Brain and Cognition 73 (2010) 146–151

Contents lists available at ScienceDirect

Brain and Cognition

journal homepage: www.elsevier .com/locate /b&c

Functional imaging of working memory and peripheral endothelial functionin middle-aged adults

Mitzi M. Gonzales a, Takashi Tarumi b, Hirofumi Tanaka b, Jun Sugawara b, Tali Swann-Sternberg a,Katayoon Goudarzi a, Andreana P. Haley a,c,*

a Department of Psychology, The University of Texas at Austin, Austin, TX, United Statesb Department of Kinesiology and Health Education, The University of Texas at Austin, Austin, TX, United Statesc University of Texas Imaging Research Center, Austin, TX, United States

a r t i c l e i n f o a b s t r a c t

Article history:Accepted 28 April 2010

Keywords:Working memoryMagnetic resonance imagingCardiovascular diseaseEndotheliumMiddle-agedCognition

0278-2626/$ - see front matter � 2010 Elsevier Inc. Adoi:10.1016/j.bandc.2010.04.007

* Corresponding author at: Department of PsycholoAustin, 108 E Dean Keaton, SEA 4.110, Austin, TX 7871471 6175.

E-mail address: haley@psy.utexas.edu (A.P. Haley)

The current study examined the relationship between a prognostic indicator of vascular health, flow-mediated dilation (FMD), and working memory-related brain activation in healthy middle-aged adults.Forty-two participants underwent functional magnetic resonance imaging while completing a 2-Backworking memory task. Brachial artery endothelial-dependent flow-mediated dilation (FMD) was assessedusing B-mode ultrasound. The relationship between FMD and task-related brain activation in a prioriregions of interest was modeled using hierarchical linear regression. Brachial FMD, was significantlyrelated to reduced working memory-related activation in the right superior parietal lobule (b = 0.338,p = 0.027), independent of age, sex, systolic blood pressure, and full scale IQ (F(5, 36) = 2.66, p = 0.038).These data provide preliminary support for the association between a preclinical marker of endothelialdysfunction and cerebral hemodynamic alterations in healthy middle-aged adults. Considering the mod-ifiable nature of endothelial function, additional investigations on the prognostic significance of FMD onfuture cognitive impairment are warranted.

� 2010 Elsevier Inc. All rights reserved.

1. Introduction et al., 2005; Haley, Forman et al., 2007a; Hoth et al., 2007; Moser

Dementia currently affects over 24 million people and itsprevalence is anticipated to double over the next 20 years (Qiu,De Ronchi, & Fratiglioni, 2007), making it one of the most signifi-cant and prevalent emerging public health issues. As successfultreatments for dementia are lacking, identification and manage-ment of modifiable risk factors are essential. Recent reportssuggest that vascular etiology contributes to over 50% of alldementia cases (Breteler, 2000). Atherosclerotic processes havebeen implicated in the pathogenesis of Alzheimer’s disease (Roheret al., 2003) and greater arterial stiffness has been recorded in indi-viduals with dementia than in age-matched healthy controls(Hanon et al., 2005). Given that vascular risk factors for dementiaare potentially modifiable, it is of utmost importance to find anefficacious method for identifying the early signs of vascular dete-rioration and their impact on cognition.

One promising approach has been to examine cardiac factors inrelation to cognitive performance and brain structure integrity innon-demented older adults with cardiovascular disease (Gunstad

ll rights reserved.

gy, The University of Texas at2, United States. Fax: +1 512

.

et al., 2008; Paul et al., 2005). Using this approach, clinicallysignificant levels of atherosclerosis and peripheral endothelialdysfunction have been associated with poorer attention-execu-tive-psychomotor performance and greater white matter damage(Forman et al., 2008; Haley, Sweet et al., 2007b; Hoth et al.,2007). While these studies have helped to elucidate the relation-ship between vascular health and brain function, their findingswere limited to elderly adults already suffering from cardiovascu-lar disease and exhibiting signs of vascular cognitive impairment.Considering that our best defense against dementia is prevention,it would be ideal to identify vascular markers that are associatedwith cerebral alterations in younger, healthier individuals.

The key challenge was to identify a vascular marker with prog-nostic significance for future vascular dysfunction. We chose to fo-cus on endothelial function since it is considered an overallindicator of vascular health (Vita & Keaney, 2002). Vascular endo-thelium plays an important role in the maintenance of vasculartone and prevention of atherosclerosis by inhibiting leukocyteand platelet adhesion to the arterial wall (Brunner et al.,2005). More importantly, endothelial dysfunction is evident be-fore the development of clinically-defined atherosclerosis (Vita& Keaney, 2002) and is considered a pathogenic factor in thedevelopment and progression of cardiovascular disease(Perticone et al., 2001).

M.M. Gonzales et al. / Brain and Cognition 73 (2010) 146–151 147

The aim of the present study was to determine if initial signs ofvascular dysfunction as indicated by poorer endothelial functionare associated with the cerebral hemodynamic response duringcognition among middle-aged adults free of cardiovascular disease.To accomplish our goal, we conducted functional magnetic reso-nance imaging (fMRI) and brachial artery flow mediated vasodila-tation (FMD) measurements on cognitively normal middle-agedadults. FMRI is unique in its sensitivity to detect early alterationsin brain function in individuals at risk for cognitive impairment(Bondi, Houston, Eyler, & Brown, 2005; Bookheimer et al., 2000;Haley et al., 2008) and FMD provides an overall indicator of endo-thelial function throughout the vasculature (Vita & Keaney, 2002).Based on previous research on elderly subjects with cardiovasculardisease (Haley, Sweet et al., 2007b; Irani et al., 2009), we hypoth-esized that decreased endothelial function would be associatedwith lower functional response to working memory and that loweractivations would be related to lower task performance.

2. Methods

2.1. Subjects

Right-handed adults between the ages of 40 and 60 were re-cruited through flyers and newspaper advertisements posted inAustin, Texas. Handedness was assessed using the EdinburghHandedness Questionnaire (Oldfield, 1971). Individuals with a his-tory of coronary artery disease, angina pectoris, myocardial infarc-tions, heart failure, and cardiac surgery were excluded. Additionalexclusion criteria included history of neurological disease (e.g.,stroke, Parkinson’s disease, clinically significant traumatic brain in-jury), major psychiatric illness (e.g. schizophrenia, bipolar disor-der), substance abuse (i.e., diagnosed abuse and/or previoushospitalization for substance abuse), metabolic disorder (i.e., dia-betes, thyroid disorder), smoking (within the last 2 years) or MRIcontraindications. The study was in accordance with the SecondDeclaration of Helsinki and was approved by the local institutionalreview committee. Forty-two adults participated in the study, pro-viding written informed consent before enrollment. Participantscharacteristics are presented in Table 1.

2.2. Neuropsychological assessment

All participants completed standard clinical neuropsychologicalinstruments with established reliability and validity (Lezak,Howieson, Loring, Hannay, & Fischer, 2004). These measures as-sessed global cognitive functioning (Mini Mental Status Exam,MMSE (Folstein, Folstein, & McHugh, 1975), full scale IQ (WechslerAbbreviated Scale of Intelligence – Two Subtest, WASI (Wechsler,

Table 1Participant characteristics (n = 42).

Participant characteristics Mean ± SD

Age, y 49.0 ± 6.3Sex (male/female) 16/26Education, y 15.4 ± 2.7

Race, n, (%)Non-Hispanic white 21 (50.0%)Hispanic 14 (33.3%)Asian American 1 (2.4%)African American 5 (11.9%)Native American 1 (2.4%)

Body mass index, kg/m2 28.5 ± 5.3Systolic blood pressure, mm Hg 124 ± 15Diastolic blood pressure, mm Hg 75 ± 8Flow-mediated dilation,% 4.2 ± 4.8

1999), language (WASI Vocabulary Subtest; Category Fluency forAnimals (Morris et al., 1989)) memory (California Verbal andLearning Test II, CVLT-II (Delis, Kramer, Kaplan, & Ober, 1987);Rey Complex Figure Test, RCF (Meyers & Meyers, 1995)), atten-tion-executive functioning (Controlled Oral Word Association Test,COWAT (Ruff, Light, Parker, & Levin, 1996); Trail Making Test A & B(Reitan, 1958), Wechsler Adult Intelligence Scale-III, WAIS-III, DigitSpan Subtest (Wechsler, 1997)), psychomotor speed (Grooved Peg-board (Ruff & Parker, 1993)) visual-spatial ability (RCF copy; WASIMatrix Reasoning Subtest) and emotional functioning (BeckDepression Inventory-II, BDI-II (Beck, Steer, & Brown, 1996); StateTrait Anxiety Inventory, STAI (Spielberger & Gorsuch, 1970)). Testswere administered and scored by a trained research assistant usingstandard administration and scoring criteria.

2.3. Cardiovascular assessment

Participants abstained from caffeine and fasted for at least fourhours prior to the assessment. After 15 min of rest in a supine po-sition, brachial blood pressure was measured using a semi-auto-mated device (Dinamap XL, Johnson & Johnson Medical Inc.,Tampa, FL). Three separate recordings were made while partici-pants sat upright in accordance with American Heart AssociationGuidelines (Perloff et al., 1993).

Endothelial function was assessed with brachial artery flow-mediated dilation. A B-mode Doppler ultrasound machine (iE 33Ultrasound System, Philips, Bothell, WA) with a customized trans-ducer holding device was used to measure brachial artery diame-ters and blood flow velocity. Brachial artery images wereobtained in a longitudinal orientation located 5–10 cm proximalto the antecubital fossa. Baseline blood flow and diameter mea-surements were made after participants rested for at least15 min in the supine position. After the acquisition of baselinemeasurements, a blood pressure cuff placed on the ipsilateral fore-arm distal to the elbow was inflated to 100 mm Hg above baselinesystolic blood pressure for 5 min using a rapid cuff inflator (E20,Hokanson, Bellevue, WA). Ultrasound-derived blood velocity anddiameter data were saved as DICOM format and transferred to acomputer using a digital image viewing software (Access Point2004, Freeland Systems; Westminster, CO) for later analyses. Allultrasound brachial images were subsequently analyzed by thesame investigator using image analysis software (VascularResearch Tool Brachial Analyzer, Medical Imaging Applications,Coralville, IA).

FMD was expressed as the percent change in brachial arterydiameters recorded during the pre and post occlusion phases andwas calculated using the equation: (maximum diameter – baselinediameter)/baseline diameter � 100. The average of 10 end-dia-stolic brachial artery diameters before blood flow occlusion wasused for baseline diameters and the average of three peak end-dia-stolic diameters during the reperfusion phase was used for maxi-mum brachial artery diameter.

2.4. Working memory task paradigm

Working memory was assessed using a verbal n-Back task(Braver et al., 1997; Cohen et al., 1997; Walter et al., 2003), consist-ing of alternating blocks of 0-Back, 2-Back, and rest conditions.During each 0- and 2-Back block, a series of twelve individual con-sonants were visually presented in random order for 500 ms eachwith a 2500 ms inter-stimulus interval. Participants responded totarget letters (33% in each block) using a two-button MR-compat-ible response box. In the 0-Back condition, the target was a pre-specified letter (H) and in the 2-Back condition, the target wasany letter that was identical to the one presented two stimuli ear-lier. Task performance was assessed by measuring mean accuracy

148 M.M. Gonzales et al. / Brain and Cognition 73 (2010) 146–151

rates and reaction time for all correct trials. During each rest block,a fixation cross appeared in the middle of the screen for 30 s. Thetask was programmed and presented using E-Prime software (Psy-chology Software Tools, Inc., Pittsburgh, PA). During neuroimaging,two consecutive 6 min runs consisting of three blocks of alternat-ing 0-Back, 2-Back, and rest conditions were presented. Beforeneuroimaging, participants were given an opportunity to practicethe task on laptop computer to ensure adequate performance.

Fig. 1. 1 = Left middle frontal gyrus; 2 = left medial frontal gyrus; 3 = right superiorparietal lobule; 4 = left inferior parietal lobule; 5 = left middle frontal gyrus;6 = right superior frontal gyrus; 7 = right middle frontal gyrus; 8 = right inferiorfrontal gyrus.

Table 3Neuropsychological test results.

Test measures Mean ± SD

Global cognitionMini Mental Status Exam (MMSE) 28.1 ± 1.5

Full scale IQWeschler Abbreviated Scale of Intelligence (WASI) 112.3 ± 16.1

LanguageWASI Vocabulary (T-score) 55.0 ± 9.2Category Fluency for Animals 23.7 ± 5.6

2.5. Neuroimaging

MRI data for each participant were acquired in a single sessionon a 3T GE Signa Excite MRI scanner equipped with a standardhead coil. T1-weighted anatomical scans of the entire brain inthe saggital plane were collected using a high-resolution SpoiledGradient Echo (SPGR) sequence (256 � 256 matrix, FOV = 24 cm2,1 mm slice thickness, 0 gap). Functional imaging was preformedwhile participants completed the 2-Back task. The task was back-projected from a laptop onto a screen positioned at the partici-pant’s head, and viewed through a double-mirror attached to thehead coil. Functional imaging was performed using a whole brainecho-planer imaging (EPI) sequence (TR = 3000 ms, TE = 30 ms,FOV = 24 cm2, 64 � 64 matrix, 42 axial slices, 3 mm slice thickness,0.3 mm gap).

All EPI images were processed using Analysis of NeuroImages(AFNI) software (Cox, 1996). Each time series was spatially regis-tered to the sixth volume of the session to reduce the effects ofhead movement. Data pre-processing also included adjustmentfor differences in adjacent slice timing due to interleaved sliceacquisition, temporal smoothing, spatial filtering, and transforma-tion to standard stereotaxic space. Task-related brain activationwas determined using voxel-wise multiple regression analyseswith the following parameters: a 0-Back/2-Back reference wave-form convolved with a gamma function and covariates accountingfor instruction screens and head movement.

In order to elaborate upon prior research examining cerebralhemodynamics and cardiovascular function, a priori regions ofinterest (ROIs) were created from published coordinates that wereempirically derived from a verbal 2-Back task in individuals withcardiovascular disease (Haley, Sweet et al., 2007b) (Table 2,Fig. 1). All sterotactic coordinates refer to Talairach space(Talairach & Tournoux, 1988). The ROIs were applied to individualdata to determine mean task-related activation intensity in eachregion.

MemoryCalifornia Verbal Learning Testing II (CVLT-II)Immediate Recall 10.3 ± 3.0Delayed Recall 11.0 ± 3.1Recognition (Yes/No) 3.1 ± 0.7Rey Complex Figure Test, RCFImmediate Recall 15.2 ± 5.6Delayed Recall 14.6 ± 5.5Recognition 19.3 ± 2.4

Attention-executive functionControlled Oral Word Association Test (COWAT) 37.6 ± 10.8Trail Making Test A, sec 30.4 ± 9.3

2.6. Statistical analyses

Variable distributions were examined using the Shapiro–Wilktest of normality. A log transformation was used to obtain a normaldistribution of the 2-Back-related activation in the right superiorfrontal gyrus (Shapiro–Wilk = 0.943, p = 0.035). All other outcomedistributions fulfilled the assumption of normality.

The association between FMD and mean 2-Back-relatedactivation intensity was analyzed within each a priori ROI using

Table 2A priori regions of interest.

Region X Y Z

Left middle frontal gyrus 33 4 56Left medial frontal gyrus 5 19 44Right superior parietal lobule 37 �63 53Left inferior parietal lobule 49 �52 44Left middle frontal gyrus 44 45 13Right superior frontal gyrus 33 48 15Right middle frontal gyrus 32 5 55Right inferior frontal gyrus 47 14 3

hierarchical linear regression. In the first block, the model was ad-justed for age, sex, systolic blood pressure, and full scale IQ. FMDwas entered into the second block in order to estimate its indepen-dent contribution to the variance in 2-Back-related activation. Thedecision to adjust the model for full scale IQ was based on researchdemonstrating a linear relationship between global intellectualability and domain specific cognitive test performance. Adults withaverage intellectual ability tend to perform worse on neuropsycho-logical tests than adults with above average and superior intellec-tual abilities, even in the absence of neurological disease (Bell &Roper, 1998; Horton, 1999; Tremont et al., 1998). Therefore, inany search for markers of cognitive vulnerability, it is importantto avoid falsely identifying adults with lower intellectual abilityas showing signs of neuropsychological impairment.

In exploratory follow-up analyses, mean 2-Back accuracy andreaction time were correlated with 2-Back activation in the singleROI that was significant in the regression model. Data were

Trail Making Test B, sec 74.5 ± 27.5Wechsler Adult Intelligence Scale-III (WAIS-III) Digit Span 16.5 ± 3.9

Visual-spatialRCF copy 30.6 ± 4.2WASI Matrix Reasoning (T-score) 57.4 ± 7.0

Psychomotor speedGrooved Peg Board, Dominant Hand, sec 76.7 ± 18.4

DepressionBeck Depression Inventory-II, BDI-II 5.9 ± 5.3

AnxietySTAI-Trait (scaled) 49.8 ± 12.6

Flow-Mediated Dilation, % Change in Vessel Diameter18.012.06.0.0− 6.0

Mea

n Ta

sk-R

elat

ed A

ctiv

atio

n In

tens

ity

10.0

7.5

5.0

2.5

.0

−12.0

Fig. 2. Scatterplot of flow-mediated dilation and mean task-related activationintensity in the right superior parietal lobule.

M.M. Gonzales et al. / Brain and Cognition 73 (2010) 146–151 149

analyzed using SPSS 16.0 computer software (SPSS Inc., Chicago,IL). A two-tailed alpha level of 0.05 was used as the criterion forsignificance for all analyses.

3. Results

Descriptive statistics on the results from the neuropsychologi-cal measures are presented in Table 3. Mean (±SD) accuracy onthe verbal 2-Back task was 76.8 ± 12.8% correct responses, andmean reaction time was 1185 ± 287 ms. All participants movedless than 1.5 mm per imaging run.

The fully adjusted regression models successfully predicted themean 2-Back-related activation intensity in the right superior pari-etal lobule (F(5, 36) = 2.66, p = 0.038) and in the left inferior parie-tal lobule (F(5, 36) = 3.46, p = 0.012). Male gender was significantlyassociated lower 2-Back-related intensity in right superior parietallobule (b = �0.346, p = 0.024) and in the left inferior parietal lobule(b = �0.431, p = 0.004). The independent effects of age, systolicblood pressure, and full scale IQ did not account for any uniquevariance in 2-Back-related activation intensity.

Decreased endothelial function, as measured by brachial FMD,was significantly associated with attenuated 2-Back-related activa-tion in the right superior parietal lobule (b = 0.338, p = 0.027), inde-pendent of age, sex, systolic blood pressure, and full scale IQ(Fig. 2). 2-Back-related activation in the same region was associ-ated with slower reaction time (r = �0.347, p = 0.026), but notaccuracy (r = 0.183, p = 0.252). The independent effects of FMDdid not reach statistical significance in the left inferior parietal lob-ule (b = 0.273, p = 0.061).

4. Discussion

The current study investigated the relationship between endo-thelial function and brain activation during a verbal working mem-ory task in healthy, middle-aged adults. We found that a sensitivemeasure of vascular health, FMD, was associated with reducedbrain activation in the superior parietal lobule. The relationship be-tween FMD and brain activation in this region was independent ofthe effects of age, sex, systolic blood pressure, and IQ. Since FMDcontributes to the pathogenesis of cardiovascular disease (Perti-cone et al., 2001), alterations in this measure appear before clinicaldiagnosis (Vita & Keaney, 2002). FMD’s prognostic significance,combined with the sensitivity of fMRI, may have enabled us to de-tect very early changes in the hemodynamic response to cognitionin healthy, cognitively intact middle-aged adults.

Superior parietal activation is a common finding across workingmemory studies (Braver et al., 1997; Cohen et al., 1997; Walteret al., 2003). This region is considered critical for sensorimotorintegration (Wolpert, Goodbody, & Husain, 1998) and respondsto higher cognitive demands by tonically increasing activation inorder to maintain higher levels of response readiness (Braveret al., 1997). Our findings are consistent with this interpretationas lower superior parietal lobule activation in our sample was asso-ciated with slower 2-Back reaction times. These results are of par-ticular interest as psychomotor slowing is among the first signs ofvascular cognitive decline (Looi & Sachdev, 2000). Our findingsindicate that the relationship between FMD and brain activationmay be sensitive enough to detect early vascular-related changesin brain function before cognition and cardiovascular function be-come substantially impaired. Longitudinal studies will be neces-sary to determine the predictive value of these findings forestimating future cognitive trajectories.

Contemplating the physiological mechanisms supporting thedetected relationship between peripheral FMD and cerebrovascu-lar response to cognition is also very interesting. Nitric oxide,which regulates endothelial dilation in FMD, is known to play animportant role in the regulation of cerebral circulation (White,Vallance, & Markus, 2000). In humans, inhibition of nitric oxide re-lease has been associated with reduced basal cerebral blood flow(White, Deane, Vallance, & Markus, 1998). These findings may bearimportance on cognition since cognitive engagement enhances thedemand for oxygen, thus increasing blood flow delivery to acti-vated brain regions (Logothetis & Wandell, 2004). With severe dys-function in the regulation of cerebral blood flow, activated brainareas may not receive adequate blood supply, inducing structuraldamage and impairments in cognition (Hoth et al., 2007;Rockwood, 2002). Consistent with this idea, patients with vasculardementia display attenuated cerebral blood flow at rest, suggestinghypoperfusion may play a role in the etiology of the disease (Yanget al., 2002). Additionally, brachial FMD has been associated withthe severity of white matter hyperintensities in the brain in elderlycardiovascular disease patients (Hoth et al., 2007), indicating arelationship between peripheral and cerebral circulation. Futurestudies will be necessary to determine if brachial FMD is indeed di-rectly related to autoregulation in the cerebral vasculature. In thiscontext, preliminary evidence is encouraging. Prior research hasfound that degree and timing of blood flow autoregulation in themiddle cerebral artery is closely associated with the flow throughperipheral arteries such as the carotid artery (White et al., 2000).

Alternatively, reduction in task-related brain activation at nor-mal behavioral performance levels could be an indicator of neuro-vascular decoupling. The blood–oxygen-level-dependent (BOLD)signal in fMRI is generated by changes in the ratio between oxy-genated and deoxygenated hemoglobin in the surrounding micro-vasculature (Logothetis & Wandell, 2004). Engagement in cognitivetasks alters the BOLD response by influencing cerebral blood flow,cerebral blood volume, and cerebral blood oxygen consumption(D’Esposito, Deouell, & Gazzaley, 2003). In healthy individuals,the change in the BOLD response closely parallels neural activity.However, in situations where the cerebrovasculature is altered,the BOLD response may not be a reliable indicator of neural activa-tion (D’Esposito, Deouell, & Gazzaley, 2003). Neurovascular decou-pling is unlikely to be the cause for our findings, because the studypopulation consisted of healthy, middle-aged adults. Evidence fordecoupling has mostly been documented in cases of severe vascu-lar pathology such as ischemic stroke and intracranial stenosis(Girouard & Iadecola, 2006). Nonetheless, assessment of neurovas-cular decoupling should be addressed in future studies.

In conclusion, the current study demonstrated that poorerperipheral endothelial function was related to reduced task-relatedsuperior parietal activation independent of the effects of age, sex,

150 M.M. Gonzales et al. / Brain and Cognition 73 (2010) 146–151

systolic blood pressure, and full scale IQ. Prior research has demon-strated a relationship between vascular dysfunction and brain acti-vation in elderly adults with cardiovascular disease (Haley, Sweetet al., 2007b; Irani et al., 2009). The current findings extend the lit-erature, by identifying a sensitive vascular factor, FMD, which wasrelated to cerebrovascular response to cognition in a healthy, mid-dle-aged sample. Due to its early role in the pathogenesis and pro-gression of vascular disease (Perticone et al., 2001), FMD may beuniquely suited for identifying cognitive vulnerability at youngerages. The ability to identify individuals at risk for vascular cognitiveimpairment among healthy, middle-aged adults would be of para-mount importance for preventive efforts, especially given the mod-ifiable nature of endothelial function (Vita & Keaney, 2002).Limitations of the current study include a relatively small samplesize and a cross-sectional design. Future longitudinal research isnecessary to determine the predictive validity of these findings forestimating individual cognitive trajectories. Studies of the relation-ship between FMD and other indicators of cerebral function such aswhite matter integrity, cerebral autoregulation, and cerebral neuro-chemistry in midlife would also help assess the value of FMD as aprognostic indicator of cognitive vulnerability. Finally, the efficacyof midlife interventions aimed at improving vascular health to delayor prevent future cognitive decline should be explored.

Acknowledgement

This work was funded by NINR center Grant P30 NR005051(APH), AHA Grant 09BGIA2060722 (APH) and the University ofTexas at Austin (APH). The authors thank the UT Imaging Centerstaff for their help with the participants.

References

Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Manual for the beck depressioninventory-II. San Antonio, TX: Psychological Corporation.

Bell, B. D., & Roper, B. L. (1998). Myths of neuropsychology: Another view. TheClinical Neuropsychologist, 12, 237–244.

Bondi, M. W., Houston, W. S., Eyler, L. T., & Brown, G. G. (2005). FMRI evidence ofcompensatory mechanisms in older adults at genetic risk for Alzheimer disease.Neurology, 64, 501–508.

Bookheimer, S. Y., Strojwas, M. H., Cohen, M. S., Saunders, A. M., Pericak-Vance, M.A., Mazziotta, J. C., et al. (2000). Patterns of brain activation in people at risk forAlzheimer’s disease. The New England Journal of Medicine, 343, 450–456.

Braver, T. S., Cohen, J. D., Nystrom, L. E., Jonides, J., Smith, E. E., Noll, D. C., et al.(1997). A parametric study of prefrontal cortex involvement in human workingmemory. Neuroimage, 5, 49–62.

Breteler, M. (2000). Vascular risk factors for Alzheimer’s disease: An epidemiologicperspective. Neurobiology of Aging, 21, 153–160.

Brunner, H., Cockcroft, J. R., Deanfield, J., Donald, A., Ferrannini, E., Halcox, J., et al.(2005). Endothelial function and dysfunction. Part II: Association withcardiovascular risk factors and diseases. A statement by the working group onendothelins and endothelial factors of the European Society of Hypertension.Journal of Hypertension, 23, 233–246.

Cohen, J. D., Perlstein, W. M., Braver, T. S., Nystrom, L. E., Noll, D. C., Jonides, J., et al.(1997). Temporal dynamics of brain activation during a working memory task.Nature, 386, 604–608.

Cox, R. W. (1996). AFNI: Software for analysis and visualization of functionalmagnetic resonance neuroimages. Computers and Biomedical Research, 29,162–173.

Delis, D. C., Kramer, J. H., Kaplan, E., & Ober, B. A. (1987). California verbal learningtest: Adult version. San Antonio, TX: The Psychological Corporation.

D’Esposito, M., Deouell, L. Y., & Gazzaley, A. (2003). Alterations in the BOLD fMRIsignal with ageing and disease: A challenge for neuroimaging. Nature ReviewsNeuroscience, 4, 863–872.

Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). Mini-mental state: A practicalmethod for grading the cognitive state of patients for the clinician. Journal ofPsychiatric Research, 12, 189–198.

Forman, D. E., Cohen, R. A., Hoth, K. F., Haley, A. P., Poppas, A., Moser, D. J., et al.(2008). Vascular health and cognitive function in older adults withcardiovascular disease. Artery Research, 2, 35–43.

Girouard, H., & Iadecola, C. (2006). Neurovascular coupling in the normal brain andin hypertension, stroke, and Alzheimer disease. Journal of Applied Physiology,100, 328–335.

Gunstad, J., Cohen, R. A., Tate, D. F., Paul, R. H., Poppas, A., Hoth, K., et al. (2005).Blood pressure variability and white matter hyperintensities in older adultswith cardiovascular disease. Blood Pressure, 14, 353–358.

Haley, A. P., Forman, D. E., Poppas, A., Hoth, K. F., Gunstad, J., Jefferson, A. L., et al.(2007). Carotid artery intima-media thickness and cognition in cardiovasculardisease. International Journal of Cardiology, 121, 148–154.

Haley, A. P., Gunstad, J., Cohen, R. A., Jerskey, B. A., Mulligan, R. C., & Sweet, L. H.(2008). Neural correlates of visuospatial working memory in healthy youngadults at risk for hypertension. Brain Imaging and Behavior, 2, 192–199.

Haley, A. P., Sweet, L. H., Gunstad, J., Forman, D. E., Poppas, A., Paul, R. H., et al.(2007). Verbal working memory and atherosclerosis in patients withcardiovascular disease: An fMRI study. Journal of Neuroimaging, 17, 227–233.

Hanon, O., Haulon, S., Lenoir, H., Seux, M. L., Rigaud, A. S., Safar, M., et al. (2005).Relationship between arterial stiffness and cognitive function in elderlysubjects with complaints of memory loss. Stroke, 36, 2193–2197.

Horton, A. M. (1999). Above-average intelligence and neuropsychological test scoreperformance. International Journal of Neuroscience, 99, 221–231.

Hoth, K. F., Tate, D. F., Poppas, A., Forman, D. E., Gunstad, J., Moser, D. J., et al. (2007).Endothelial function and white matter hyperintensities in older adults withcardiovascular disease. Stroke, 38, 308–312.

Irani, F., Sweet, L. H., Haley, A. P., Gunstad, J. J., Jerskey, B. A., Mulligan, R. C., et al.(2009). A fMRI study of verbal working memory, cardiac output, and ejectionfraction in elderly patients with cardiovascular disease. Brain Imaging andBehavior, 3, 350–357.

Lezak, M., Howieson, D., Loring, D., Hannay, H., & Fischer, J. (2004).Neuropsychological assessment (4th ed.). New York, NY: Oxford University Press.

Logothetis, N. K., & Wandell, B. A. (2004). Interpreting the BOLD signal. AnnualReviews of Physiology, 66, 735–769.

Looi, J. C., & Sachdev, P. S. (2000). Vascular dementia as a frontal subcortical systemdysfunction. Psychological Medicine, 30, 997–1003.

Meyers, J. E., & Meyers, K. R. (1995). Rey complex figure test and recognition trial:Professional manual. Odessa, Fl: Psychological Assessment Resources.

Morris, J. C., Heyman, A., Mohs, R. C., Hughes, J. P., Van Belle, G., Fillenbaum, G., et al.(1989). The consortium to establish a registry for Alzheimer’s disease (CERAD).I. Clinical and neuropsychological assessment of Alzheimer’s diseases.Neurology, 39, 1159–1165.

Moser, D. J., Miller, I. N., Hoth, K. F., Correia, M., Arndt, S., & Haynes, W. G. (2008).Vascular smooth muscle function is associated with initiation and processingspeed in patients with atherosclerotic vascular disease. Journal of theInternational Neuropsychological Society, 14, 535–541.

Oldfield, R. C. (1971). The assessment and analysis of handedness: The Edinburghinventory. Neuropsychologia, 9, 97–113.

Paul, R. H., Gunstad, J., Poppas, A., Tate, D. F., Foreman, D., Brickman, A. M., et al.(2005). Neuroimaging and cardiac correlates of cognitive function amongpatients with cardiac disease. Cerebrovascular Disease, 20, 129–133.

Perloff, D., Grim, C., Flack, J., Frohlich, E. D., Hill, M., McDonald, M., et al. (1993).Human blood pressure determination by sphygmomanometry. Circulation, 88,2460–2470.

Perticone, F., Ceravolo, R., Pujia, A., Ventura, G., Iacopino, S., Scozzafava, A., et al.(2001). Prognostic significance of endothelial dysfunction in hypertensivepatients. Circulation, 104, 191–196.

Qiu, C., De Ronchi, D., & Fratiglioni, L. (2007). The epidemiology of the dementias:An update. Current Opinion in Psychiatry, 20, 380–385.

Reitan, R. M. (1958). Validity of the trail making test as an indicator of organic braindamage. Perceptual and Motor Skills, 8, 271–276.

Rockwood, K. (2002). Vascular cognitive impairment and vascular dementia. Journalof the Neurological Sciences, 203, 23–27.

Roher, A. E., Esh, C., Kokjohn, T. A., Kalback, W., Luehrs, D. C., Seward, J. D., et al.(2003). Circle of Willis atherosclerosis is a risk factor for sporadic Alzheimer’sdisease. Arteriosclerosis, Thrombosis, and Vascular Biology, 23, 2055–2062.

Ruff, R. M., Light, R. H., Parker, S. B., & Levin, H. S. (1996). Benton controlled oralword association test: Reliability and updated norms. Archives of ClinicalNeuropsychology, 11, 329–338.

Ruff, R. M., & Parker, S. B. (1993). Gender-and age-specific changes in motor speedand eye-hand coordination in adults: Normative values for the finger tappingand grooved pegboard tests. Perceptual and Motor Skills, 76, 1219–1230.

Spielberger, C. D., & Gorsuch, R. L. (1970). STAI manual for the state-trait anxietyinventory (‘‘self-evaluation questionnaire”). Palo Alto, CA: ConsultingPsychologists Press.

Talairach, J., & Tournoux, P. (1988). Co-planar stereotaxic atlas of the human brain.Stuttgart: Thieme.

Tremont, G., Hoffman, R. G., Scott, J. G., & Adams, R. L. (1998). Effect of intellectuallevel on neuropsychological test performance: A response to Dodrill (1997). TheClinical Neuropsychologist, 12, 560–567.

Vita, J. A., & Keaney, J. F. Jr., (2002). Endothelial function: A barometer forcardiovascular risk? Circulation, 106, 640–642.

Walter, H., Bretschneider, V., Grön, G., Zurowski, B., Wunderlich, A. P., Tomczak, R.,et al. (2003). Special issue evidence for quantitative domain dominance forverbal and spatial working memory in the frontal and parietal cortex. Cortex, 39,897–911.

Wechsler, D. (1997). Administration and scoring manual for the Wechsler adultintelligence scale (3rd ed.). San Antonio, TX: The Psychological Corporation.

Wechsler, D. (1999). Wechsler abbreviated scale of intelligence. San Antonio, TX: ThePsychological Corporation.

White, R. P., Deane, C., Vallance, P., & Markus, H. S. (1998). Nitric oxide synthaseinhibition in humans reduces cerebral blood flow but not the hyperemicresponse to hypercapnia. Stroke, 29, 467–472.

M.M. Gonzales et al. / Brain and Cognition 73 (2010) 146–151 151

White, R. P., Vallance, P., & Markus, H. S. (2000). Effect of inhibition of nitric oxidesynthase on dynamic cerebral autoregulation in humans. Clinical Science, 99,555–560.

Wolpert, D. M., Goodbody, S. J., & Husain, M. (1998). Maintaining internalrepresentations: The role of the human superior parietal lobe. NatureNeuroscience, 1, 529–533.

Yang, D. W., Kim, B. S., Park, J. K., Kim, S. Y., Kim, E. N., & Sohn, H. S. (2002). Analysisof cerebral blood flow of subcortical vascular dementia with single photonemission computed tomography: Adaptation of statistical parametric mapping.Journal of the Neurological Sciences, 203, 199–205.