Editor of Neuropsychology until a new editor

can be appointed. A search for a new editor for the term beginning in 2002 is currently

under way and is expected to be completed in March. Manuscripts that Dr. Cermak was

handling have been forwarded to Dr. Sutker for final action.

New submissions should continue to be sent in quadruplicate to the Boston office for

processing, as follows:

Neuropsychology

c/o Mary R. Fitzgerald, Editorial Assistant

Psychology Research Service (151 A)

Department of Veterans Affairs Medical Center

150 S. Huntington Avenue

Boston, MA 02130