Article

An fMRI Stroop Task Study of Ventromedial Prefrontal

Cortical Function in Pathological Gamblers

Marc N. Potenza, M.D., Ph.D. Objective: Function of the ventromedial demonstrated decreased activity in the
prefrontal cortex has been implicated in left ventromedial prefrontal cortex rela-
Hoi-Chung Leung, Ph.D. impulse control. The authors used the tive to the comparison subjects. Both
Stroop paradigm to test attention and re- groups demonstrated similar activity
Hilary P. Blumberg, M.D. sponse inhibition during the presentation c h an g e s in mu lt ip le br ain re g io n s ,
of congruent and incongruent stimuli in including activation of the dorsal ante-
Bradley S. Peterson, M.D. male pathological gamblers and a group rior cingulate and dorsolateral frontal
of comparison subjects. cortex.
Robert K. Fulbright, M.D. Method: Event-related functional mag- Conclusions: P at h o l og i c a l g a m b le rs
netic resonance imaging was used to ex-
Cheryl M. Lacadie, B.S. share many neural correlates of Stroop
amine ventromedial prefrontal cortex task performance with healthy subjects
Pawel Skudlarski, Ph.D. function during Stroop performance. but differ in a brain region previously im-
Results: In response to infrequent in- plicated in disorders characterized by
John C. Gore, Ph.D. congruent stimuli, pathological gamblers poor impulse control.

(Am J Psychiatry 2003; 160:1990–1994)

D espite significant prevalence estimates and associa-

tions with adverse consequences (1), pathological gam-
ing the Structured Clinical Interview for DSM-IV (SCID) (14) for
the pathological gamblers but not the comparison subjects. The
Structured Clinical Interview for Pathological Gambling (unpub-
bling has received relatively little study, particularly in its
lished, available from the first author upon request), a SCID-com-
neurobiological underpinnings (2, 3). Function of the ven- patible module based on DSM-IV criteria, was used to confirm
tromedial prefrontal cortex has been widely implicated in the diagnosis of pathological gambling.
impulse regulation (4–9). Previously, we found that patho- Data were obtained from 13 pathological gamblers and 11
logical gamblers exhibit decreased activity in the ventro- healthy comparison subjects. Pathological gamblers and com-
medial prefrontal cortex during presentation of gambling parison subjects were similar in age (mean=35.15 years [SD=7.97]
and 29.00 years [SD=7.81], respectively; χ2=3.62, df=1, p=0.07). Six
cues (10). Our group has used an event-related Stroop
pathological gamblers were considered nicotine dependent
paradigm (involving the infrequent presentation of mis- (Fagerstrom Test for Nicotine Dependence scores ≥5); no com-
matched color-word stimuli and frequent presentation of parison subjects reported smoking. All subjects were high school
matched color-word stimuli) in functional magnetic reso- graduates. The racial/ethnic composition for the pathological
nance imaging (fMRI) studies to identify the neural corre- gamblers was Caucasian (N=8), African American (N=4), and His-
panic (N=1); among the comparison subjects, nine were Cauca-
lates of the Stroop effect (11, 12). Since the incongruent
sian, one was Hispanic, and one was of unknown ethnicity. All
stimuli presentations in the Stroop paradigm require re- subjects denied psychoactive drug use with the exception of nic-
sponse inhibition and pathological gambling involves otine or caffeine for the 72 hours before the fMRI protocol.
impaired inhibitory control of gambling behaviors, we Subjects performed the Stroop task as previously described
predicted that pathological gamblers would differ from (11, 12). Briefly, 9–10 runs of 102 stimuli were presented. Each
comparison subjects in ventromedial prefrontal cortex ac- stimulus was presented for 1300 msec, with an interstimulus in-
terval of 350 msec. Incongruent stimuli were presented pseudo-
tivity during Stroop task performance. randomly every 13–16 congruent stimuli (i.e., 21.45–26.4 seconds
apart). Subjects practiced aloud the task for one or two runs be-
Method fore scanning and silent naming was performed during imaging
to minimize motion artifact. Performance was assessed following
This research was approved by the Yale Human Investigations scanning to measure reaction time (difference in response times
Committee, and all subjects provided written informed consent. to incongruent versus congruent stimuli) and percentage of cor-
All subjects were male, 18–65 years of age, native English speak- rect responses to incongruent stimuli (6, 11, 12).
ers, and without past or present major neurological injury or ill-
Image Acquisition
ness. Two pathological gamblers were left-handed, and all other
subjects were right-handed. Pathological gamblers had an aver- Images were obtained with a 1.5-T GE Signa MRI system
age South Oaks Gambling Screen (13) score of 12.62 (SD=3.93). equipped with an echo-planar imaging system, a standard quadra-
Pathological gamblers met DSM-IV-TR criteria for pathological ture head coil, and a T2*-sensitive gradient-recalled, single-shot,
gambling and were free of any other active axis I disorder except echo-planar pulse sequence (6, 11, 12). Conventional T1-weighted
nicotine dependence. Comparison subjects were free of any ac- spin-echo sagittal anatomic images (TE=11 msec, TR=667 msec,
tive axis I disorder. Diagnostic information was determined by us- field of view=24 cm, slice thickness=5 mm, gap=1 mm, 256×128

1990 http://ajp.psychiatryonline.org Am J Psychiatry 160:11, November 2003

 POTENZA, LEUNG, BLUMBERG, ET AL.

data matrix) were obtained for slice localization. Ten T1-weighted stimuli, pathological gamblers and comparison subjects
oblique axial slices (TE=13 msec, TR=500 msec, field of view=40×40 demonstrated activity changes in many of the same brain
cm, gap=1 mm, 256×192 data matrix) were obtained parallel to the
regions (Figure 1, Table 1). Activations observed for both
plane transecting the anterior and posterior commissures (6, 11,
12). Functional images were obtained by using a single-shot, echo- groups involved the dorsal anterior cingulate, right middle
planar gradient-echo sequence (TR=1.65 seconds; TE=60 msec; and inferior frontal gyri, bilateral inferior frontal gyri, right
flip angle=60°; field of view=40×20 cm; slice thickness=7 mm, with insula, and right thalamus. Both groups demonstrated
skip varying between 0.5 and 2 mm to keep strict correspondence decreased activity in the ventral anterior cingulate. Patho-
to Talairach space; in-plane resolution=3.12×3.12 mm; number of
logical gamblers and comparison subjects were distin-
images per slice=102) at the same locations.
guished by differences in signal change in the left ventro-
Data Analysis medial prefrontal cortex (Figure 1, Table 1). This region of
Data were motion corrected for three translational directions the ventromedial prefrontal cortex involves the left middle
and three possible rotations (15). Runs with motion in excess of and superior frontal gyri and borders the superior frontal
1.5 mm displacement and 2 degrees rotation were rejected. Sim- sulcus laterally and the orbitofrontal cortex ventrally. The
ilar numbers of runs per subject were used from pathological difference in signal change was attributable largely to de-
gamblers (mean=8.69, SD=1.65) and comparison subjects
creased activity in pathological gamblers, with a lesser
(mean=9.00, SD=0.89) (χ2=0.30, df=1, p=0.59). Corrected images
were spatially filtered by using a Gaussian filter with a full width contribution from increased activity in healthy subjects
at half maximum of 6.25 mm. (observed in comparison maps of less stringent signifi-
T1-weighted axial anatomical images and corresponding func- cance thresholds than p<0.01 [not shown]). Correlation
tional images for each subject were transformed into a common analyses suggest that the between-group difference in the
stereotactic space by piece-wise linear warping. Pixel-based left ventromedial prefrontal cortex was not attributable to
changes in fMRI signal associated with incongruent stimuli pre-
sentation were calculated as follows. The average signal of the six
between-group differences in age or smoking status (data
images preceding incongruent stimuli presentation was sub- not shown).
tracted from each of the eight images following the incongruent
stimuli at each pixel by using voxel-based statistical mapping
methodologies to generate activation time courses as described
Discussion
previously (6, 10–12). Linear contrasts between conditions were
used to calculate t values for each voxel and data randomizations
Past investigations have reported cognitive biases (16)
using nonparametric statistics employed to create distributions and attentional deficits (17) in pathological gamblers, and
with which to assign statistical significance to observed between- our group has found decreased activity in the ventrome-
group differences at each voxel location (p-map generation). dial prefrontal cortex in pathological gamblers during
Time course p-maps (at p=0.05 with voxels in contiguous sets of gambling cue presentation (10). The Stroop paradigm was
20 voxels each) were generated to identify between-group differ-
ences in ventromedial prefrontal cortex activity (12). The largest
selected to investigate aspects of attention and response
between-group differences were observed in the third and fourth inhibition and ventromedial prefrontal cortex function in
images. Therefore, the average signal of the six images preceding pathological gamblers and healthy comparison subjects.
incongruent stimuli presentation was subtracted from the aver- Behavioral performance measures of the Stroop task were
age of the third and fourth images following the incongruent similar across groups, although sensitivity is decreased by
stimuli at each pixel at a significance threshold of p=0.01 for each
subject group and between subject groups (11, 12). Significance
the sizable variance. Task-related brain activities were
was assigned by using voxel-based data randomizations as de- largely similar across groups, suggesting that brain func-
scribed previously (6, 10–12). Within-group correlation analyses tioning and its behavioral consequences were similar in
were performed to investigate the potential effects of age and pathological gamblers and healthy subjects. In particular,
smoking status (Fagerstrom score) on observed between-group both groups demonstrated activation of the dorsolateral
differences in signal activity. To ensure that signal arising at the
lower part of the brain was not significantly impacted by suscep-
prefrontal cortex and dorsal anterior cingulate, brain re-
tibility artifact, the signal from lower brain regions was evaluated gions previously implicated in conflict monitoring and
in all cases to verify that it was >75% of the signal from cortical re- cognitive control and identified as activated in the Stroop
gions in more uniform areas. SUPER-ANOVA (Abacus Concepts, paradigm (6, 11, 12, 18–20).
Inc.; Berkeley, Calif.) was used to perform t tests to analyze data
from Stroop performance (reaction time and correct responses)
The most robust between-group difference was ob-
(11, 12). served in the left ventromedial prefrontal cortex, a region
involving the superior aspect of the orbitofrontal cortex.
Results The ventromedial prefrontal cortex has been implicated in
decision making (4), and its orbitofrontal component is
The pathological gamblers and comparison subjects thought to participate in the processing of rewards during
performed similarly on the Stroop task in terms of incor- the expectancy and experiencing of monetary gains or
rect response percentage (mean=12.09% [SD=16.34%] and losses (21). The orbitofrontal cortex also appears to be-
7.14% [SD=10.10%], respectively; χ2=0.70, df=1, p=0.41) come activated when there is insufficient information to
and reaction times to incongruent stimuli (mean=215.77 determine the appropriate course of action or when deci-
msec [SD=126.20] and 197.55 msec [SD=66.77]; χ2=0.14, sion-making action requires suppression of previously re-
df=1, p=0.71). Following presentation of incongruent warded responses (22). Individuals with impaired impulse

Am J Psychiatry 160:11, November 2003 http://ajp.psychiatryonline.org 1991

STROOP TASK PERFORMANCE OF PATHOLOGICAL GAMBLERS

FIGURE 1. Activation Differences in Pathological Gamblers and Comparison Subjects Following Presentation of Incongruent
Stimuli (Stroop task)

Right Left

z=50

Increased Activity
p<0.001

z=41

z=32

p<0.01

z=23

Decreased Activity
z=14 p<0.01

z=5

p<0.001
z=–5

Pathological Comparison Pathological

Gamblers (N=13) Subjects (N=11) Gamblers (N=13)
Minus Comparison
Subjects (N=11)

control have demonstrated orbitofrontal cortex/ventro- tex activation in pathological gamblers during incongru-
medial prefrontal cortex dysfunction (5–8), including ent stimulus presentation suggests dysfunction in this
during Stroop task performance (23). The finding of de- brain region. Further investigations are needed to exam-
creased orbitofrontal cortex/ventromedial prefrontal cor- ine directly the extent to which ventromedial prefrontal

1992 http://ajp.psychiatryonline.org Am J Psychiatry 160:11, November 2003

 POTENZA, LEUNG, BLUMBERG, ET AL.

TABLE 1. Regional Brain Activity Changes in Pathological Gamblers and Comparison Subjects Following Presentation of
Incongruent Stimuli (Stroop task)a
Slices Talairach
Involved Coordinates Size
Subject Group and Activity Change (z levels) Brain Regions (x, y, z) (mm3)
Pathological gamblers
Increased activity 41 Right parietal cortex (inferior parietal lobule) –41, –45, 41 4,433
41, 32 Left parietal cortex (inferior parietal lobule) 50, –41, 38 1,904
41, 32, 23 Dorsal anterior cingulate 2, 15, 34 9,759
41, 32, 23 Right prefrontal cortex (inferior and middle frontal gyri) –42, 5, 30 7,587
23 Left prefrontal cortex (inferior frontal gyrus) 46, –1, 23 1,547
23 Right prefrontal cortex (middle frontal gyrus) –39, 41, 23 625
5 Right thalamus –5, –18, 5 803
5, –5 Right insula/inferior frontal gyrus –33, 12, 1 15,858
5, –5 Left insula/inferior frontal gyrus 38, 8, 0 11,009
Decreased activity –5 Ventral anterior cingulate –3, 36, –5 4,582
Comparison subjects
Increased activity 50, 41, 32 Dorsal anterior cingulate –1, 11, 42 7,885
41 Right parietal cortex (inferior parietal lobule) –40, –56, 41 2,668
41, 32, 23 Right prefrontal cortex (inferior and middle frontal gyri) –43, 7, 35 3,183
32 Left parietal cortex (inferior parietal lobule) 32, –50, 32 952
23 Right prefrontal cortex (middle frontal gyrus) –39, 41, 23 1,934
23 Anterior cingulate –2, 31, 23 1,101
14 Right thalamus –11, –7, 14 1,130
5 Left insula 28, 11, 5 1,190
5 Right insula/inferior frontal gyrus –36, 13, 5 4,552
–5 Right basal ganglia –15, 5, –5 833
Decreased activity –5 Ventral anterior cingulate 2, 40, –5 982
Pathological gamblers minus comparison subjects
Decreased activity –5 Left ventromedial prefrontal cortex 20, 36, –5 893
a Coordinates listed identify approximate x, y, and z coordinates based on the average center of mass for an activity change. Activity changes
listed were grouped if they were contiguous across z-level slices. Negative x values indicate the right side of the brain, negative y values in-
dicate brain posterior to the anterior commissure, and negative z values indicate brain regions inferior to the plane defined by the anterior
and posterior commissures. An additional activity change (decrease) was observed in the comparison group (centered at –24, –33, 23, with
a size of 1,190 mm3) but was not tabulated as the change appeared to be artifactual.

cortex dysfunction underlies specific aspects of the pa- The authors thank Hedy Sarofin, Cheryl McMurray, Terry Hickey,
Mary Wilber, Erin Reutenauer, and Kathleen Colonese for technical
thology of pathological gambling, including poor decision support.
making and rapid temporal discounting of rewards (24).
Future investigations should address limitations of the
present study by using a larger, more diverse sample (with References
women); matching precisely for age and race/ethnicity; 1. Potenza MN, Kosten TR, Rounsaville BJ: Pathological gambling.
examining the potential effect of tobacco use; obtaining JAMA 2001; 286:141–144
2. Eber GB, Shaffer HJ: Trends in bio-behavioral gambling studies
structured clinical interview data on all subjects; and cor-
research: quantifying citations. J Gambl Stud 2000; 16:461–467
relating on-line Stroop task performance with measures of 3. Potenza MN: The neurobiology of pathological gambling.
brain activity. Semin Clin Neuropsychiatry 2001; 6:217–226
4. Bechara A: Neurobiology of decision-making: risk and reward.
Semin Clin Neuropsychiatry 2001; 6:205–216
Received Oct. 29, 2002; revision received Jan. 13, 2003; accepted Jan.
5. Siever LJ, Buchsbaum MS, New AS, Spiegel-Cohen J, Wei T,
17, 2003. From the Department of Psychiatry and the Department of
Diagnostic Radiology, Yale University School of Medicine; the Connecti-
Hazlett EA, Sevin E, Nunn M, Mitropoulou V: d,l-Fenfluramine
cut Mental Health Center; and the West Haven VA Hospital, West Ha- response in impulsive personality disorder assessed with
ven, Conn. Address reprint requests to Dr. Potenza, Yale University [18F]fluorodeoxyglucose positron emission tomography. Neu-
School of Medicine, Connecticut Mental Health Center, Room S-104, 34 ropsychopharmacology 1999; 20:413–423
Park St., New Haven, CT 06519; [email protected] (e-mail). 6. Blumberg HP, Stern E, Ricketts S, Martinez D, de Asis J, White T,
Supported by Young Investigator Awards from the National Alli- Epstein J, Isenberg N, McBride PA, Kemperman I, Emmerich S,
ance for Research in Schizophrenia and Depression (Drs. Potenza Dhawan V, Eidelberg D, Kocsis JH, Silbersweig DA: Rostral and
and Blumberg); a Drug Abuse Research Scholar Program in Psychia- orbital prefrontal cortex dysfunction in the manic state of bi-
try Award from APA (Dr. Potenza); National Institute on Drug Abuse polar disorder. Am J Psychiatry 1999; 156:1986–1988
grant DA-00366 (Dr. Potenza); funding from the National Center for 7. Pietrini P, Guazzelli M, Basso G, Jaffe K, Grafman J: Neural cor-
Responsible Gaming (Dr. Potenza), the Stanley Foundation (Dr. Blum- relates of imaginal aggressive behavior assessed by positron
berg), the Charles A. Dana Foundation (Dr. Peterson), and the Su- emission tomography in healthy subjects. Am J Psychiatry
zanne Crosby Murphy Endowment at Columbia University College of 2000; 157:1772–1781
Physicians and Surgeons (Dr. Peterson); VA grants for Research Career 8. New AS, Hazlett EA, Buchsbaum MS, Goodman M, Reynolds D,
Development (Dr. Blumberg) and Merit Review (Dr. Blumberg) and Mitropoulou V, Sprung L, Shaw RB Jr, Konigsberg H, Platholi J,
the Research Enhancement Award Program (Drs. Leung and Blum- Silverman J, Siever LJ: Blunted prefrontal cortical 18-fluorode-
berg); and NIMH grants MH-01232 and MH-59139 (Dr. Peterson). oxyglucose positron emission tomography response to meta-

Am J Psychiatry 160:11, November 2003 http://ajp.psychiatryonline.org 1993

STROOP TASK PERFORMANCE OF PATHOLOGICAL GAMBLERS

chlorophenylpiperazine in impulsive aggression. Arch Gen Psy- 16. McCusker CG, Gettings B: Automaticity of cognitive biases in ad-
chiatry 2002; 59:621–629 dictive behaviours: further evidence with gamblers. Br J Clin
9. Best M, Williams JM, Coccaro EF: Evidence for a dysfunctional Psychol 1997; 36(part 4):543–554
prefrontal circuit in patients with an impulsive aggressive dis- 17. Rugle L, Melamed L: Neuropsychological assessment of atten-
order. Proc Natl Acad Sci USA 2002; 99:8448–8453 tion problems in pathological gamblers. J Nerv Ment Dis 1993;
181:107–112
10. Potenza MN, Steinberg MA, Skudlarski P, Fulbright RK, Lacadie
18. Bush G, Luu P, Posner MI: Cognitive and emotional influences
CM, Wilber MK, Rounsaville BJ, Gore JC, Wexler BE: Gambling
in anterior cingulate cortex. Trends Cogn Sci 2000; 4:215–222
urges in pathological gambling: a functional magnetic reso-
19. Peterson BS, Skudlarski P, Gatenby JC, Zhang H, Anderson AW,
nance imaging study. Arch Gen Psychiatry 2003; 60:828-836
Gore JC: An fMRI study of Stroop word-color interference: evi-
11. Leung H-C, Skudlarski P, Gatenby JC, Peterson BS, Gore JC: An dence for cingulate subregions subserving multiple distributed
event-related functional MRI study of the Stroop color word in- attentional systems. Biol Psychiatry 1999; 45:1237–1258
terference task. Cereb Cortex 2000; 10:552–560 20. van Veen V, Cohen JD, Botvinick MM, Stenger VA, Carter CS: An-
12. Peterson BS, Kane MJ, Alexander GM, Lacadie C, Skudlarski P, terior cingulate cortex, conflict monitoring, and levels of pro-
Leung H-C, May J, Gore JC: An event-related functional MRI cessing. Neuroimage 2001; 14:1302–1308
study comparing interference effects in the Simon and Stroop 21. Breiter HC, Aharon I, Kahneman D, Dale A, Shizgal P: Func-
task. Cogn Brain Res 2002; 13:427–440 tional imaging of neural responses to expectancy and experi-
ence of monetary gains and losses. Neuron 2001; 30:619–639
13. Lesieur HR, Blume SB: The South Oaks Gambling Screen (SOGS):
22. Elliott R, Dolan RJ, Frith CD: Dissociable functions in the medial
a new instrument for the identification of pathological gam-
and lateral orbitofrontal cortex: evidence from human neu-
blers. Am J Psychiatry 1987; 144:1184–1188
roimaging studies. Cereb Cortex 2000; 10:308–317
14. First MB, Spitzer RL, Gibbon M, Williams JBW: Structured Clini-
23. Blumberg HP, Leung HC, Skudlarski P, Lacadie CM, Fredericks
cal Interview for DSM-IV Disorders, Patient Edition (SCID-P). CA, Harris BC, Charney DS, Gore JC, Krystal JH, Peterson BS: A
New York, New York State Psychiatric Institute, Biometrics Re- functional magnetic resonance imaging study of bipolar disor-
search, 1995 der: state- and trait-related dysfunction in ventral prefrontal
15. Friston KJ, Holmes AP, Worsley KJ, Pooline J-B, Frith CD, Frack- cortices. Arch Gen Psychiatry 2003; 60:601–609
owiak RSJ: Statistical parametric maps in functional imaging: a 24. Bechara A: Risky business: emotion, decision-making, and ad-
general linear approach. Hum Brain Mapp 1995; 2:184–210 diction. J Gambl Stud 2003; 19:23–51

1994 http://ajp.psychiatryonline.org Am J Psychiatry 160:11, November 2003