NeuroImage 13, 68–75 (2001)doi:10.1006/nimg.2000.0662, available online at http://www.idealibrary.com on
The Effect of Medial Frontal and Posterior Parietal DemyelinatingLesions on Stroop Interference
Jesus Pujol,* Pere Vendrell,† Joan Deus,* Carme Junque,† Joan Bello,‡Josep L. Martı-Vilalta,§ and Antoni Capdevila*
*Magnetic Resonance Center of Pedralbes, Barcelona; †Department of Psychiatry and Clinical Psychobiology, University of Barcelona;‡Neurology Unit, Creu Roja Hospital, Hospitalet de Llobregat, Barcelona; and §Department of Neurology,
Santa Creu i Sant Pau Hospital, Autonomous University of Barcelona, Spain
Received April 10, 2000; published online November 7, 2000
The study of the cerebral structures involved in theStroop task has contributed to a better understandingof the functional anatomy of the major attentional sys-
tems in the human brain (Pardo et al., 1990; Bench et
al., 1993; Peterson et al., 1999). Indeed, basic aspects ofattention may be explored in situations of conflict be-tween selective attention and distracting concurrentstimulation. In the Stroop test, the time taken to namethe color of a presented stimulus increases if that stim-ulus is a printed word that signifies a different color,such as the word blue written in red ink (MacLeod,1991).
During the last decade, functional imaging has de-lineated the anatomical substratum of the Stroop phe-nomenon. These studies have confirmed that the mainattention-related areas of the brain are active duringthis attentional conflict. Functional results have beennotably coincident and, although the specific anatomi-cal pattern observed varied as a function of differencesin task constraints, two definite brain regions haverepetitively shown increased activity during Strooptesting: the medial frontal cortex, particularly the an-terior cingulate, and the posterior parietal cortex(Bench et al., 1993; George et al., 1994; Carter et al.,1995; Bush et al., 1998; Brown et al., 1999; Peterson etal., 1999).
Interestingly, these two regions appear to processnotably different aspects of attention. The anterior cin-gulate seems to be related to executive operations fo-cusing attention on response selection (Corbetta et al.,1991; Carter et al., 1995; Passingham, 1996). By con-trast, the role of the parietal lobe is better described interms of stimulus-driven orientation of attention withautomatic enhancement of sensory processing (Posneret al., 1984; Pardo et al., 1991; Corbetta, 1998). In such
context, it is possible that anterior cingulate favorsaster responses in the Stroop situation, whereas pari-tal cortex mediates automatic engagement of the in-erfering reading stimulus that slows color naming.evertheless, functional imaging activity changes doot directly reflect facilitation and interference effectsBench et al., 1993). Studies analyzing correlations be-
tween regional activations and Stroop reaction time
showed that cingulate activity was inversely related to
4
69STROOP AND MULTIPLE SCLEROSIS
subjects’ efficiency in the task (George et al., 1994;Taylor et al., 1994; Bush et al., 1998).
We conducted a lesion study to directly investigatethe influence of these two brain regions on task perfor-mance. Specifically, we studied patients with multiplesclerosis (MS), a neurological disorder that clinicallypresents varying degrees of Stroop interference (Rao etal., 1991) and pathologically shows demyelination,which produces measurable slowing of neural trans-mission and is exquisitely detectable with currentstructural imaging. In MRI studies, we measured ar-eas of demyelination on medial frontal and posteriorparietal white matter and determined their specificcontribution to Stroop response delay.
MATERIALS AND METHODS
Subjects
Multiple sclerosis outpatients were consecutively se-lected for this study according to the following criteria:(1) positive diagnosis of “clinically definite” MS (Poseret al., 1983), (2) evidence of cerebral demyelinatinglesions in routine MRI, (3) report of a minimal 3-monthperiod prior to recruitment without clinical relapse (orrelevant increase of the progression rate), (4) no steroidtherapy, (5) absence of concomitant medical processesthat could mask MS symptoms, (6) no DSM-III-R cri-teria of severe dementia, and (7) no severe dysarthriaor visual disturbances that could preclude Stroop as-sessment. Written informed consent was obtained ineach case and the study was approved by the Institu-tional Research Committee.
Forty-five patients (30 women and 15 men) fulfilledthe criteria and underwent a comprehensive clinical,neuropsychological, and imaging evaluation to investi-gate the Stroop phenomenon and other behavioral as-pects (Pujol et al., 1997). The average age of patientswas 39.7 6 11.9 years and the time since originaldiagnosis was 7.4 6 6.9 years. Fourteen patients hadthe progressive (primary or secondary) form and 31had the relapsing-remitting form of the disease. TheKurtzke Expanded Disability Status Scale (Kurtzke,1983) rated a mean of 3.3 6 1.8 and the NeurologicRating Scale (Sipe et al., 1984) a mean of 81.1 6 16.3.The mean level of education attained was 10.0 6 3.7years, the vocabulary subtest of the Wechsler AdultIntelligence Scale (Wechsler, 1955) scored a mean of12.3 6 2.5 and similarities a mean of 12.4 6 2.1.
No patient showed any major impediment in namingStroop colors. Group mean errors registered duringsimple color naming in this test was 0.4 6 1.2. Simi-larly, no patient had any major impediment in readingStroop words. Each patient was able to read a complextext successfully (Pujol et al., 1997). All patients scored
points or less (maximum test score 5 6) in the visual
assessment part of the Kurtzke Expanded Disability
Status Scale and none of them showed a visual-fieldhemianopic deficit.
A series of 30 volunteers (19 women and 11 men)with similar mean age (39.4 6 14.9 years) and educa-tional level (10.7 6 3.7 years) were recruited fromrelatives of patients to obtain reference values ofStroop performance. None of them had history of neu-rological or psychiatric disease, alcoholism, or sub-stance abuse.
Stroop Test
A verbal version of the Stroop test was used in whicheight sets of six stimuli were presented on the center ofa computer monitor. In odd sets, patients were re-quired to name the color (blue, green, red, yellow) inwhich a string of dots was presented and, in even sets,the color of letters signifying an incongruent colorword. Subjects were required to name the presentedcolors as quickly as possible while trying to avoid er-rors. A voice-activated relay was used to record inmilliseconds the vocal reaction time of correct re-sponses using the procedures adopted by Vendrell et al.(1995). To minimize the effect of possible extreme val-ues, we used the median of reaction time, instead of themean, as a measurement of test performance. The dif-ference of median reaction time of interfered minusnoninterfered naming conditions was taken as the“Stroop interference.” Response errors were also re-corded.
MRI Procedure
A superconductive 1.5-Tesla magnet (Signa system,GE Medical Systems, Milwaukee, WI) was used to ex-amine patients within the same week of neuropsycho-logical assessment with a specific MRI protocol (Pujolet al., 1997), including sagittal, axial (spin-echo), andcoronal (inversion-recovery) sequences. Demyelinatinglesion measurements were performed using the coro-nal projection, as it provides better anatomical depic-tion of the white matter regions analyzed in this study.Moreover, acquisition times of this inversion-recoverysequence were adjusted to produce T1-weighted im-ages (see Fig. 1), which may be more specific for myelinloss evaluation. TR/TE and TI of 1,500/20, and 650 mswere used to obtain 15 parallel, 5-mm-thick slices, withan interslice gap of 2.5 mm within a matrix size of256 3 256 pixels and a 22-cm field of view.
Region delimitation. A reference slice was estab-lished on the sagittal view in each examination follow-ing the commissural–obex reference line. The medialfrontal region was defined in this reference slice and ineach of the eight contiguous slices anterior to it. Thelimits of this region were the corpus callosum (inferi-orly) and a vertical line passing through the superiorfrontal sulcus (laterally), as is shown in Fig. 1. This
medial frontal region was further sectored in five divi-
1map
csetgdml
s
70 PUJOL ET AL.
sions (Fig. 2, left), corresponding to a subgenual divi-sion (slices 1, 2, and 3, below the rostrum of corpuscallosum), a pregenual division (slices 1, 2, and 3,above the rostrum of corpus callosum), an anteriorsupracallosal division (slices 4 and 5), a midsupracal-losal division (slices 6 and 7), and a posterior supracal-losal division (slices 8 and 9).
The parietal region was defined in each of the fourcontiguous slices posterior to the reference slice (Fig.1). An inferomedial limit of this region was establishedby means of an oblique line traced from the cingulatesulcus to the sylvian fissure (or a prolongation of itshorizontal segment in the most posterior slices). There-fore, the parietal region considered in this study rep-resents the posterior part of the dorsal and lateralparietal lobe and includes a similar extent of whitematter above and below the intraparietal sulcus.
Measurements. Lesions were identified and mea-sured in each slice and the obtained areas were addedto compute a lesion score for each defined region inboth hemispheres. Lesions detected in the rest of thebrain were also measured and a total lesion score wascomputed to represent the lesion load. Lesions weremeasured in square millimeters using the analysis sys-
FIG. 1. Region delimitation. The medial frontal region (red) was(RS). Three representative coronal views show the extent of this regiregion (green) are illustrated.
tem and the procedures adopted by Pujol et al. (1993). c
The main steps were to magnify pictures and to isolatelesions using, first, coarse contouring and, then, finesegmentation procedures. Measurements were per-formed by a single investigator blind to the patientdata. The reproducibility of our measurement methodhad previously proved to be good (Pujol et al., 1993,997). In the present study, we measured the rightedial frontal region twice in each subject and found
n intraclass correlation coefficient of 0.94 between 45airs of measurements.Statistical analysis. The Student t test was used to
ompare study variables between both groups. Pear-on’s correlation and partial correlations were used tostablish the relationship between Stroop interferenceime and lesion measurements. Stepwise multiple-re-ression analyses were used to determine the best pre-iction of the Stroop interference from regional lesioneasurements. All analyses were 2-tailed and alpha
evel of significance was set at 0.05.
RESULTS
Task performance. Multiple sclerosis patientshowed a mean (6SD) reaction time during baseline
aterally defined in the first nine slices including the reference sliceSimilarly, two of the four slices used to specify the posterior parietal
bilon.
olor naming of 689.9 (6161.7) ms, whereas the refer-
71STROOP AND MULTIPLE SCLEROSIS
ence control group showed a mean of 619.2 (693.2) ms.The 70.7 ms mean difference observed between bothgroups in this part of the test was statistically signifi-cant (t 5 2.4, df 5 71.7, P 5 0.019, pooled varianceestimate), showing a 95% confidence interval (CI) rang-ing from 11.9 to 129.5 ms. Nevertheless, patient slow-ness in naming was not uniform, as greater dispersionof the measured responses occurred in this group (Lev-ene’s test for equality of variances showed F 5 4.8 andP 5 0.032).
Stroop interference (incongruent minus congruentnaming) was also varied in patients (208.0 6 123.9ms), again showing significant differences with controlsubjects (198.6 6 77.2 ms) when tested for equality ofvariances (F 5 4.4, P 5 0.040). In this case, however,the 9.4 ms mean difference observed between bothgroups was not significant (95% CI ranged from 236.9to 55.7 ms). In addition, a similar low error rate wasregistered in both study groups (1.1 6 1.6 errors inpatients versus 0.8 6 1.1 in reference subjects).
Correlation analysis. A mild slowing effect of lesionload on the Stroop response was observed, as totallesion area correlated with Stroop interference of pa-
FIG. 2. Three-dimensional head models illustrating study findinred (left). The divisions in which lesions were more significantly relaa darker shade of red in this picture. Bicommissural reference lincomparisons with other studies. The left posterior parietal region, widentified in green on a surface-rendered model of the brain (right).
tients showing r 5 0.36 (Table 1). Correlations were
stronger for medial frontal regions, particularly for theright side, whereas no direct relationship was foundbetween posterior parietal lesion measurements andStroop interference.
In order to test the effect of each region on Stroopinterference more specifically, the correlation analysiswas repeated using partial correlations to remove theeffect of total lesion area (this global score correlatedwith region measurements always showing r . 0.70).We observed that the right, and not the left, medialfrontal region remained significantly related to theStroop interference (Table 1) and, remarkably, all pa-rietal measurements appeared with strong negativecorrelations, showing that the larger the lesion area,particularly in the left hemisphere, the lesser theStroop interference. Interestingly, total lesion area im-proved its correlation with Stroop interference to r 50.63, after correcting for posterior parietal lesion score.
The best prediction of the Stroop interference wasdetermined by means of a stepwise multiple regressionanalysis, including left and right frontal and parietallesion scores. We found that right medial frontal regionentered at the first step, showing r 5 0.48 (F 5 12.5,
The right medial frontal region and its five divisions are depicted into slowed Stroop responses (midsupracallosal and pregenual) show
(Talairach and Tournoux, 1988) are superimposed to allow resulte demyelination was associated with reduced Stroop interference, is
gs.tedesher
P 5 0.001) and left posterior parietal region in the
(
5c0ape
rsrmiy
ll(
(
72 PUJOL ET AL.
second and last step with an r square change of 0.20final multiple r 5 0.65, df 5 2,42, F 5 15.6, P , 0.001).
In the final equation, right medial frontal regionshowed a partial correlation of 0.65 (beta 5 0.81, t 5.6, P , 0.001) and left posterior parietal a partialorrelation of 20.51 (beta 5 20.56, t 5 23.8, P ,.001). Thus, lesions related to greater response delaynd those related to lesser degree of interference com-lemented each other to account for Stroop interfer-nce time variance.Medial frontal region analysis. The medial frontal
egion was further divided into 5 sectors in each hemi-phere as described in Methods. Table 2 shows theesults from the correlation analysis. We found rightidsupracallosal and right pregenual divisions show-
ng the strongest correlations, both in the direct anal-sis and after removing the lesion load effect.A stepwise multiple regression including these 10
esion measurements showed that right midsupracal-osal division entered at the first step with r 5 0.53F 5 16.4, P , 0.001) and right pregenual in the second
and last step with a significant r square change of 0.07
final multiple r 5 0.59, df 5 2,42, F 5 11.3, P , 0.001).In this case, both contributors showed positive corre-lations. The right midsupracallosal division showed apartial correlation of 0.33 (beta 5 0.34, t 5 2.3, P 50.028) and the right pregenual a partial correlation of0.32 (beta 5 0.33, t 5 2.2, P 5 0.035). It is remarkablein this analysis that two portions of the same rightmedial frontal region explained different parts of theStroop interference variance.
The left posterior parietal lesion measurement wasintroduced in this regression analysis and we ob-served a significant further contribution from thisvariable showing an r square change of 0.14 (partialcorrelation 5 20.46, t 5 23.4, P 5 0.002,). The finalcombination of right frontal and left parietal mea-surements accounted for 45% of Stroop interferencevariance (multiple r 5 0.70, df 5 3,41, F 5 13.1,P , 0.001, adjusted r square 5 0.45). Additionally,age, gender, years of education and total lesion areawere tested in this regression model, but none ofsuch variables was able to modify the presented re-sults.
Regional predominance of lesions. Our analysis in-dicates that patients with more right frontal than leftparietal lesions show the largest Stroop effect, whereasa relative predominance of parietal lesions is associ-ated with reduced interference. The difference betweenright frontal (midsupracallosal plus pregenual) andleft parietal measurements may be a simple form ofexpressing “lesion predominance” and allows a directvisual illustration of the study’s main finding (Fig. 3).
The effect of the anatomical distribution of lesions onthe Stroop response may be behaviorally relevant. Wefound that the 15 patients with the greatest right fron-tal lesion predominance showed a mean Stroop inter-ference 115.3 ms slower than control subjects (95% CIof this difference, 30.4 to 200.2 ms). By contrast, the 15patients with the greatest left parietal lesion predom-inance showed a mean Stroop interference 46.2 msfaster than the control group (95% CI, 291.8 to 20.5ms). Thus, paradoxically, there were lesions in thisstudy that “improved” Stroop performance.
The effect of baseline color naming. The baselinetask of naming colored dots was significantly related tothe Stroop interference in patients with r 5 0.59 andP , 0.001, showing that the slower the color naminghe greater the Stroop interference in this group. Thisas not the case in reference control subjects, wheree found r 5 0.08 and P 5 0.663 for this correlation. Inddition, the color naming reaction time was also sig-ificantly related to the “lesion predominance” scoreith r 5 0.41 and P 5 0.005, again showing slower
esponses for the patients with relatively more rightrontal and less left parietal lesions.
To further delimitate the specific effect of lesions on
FIG. 3. Plot of lesion predominance with Stroop interference.Lesion predominance resulted from the difference between rightmedial frontal (midsupracallosal plus pregenual) and left posteriorparietal demyelination areas and correlated with Stroop interferencetime showing r 5 0.69 and P , 0.001.
troop interference and to control any influence of the d
aseline color naming, a new regression was performedncluding both color naming and Stroop interference inhe prediction of the composite “right frontal minus leftarietal” lesion measurement. We obtained a partialorrelation of 0.61 (t 5 5.0, P , 0.001) for Stroopnterference and a partial correlation of 0.02 (t 5 0.1,
5 0.918) for color naming. These results indicate thathe main effect of this lesion combination was ontroop interference rather than on the baseline tasksed. The lesion pattern that disturbs color namingay be different. The slow color naming of our patientsay be better explained when left (as opposed to right)
rontal lesions and the occipital lobe lesions surround-ng the calcarine sulcus (a nonstudy region) are in-luded in the regression. This lesion combination ac-ounted for 42% of the color naming time variance.
DISCUSSION
We studied the effect of medial frontal and posteriorarietal demyelinating lesions on Stroop reaction timen a series of MS patients showing a non-uniformtroop effect. We found that a combination of frontalnd parietal lesion measurements accounted for 45% ofhe Stroop interference variance. Patients with moreight frontal than left parietal demyelination showedlowed Stroop responses, whereas predominance of le-ions in the left posterior parietal region was associ-ted with a reduced Stroop effect.The study of a large series representing the generalS population showed significant Stroop interference
ifferences between patients and control subjects, al-hough most of the patients were clinically normal inhe execution of this test (Rao et al., 1991). There areeports of patients showing the combination, as in ourtudy, of slowed color naming with a non-significanttroop effect (van den Burg et al., 1987; Jennekens-chinkel et al., 1990; van Dijk et al., 1992), while othertudies have found increased Stroop effect in the spe-ific subgroups of cognitively deteriorated patientsKujala et al., 1995) or those presenting pathologicalaughing and crying (Feinstein et al., 1999). The pre-ious data, therefore, emphasize the heterogeneity oftroop interference in MS.Our results suggest that, apart from clinical vari-
bles, lesion distribution partly accounts for the widetroop effect range (lesions related to both increasednd decreased Stroop interference) and the lack ofean differences between patients and control subjects
n several series. It seems that the correlation of lesionoad with Stroop interference may be truncated by theresence of parietal lesions. Indeed, we found a corre-ation between total lesion area and Stroop interfer-nce of 0.36, similar to that reported by Rao et al.1989). Nevertheless, upon previously using the Stroopask to test 41 cerebrovascular risk patients with pre-
ominantly periventricular (with few parietal) white
air
taarrborb(mlctt
pqll(iirtiseoi
btsm(
oigm1ltlle
74 PUJOL ET AL.
matter lesions, we found a correlation of 0.61 betweenquantified lesions and Stroop interference (Junque et
l., 1990). Interestingly, this correlation was very sim-lar to that observed in the present study (0.63) afteremoving parietal lesion effects.
The findings of our current report may help to definehe significance of medial frontal and posterior parietalctivations seen in functional imaging studies. Ingreement with the general view, the association ofight medial frontal demyelination with slowed Stroopesponses suggests that this region supports part of therain resources favoring Stroop conflict resolution. Theddest finding, however, was that parietal lesions wereelated to a reduced Stroop effect. This part of therain deals with enhancement of stimulus attentionCorbetta, 1998) that, in the specific Stroop situation,ay jointly involve both voluntary attention to stimu-
us color and automatic processing of the interferingolor word. The nature of our correlation suggests thathe net effect of parietal operations during the Stroopask interferes with performance.
Alternatively, a direct disturbance in the readingrocess itself may also explain parietal lesion conse-uences. The parietal region includes part of the angu-ar and supramarginal gyri, which were classically re-ated to the phonological encoding of the word stimulusGeschwind, 1979). Although functional imaging stud-es show that the most active areas during word read-ng are inferior to these gyri and to our defined parietalegion (Fiez and Petersen, 1998), connections from pos-erior reading areas to language motor output could benvolved. Unfortunately, anatomical limitations in ourtudy prevent us from ascertaining whether the criticalvent is disruption of the reading process, weakeningf the parietal attentional source enhancing the read-ng (Corbetta, 1998) or both circumstances together.
Findings from the frontal division analysis may alsoe relevant. Two right medial frontal divisions enteredhe regression, thus suggesting that different cingulateubsystems contribute to modulating Stroop perfor-ance, as was pointed out by previous researchers
George et al., 1994; Whalen et al., 1998; Derbyshire etal., 1998; Brown et al., 1999; Peterson et al., 1999).Whalen et al. (1998) functionally differentiated twoparts in the anterior cingulate. They described an af-fective pregenual division that is active when emotion-ally negative words are presented as interfering stim-ulus (Whalen et al., 1998) and a cognitive supracallosaldivision that responds during classic Stroop and itscognitive variants (Bush et al., 1998). Our pregenualand midsupracallosal regions roughly correspond tothose affective and cognitive divisions. Therefore, bothaffective and cognitive resources may be relevant inMS patients to solve the Stroop conflict. In this casealso, reduced anatomical discrimination does not allowus to distinguish between the effects of anterior cingu-
late and medial premotor areas.
We previously reported an association between dis-crete lesions of the right lateral prefrontal cortexand response errors using the same Stroop version(Vendrell et al., 1995). The present study did not spe-cifically investigate the effect of lateral prefrontal de-myelination. This is not a brain region where demyeli-nation is prominent (Lumsden, 1970). Moreover,reaction time alteration, rather than Stroop errors,typically occurs in MS. Our reports, together withother studies (Junque et al., 1990; Taylor et al., 1997;Brown et al., 1999), suggest that medial frontal areasperate mainly by modulating Stroop responses andnfluencing reaction time, whereas lateral frontal re-ions may supervise task accuracy by attempting toinimize error (Vendrell et al., 1995; Peterson et al.,
999). Furthermore, right hemisphere response modu-ation and accuracy control may be seen as mere atten-ional adjustments of the Stroop response, whereas theocus of Stroop interference most probably involvesanguage areas of the left lateral frontal cortex (Taylort al., 1997; Brown et al., 1999).
Finally, our results may emphasize that the Strooptask, as a neuropsychological test, does not assess justa single dimension of attention or even a specific fron-tal function, but rather the combined effect of anatom-ically segregated processes. Therefore, focal brain le-sions, as we observed, can either increase or decreaseStroop reaction time depending on their location.
ACKNOWLEDGMENTS
This study was supported in part by the Institut de Recerca del’Hospital de la Santa Creu i Sant Pau of Barcelona and the DireccioGeneral de Recerca de la Generalitat de Catalunya (Grant 1999SGR00328). We thank Gerald Fannon, PhD, for the revision of the manu-script.
REFERENCES
Bench, C. J., Frith, C. D., Grasby, P. M., Friston, K. J., Paulesu, E.,Frackowiak, R. S. J., and Dolan, R. J. 1993. Investigations of thefunctional anatomy of attention using the Stroop test. Neuropsy-chologia 31: 907–922.
Brown, G. G., Kindermann, S. S., Siegle, G. J., Granholm, E., Wong,E. C., and Buxton, R. B. 1999. Brain activation and pupil responseduring covert performance of the Stroop Color Word task. J. Int.Neuropsychol. Soc. 5: 308–319.
van den Burg, W., van Zomeren, A. H., Minderhoud, J. M., Prange,A. J., and Meijer, N. S. 1987. Cognitive impairment in patientswith multiple sclerosis and mild physical disability. Arch. Neurol.44: 494–501.
Bush, G., Whalen, P. J., Rosen, B. R., Jenike, M. A., McInerney, S. C.,and Rauch, S. L. 1998. The counting Stroop: An interference taskspecialized for functional neuroimaging: validation study withfunctional MRI. Hum. Brain Mapp. 6: 270–282.
Carter, C. S., Mintun, M., and Cohen, J. D. 1995. Interference andfacilitation effects during selective attention: An H2
15O PET studyof Stroop task performance. Neuroimage 2: 264–272.
Corbetta, M., Miezin, F. M., Dobneyer, S., Shulman, G. L., and
Petersen, S. E. 1991. Selective and divided attention during visual
G
J
J
P
P
P
V
W
W
75STROOP AND MULTIPLE SCLEROSIS
discriminations of shape, color, and speed: Functional anatomy bypositron emission tomography. J. Neurosci. 11: 2383–2402.
Corbetta, M. 1998. Frontoparietal cortical networks for directingattention and the eye to visual locations: Identical, independent, oroverlapping neural systems? Proc. Natl. Acad. Sci. USA 95: 831–838.
Derbyshire, S. W., Vogt, B. A., and Jones, A. K. 1998. Pain andStroop interference tasks activate separate processing modules inanterior cingulate cortex. Exp. Brain Res. 118: 52–60.
van Dijk, J. G., Jennekens-Schinkel, A., Caekebeke, J. F., and Zwin-derman, A. H. 1992. Are event-related potentials in multiple scle-rosis indicative of cognitive impairment? Evoked and event-relatedpotentials, psychometric testing and response speed: A controlledstudy. J. Neurol. Sci. 109: 18–24.
Feinstein, A., O’Connor, P., Gray, T., and Feinstein, K. 1999. Patho-logical laughing and crying in multiple sclerosis: A preliminaryreport suggesting a role for the prefrontal cortex. Mult. Scler. 5:69–73.
Fiez, J. A., and Petersen, S. E. 1998. Neuroimaging studies of wordreading. Proc. Natl. Acad. Sci. USA 95: 914–921.
George, M. S., Ketter, T. A., Parekh, P. I., Rosinsky, N., Ring, H.,Casey, B. J., Trimble, M. R., Horwitz, B., Herscovitch, P., and Post,R. M. 1994. Regional brain activity when selecting a responsedespite interference: An H2
15O PET study of the Stroop and anemotional Stroop. Hum. Brain Mapp. 1: 194–209.eschwind, N. 1979. Specialization of the human brain. Sci. Am.241: 180–199.
ennekens-Schinkel, A., Lanser, J. B., van der Velde, E. A., Sanders,E. A. 1990. Performances of multiple sclerosis patients in tasksrequiring language and visuoconstruction. Assessment of outpa-tients in quiescent disease stages. J. Neurol. Sci. 95: 89–103.
unque, C., Pujol, J., Vendrell, P., Bruna, O., Jodar, M., Ribas, J. C.,Vinas, J., Capdevila, A., and Martı-Vilalta, J. L. 1990. Leuko-araiosis on magnetic resonance imaging and speed of mental pro-cessing. Arch. Neurol. 47: 151–156.
Kujala, P., Portin, R., Revonsuo, A., and Ruutiainen, J. 1995. Atten-tion related performance in two cognitively different subgroups ofpatients with multiple sclerosis. J. Neurol. Neurosurg. Psychiatry59: 77–82.
Kurtzke, J. F. 1983. Rating neurologic impairment in multiple scle-rosis: An expanded disability status scale (EDSS). Neurology 33:1444–1452.
Lumsden, C. E. 1970. The neuropathology of multiple sclerosis. InHandbook of Clinical Neurology (P. J. Vinken and G. W. Bruyn,Eds.), pp. 217–309. North-Holland, Amsterdam.
MacLeod, C. M. 1991. Half a century of research on the Stroop effect:An integrative review. Psychol. Bull. 109: 163–203.
Pardo, J. V., Pardo, P. J., Janer, K. W., and Raichle, M. E. 1990. Theanterior cingulate cortex mediates processing selection in theStroop attentional conflict paradigm. Proc. Natl. Acad. Sci. USA87: 256–259.
Pardo, J. V., Fox, P. T., and Raichle, M. E. 1991. Localization of ahuman system for sustained attention by positron emission tomog-
raphy. Nature 349: 61–64.
Passingham, R. E. 1996. Attention to action. Philos. Trans. R. Soc.Lond. B Biol. Sci. 351: 1473–1479.
Peterson, B. S., Skudlarski, P., Gatenby, J. C., Zhang, H., Anderson,A. W., and Gore, J. C. 1999. An fMRI study of Stroop word-colorinterference: Evidence for cingulate subregions subserving multi-ple distributed attentional systems. Biol. Psychiatry 45: 1237–1258.
Poser, C. M., Paty, D. W., Scheinberg, L., McDonald, W. I., Davis,F. A., Ebers, G. C., Johnson, K. P., Sibley, W. A., Silberberg, D. H.,and Tourtellotte, W. W. 1983. New diagnostic criteria for multiplesclerosis: Guidelines for research protocols. Ann. Neurol. 13: 227–231.
osner, M. I., Walker, J. A., Friedrich, F. J., and Rafal, R. D. 1984.Effects of parietal injury on covert orienting of attention. J. Neu-rosci. 4: 1863–1874.
ujol, J., Vendrell, P., Junque, C., Martı-Vilalta, J. L., and Capdev-ila, A. 1993. When does human brain development end? Evidenceof corpus callosum growth up to adulthood. Ann. Neurol. 34: 71–75.
ujol, J., Bello, J., Deus, J., Marti-Vilalta, J. L., and Capdevila, A.1997. Lesions in the left arcuate fasciculus region and depressivesymptoms in multiple sclerosis. Neurology 49: 1105–1110.
Rao, S. M., Leo, G. J., Haughton, V. M., Aubin-Faubert, P. St., andBernardin, L. 1989. Correlation of magnetic resonance imagingwith neuropsychological testing in multiple sclerosis. Neurology39: 161–166.
Rao, S. M., Leo, G. J., Bernardin, L., and Unverzagt, F. 1991. Cog-nitive dysfunction in multiple sclerosis. I. Frequency, patterns,and prediction. Neurology 41: 685–691.
Sipe, J. C., Knobler, R. L., Braheny, S. L., Rice, G. P., Panitch, H. S.,and Oldstone, M. B. 1984. A neurologic rating scale (NRS) for usein multiple sclerosis. Neurology 34: 1368–1372.
Talairach, J., and Tournoux, P. 1988. Co-Planar Stereotaxic Atlas ofthe Human Brain. Thieme Medical, New York.
Taylor, S. F., Kornblum, S., Minoshima, S., Oliver, L. M., andKoeppe, R. A. 1994. Changes in medial cortical blood flow with astimulus-response compatibility task. Neuropsychologia 32: 249–255.
Taylor, S. F., Kornblum, S., Lauber, E. J., Minoshima, S., andKoeppe, R. A. 1997. Isolation of specific interference processingin the Stroop task: PET activation studies. Neuroimage 6: 81–92.endrell, P., Junque, C., Pujol, J., Jurado, M. A., Molet, J., andGrafman, J. 1995. The role of prefrontal regions in the Stroop task.Neuropsychologia 33: 341–352.echsler, D. 1955. Wechsler Adult Intelligence Scale. PsychologicalCorp., New York, NY.halen, P. J., Bush, G., McNally, R. J., Wilhelm, S., McInerney,S. C., Jenike, M. A., and Rauch, S. L. 1998. The emotional countingStroop paradigm: A functional magnetic resonance imaging probeof the anterior cingulate affective division. Biol. Psychiatry 44:
