Increased Dopamine D2/D3 Receptor Binding After
Recovery from Anorexia Nervosa Measured by
Positron Emission Tomography and [11C]Raclopride
Guido K. Frank, Ursula F. Bailer, Shannan E. Henry, Wayne Drevets, Carolyn C. Meltzer, Julie C. Price,
Chester A. Mathis, Angela Wagner, Jessica Hoge, Scott Ziolko, Nicole Barbarich-Marsteller,
Lisa Weissfeld, and Walter H. Kaye
Background: Several lines of evidence support the possibility that disturbances of dopamine (DA) function could contribute to
alterations of weight, feeding, motor activity, and reward in anorexia nervosa (AN).
Methods: To assess possibly trait-related disturbances but avoid confounding effects of malnutrition, 10 women who were recovered
from AN (REC AN) were compared with 12 healthy control women (CW). Positron emission tomography with [11C]raclopride was used
to assess DA D2/D3 receptor binding.
Results: The women who were recovered from AN had significantly higher [11C]raclopride binding potential in the antero-ventral
striatum than CW. For REC AN, [11C]raclopride binding potential was positively related to harm avoidance in the dorsal caudate and
Conclusions: These data lend support for the possibility that decreased intrasynaptic DA concentration or increased D2/D3 receptor
density or affinity is associated with AN and might contribute to the characteristic harm avoidance or increased physical activity found
in AN. Most intriguing is the possibility that individuals with AN might have a DA related disturbance of reward mechanisms
contributing to altered hedonics of feeding behavior and their ascetic, anhedonic temperament.
Key Words: Anorexia nervosa, dopamine, positron emission to-
mography, [11C]raclopride, eating disorders, brain imaging
Psychiatric Association 2000). Women with AN often have a
cluster of stereotypic behaviors that include extremes of food
ingestion, anxious and obsessive thoughts, and increased physical
activity. Individuals with AN have anhedonic, overcontrolled, per-
fectionistic personalities (Anderluh et al 2003). They seem to find
little that is rewarding in life, aside from losing weight.
Several lines of evidence suggest dopamine (DA) function
could contribute to the pathophysiology of AN. Dopamine
function contributes to the modulation of motor activity (Alex-
ander et al 1990), weight and feeding behaviors (Halford et al
2004), and reinforcement and reward (Volkow et al 2002). There
is some indication that individuals with AN respond to typical
and atypical neuroleptics (Brewerton 2004; Cassano et al 2003).
Women recovered from restricting type AN (REC AN-R) had a
norexia nervosa (AN) is a disorder of unknown etiology
that tends to begin in adolescence in women who
develop a relentless desire to lose weight (American
reduction of the DA metabolite homovanillic acid (HVA) in
cerebrospinal fluid (CSF) compared with control women (CW)
(Kaye et al 1999). Finally, individuals with AN have altered
frequency of functional polymorphisms of D2 receptor genes
that might affect receptor transcription and translation efficiency
(Bergen et al, in press).
This study used positron emission tomography (PET) imaging
with the radioligand [11C]raclopride to assess DA D2/D3 receptor
function in AN. To avoid the confounding effects of malnutrition
on DA activity, women were investigated who had recovered for
one or more years from AN. Recent studies suggest that temper-
ament and personality traits occur premorbidly and persist after
recovery from AN (Bulik et al 1997; Srinivasagam et al 1995), as
do alterations in DA activity (Kaye et al 1999). We hypothesized
that REC AN would have reduced intrasynaptic DA in reward-
related brain regions such as the antero-ventral striatum (includ-
ing the nucleus accumbens) and this would be reflected by
increased DA receptor binding. Such a disturbance could then
suggest a trait disturbance.
Methods and Materials
Ten subjects who were recovered from AN (3 pure restricting-
type [AN-R], 1 restricting-purge-type [AN-P], and 6 binge-purge-
type AN [AN-BP]) were compared with 12 CW. Subjects were
previously treated in the eating disorders treatment program at
the Western Psychiatric Institute and Clinic, Pittsburgh, Pennsyl-
vania, or were recruited through advertisements. To be consid-
ered “recovered,” for at least one year before the study, subjects
had to 1) maintain a weight above 85% average body weight
(Metropolitan Life Insurance 1983), 2) have regular menstrual
cycles, and 3) have not binged, purged, or engaged in significant
restrictive eating patterns. Additionally, subjects must not have
used psychoactive medication such as antidepressants and not
met criteria for alcohol or drug abuse or dependence, major
depressive disorder, or severe anxiety disorder within three
months of the study. During the ill state, all studied recovered
subjects fulfilled criteria for full AN diagnostic criteria (American
From the Departments of Radiology (WD, CCM, JCP, CAM, JH, SZ) and Neu-
rology (CCM), Presbyterian University Hospital; Department of Psychia-
try, Western Psychiatric Institute and Clinic (GKF, UFB, SHE, WD, CCM,
AW, WK), School of Medicine; Department of Biostatistics (LW), Univer-
sity of Pittsburgh, Pittsburgh, Pennsylvania; Department of Child and
San Diego, San Diego, California; Department of General Psychiatry
(UFB), University Hospital of Psychiatry, Medical University of Vienna,
Vienna, Austria; Neuroimaging Section (WD), Mood and Anxiety Disor-
ders Program, National Institute of Mental Health, National Institutes of
(N B-M), Department of Neurobiology and Behavior, State University of
New York at Stony Brook, Stony Brook, New York.
Address reprint requests to Walter H. Kaye, M.D., University of Pittsburgh,
Western Psychiatric Institute and Clinic, 3811 O’Hara Street, Iroquois
Building, Suite 600, Pittsburgh, PA 15213; E-mail: firstname.lastname@example.org.
Received February 14, 2005; revised April 19, 2005; accepted May 3, 2005.
BIOL PSYCHIATRY 2005;58:908–912
© 2005 Society of Biological Psychiatry
Psychiatric Association 2000). Subjects were interviewed by a
psychiatrist who ascertained the absence of pathological eating
patterns or other psychopathology. Eating disorder behaviors
were assessed with the Eating Disorders Inventory (EDI-2 (Gar-
ner 1991), perfectionism with the Multidimensional Perfection-
ism Scale (Frost et al 1990), and temperament was assessed with
the Temperament and Character Inventory (TCI) (Cloninger et al
1994). This study was approved by the University of Pittsburgh
Istitutional Review Board, and all subjects gave written informed
consent. The PET imaging was performed during the first 10 days
of the follicular phase of the menstrual cycle for all subjects. The
follicular phase was determined by history. Subjects were admit-
ted to a research laboratory on the eating disorders unit of
Western Psychiatric Institute and Clinic at 9:00 PM of the day
before the PET study for adaptation to the laboratory and for
psychological assessments. The PET study was done the next
day. All subjects had the same standardized, monoamine con-
trolled (low protein) breakfast on the morning of the study.
All subjects underwent magnetic resonance (MR) imaging
before the PET scan, and MR and PET image data were co-
registered as previously described (Frank et al 2002). The PET
scans were acquired on an ECAT HR? PET scanner (CTI PET
systems, Knoxville, Tennessee) in three-dimensional mode (63
transaxial planes, 2.4-mm thickness; in-plane resolution ? 4.1
mm full-width at half-maximum over a 15.2-cm field of view).
Twenty to 25 frames were collected over an acquisition time of
120 min. The PET frames were visually inspected for head
motion and a postprocessing correction was performed. The
regions of interest (ROIs) were hand drawn (GKF, who was blind
to the diagnosis) on the coregistered MR images as described by
Drevets et al (2001) and applied to the dynamic PET data to
generate time-activity curves. Left and right ROIs were analyzed
and then combined. A ROI sampling of the cerebellum was also
performed and used as reference region for non-specific binding.
We selected the antero-ventral striatum including the nucleus
accumbens as primary ROI; however, to test for spill-over effects
as well as to conform with convention, the ventral putamen,
dorsal caudate nucleus, middle caudate nucleus, and the dorsal
putamen were also assessed. We did not expect alterations in
those areas. High specific activity [11C]raclopride was used as
radio tracer, and methods for synthesis, administration, and
blood sampling in our group have been described previously
(Drevets et al 2001).
For the imaging data analysis, two methods were applied.
First, the Logan graphical method (Logan et al 2001) was used as
described previously (Drevets et al 2001). For this method, the
binding measure is based upon the ratio of each ROI distribution
volume (Dv) value to the cerebellar Dv value (Dv ROI/Dv
Cer ? DvRatio, DvR) and the binding potential (BP ? DvR-1).
The Logan method has the advantage of providing information
on the cerebellum, wherein the tissue radioactivity concentra-
tions predominantly reflect free and nonspecifically bound ra-
diotracer. The disadvantage is that arterial blood sampling is
required during the scanning procedure and this is not always
available. We calculated results for the Logan method in all
subjects that had arterial blood sampling during the study. If
results are robust, the results from the Logan method should be
similar to the second method applied, the Reference Tissue
Model (Lammertsma and Hume 1996). Here, the [11C]raclopride
binding measure is derived from the kinetic rate constant k3/k4,
which is equivalent to Dv ratio-1.
A previously validated two-component MR-based partial vol-
ume correction algorithm was applied to correct the PET data for
the potential dilutional effect of expanded CSF (Meltzer et al
1999), because cerebral volume loss in AN, even after recovery,
has been reported (Frank et al 2004).
The SPSS statistical software package (SPSS, Chicago, Illinois)
was used for all statistical analyses. Owing to the relatively small
sample size, between-group comparisons were made with non-
parametric Mann-Whitney U Two-Independent-Samples Tests,
calculating two-sided exact significance levels. Correlations were
examined with Spearman correlation coefficients. All values are
expressed as mean ? SD. As level of significance, a p value of p
? .05 was selected.
Subject groups were of similar age (CW 27 ? 6 years, REC AN
24 ? 5 years, p ? .2) and body mass index (CW 23 ? 2, REC AN
22 ? 3, p ? .3). The mean age of onset of AN was 15 ? 2 years,
and REC AN subjects had been recovered for between 12 and 66
months. Compared with CW, REC AN scored higher on EDI-2
drive for thinness-worst ever (REC AN 19.7 ? 2, CW .3 ? 1, p ?
.001) and worst ever Perfectionism (REC AN 108 ? 13, CW 53 ?
11, p ? .001). Current measures of Harm Avoidance (REC AN 13
? 9, CW 9 ? 5, p ? .5), Novelty Seeking (REC AN 22 ? 6, CW
20 ? 5, p ? .4), and Reward Dependence (REC AN 18 ? 3, CW
19 ? 3, p ? .5) were similar between groups.
Eight REC AN and eight CW had arterial blood sampling
during PET scanning. This method showed increased antero-
ventral striatal [11C]raclopride binding in REC AN (2.3 ? .4)
compared with CW (1.91 ? .3) (p ? .028). Other ROIs, including
the cerebellum, did not show significant differences between
groups. There was a trend, however, toward higher [11C]raclo-
pride binding for REC AN compared with CW in the ventral
putamen (CW 2.79 ? .2, REC AN 2.98 ? .3; p ? .1) and the
middle caudate (CW 2.33 ? .11, REC AN 2.56 ? .2; p ? .05).
The reference tissue model was then applied to the full study
sample. The 10 REC AN studied had significantly higher [11C]ra-
clopride binding potential (BP) (Figure 1) than the 12 CW for the
antero-ventral striatum (CW 1.98 ? .4, REC AN 2.33 ? .3, p ?
Figure 1. Increased [11C]raclopride binding potential (BP) in the antero-
ventral striatum in recovered anorexic women (REC AN) compared with
control women (CW).
G.K. Frank et al
BIOL PSYCHIATRY 2005;58:908–912 909
.036). This suggested a robust finding, and data are presented for
this larger sample size. There were no significant group differ-
ences in [11C]raclopride BP for the ventral putamen (CW 2.79 ?
.3, REC AN 2.97 ? .3, p ? .3), dorsal putamen (CW 2.97 ? .4, REC
AN 2.78 ? .2, p ? .08), middle caudate (CW 2.30 ? .3, REC AN
2.47 ? .2, p ? .1), and dorsal caudate (CW 2.47 ? .3, RAN AN
2.26 ? .3, p ? .09). Figure 2 shows increased [11C]raclopride
binding in an REC AN subject compared with an age matched
The injected dose of [11C]raclopride was similar between
groups (p ? .6), and repeated measures analysis in subjects with
arterial blood sampling showed similar metabolites in plasma (p
Neither age, body mass index, nor length of recovery were
correlated with [11C]raclopride binding. Antero-ventral striatal
values for REC AN-R tended to be higher compared with REC
AN-BP type, but this was statistically not significant, although this
is a relatively small sample size with limited power. A lifetime
diagnosis of major depression (n ? 6) or OCD (n ? 5) was not
related to [11C]raclopride BP. For REC AN, Harm Avoidance was
significantly and positively associated with [11C]raclopride bind-
ing in the dorsal caudate (? ? .9, p ? .0004, p ? .008 Bonferroni
corrected, Figure 3) and dorsal putamen (? ? .8, p ? .004), but
not in the antero-ventral striatum (rho ? ?.3, p ? .3).
This is the first study to investigate D2/D3 receptors in AN in
vivo with PET. A mixed group of REC AN-R and REC AN-BP
women had increased [11C]raclopride BP in the antero-ventral
striatum. Because PET measures of [11C]raclopride binding are
sensitive to endogenous DA concentrations (Drevets et al 2001),
this difference could be accounted for either by a reduction in the
intrasynaptic DA concentration or by an elevation of the density
and/or affinity of the D2/D3 receptors in this region.
People with AN are motorically restless and driven to over-
exercise in a manner resembling stereotypy. In animals, drugs
that stimulate DA receptors induce hyperactivity and stereotypy,
and Parkinson’s disease, a degeneration of DA pathways, is
associated with reduced motor activity (Maruya et al 2003). A
previous study from our group (Kaye et al 1999) found reduced
CSF HVA in REC AN-R. That finding lead to the hypothesis of
possibly increased postsynaptic DA receptor binding in AN in a
negative feedback fashion (Cooper et al 2003). For example, a
recent [11C]raclopride imaging study showed reduced plasma
HVA and increased D2 binding in response to catecholamine
depletion (Verhoeff et al 2003). Moreover, studies in rodents find
inverse relationships between DA metabolism and D2 receptor
binding (Galkina and Podgornaya 1996). Thus, reduced CSF
HVA might be consistent with the results described herein.
Whether AN have a compensatory up-regulation of the D2/D3
receptors (Cooper et al 2003) is not known. Motor hyperactivity
in AN, however, might be consistent with increased DA neuro-
Another psychiatric disorder, attention-deficit/hyperactivity
disorder (ADHD), has been associated with increased D2/D3
receptor binding in a small group of ADHD youth that was
related to reduced brain blood perfusion during birth (Lou et al
2004). A more recent study did not find D2/D3 receptor binding
differences in adolescents compared with young adults (Jucaite
et al 2005), although striatal mean receptor binding was higher in
the ADHD group. Attention-deficit/hyperactivity disorder is not
associated with a primary disturbance of eating behavior, al-
though individuals with ADHD often respond to stimulants such
as methylphenidate, which is relatively selective for DA neuro-
transmission (Kuczenski and Segal 1997), and loss of appetite
might occur as a side effect of this medication.
It is of potential relevance that decreased D2/D3 receptor
binding has been found in obese subjects (Wang et al 2004). That
finding supports the possibility that D2/D3 receptor binding
might be inversely related to weight and eating, with AN on one
end, and obesity on the other end of the spectrum. Thus,
increased D2/D3 receptor availability in AN might contribute to
a drive to become emaciated. It is worth noting that food
restriction has been shown to increase D2/D3 receptor activity in
rats (Carr et al 2003). Although REC AN subjects continue to have
some aberrant meal patterns and tend to still have desires to
restrict their food, we do not think food restriction accounts for
our findings in AN, because REC AN subjects were at normal
Figure 2. Comparison of one recovered anorexic subject (REC AN) and an
age-matched control woman (CW). MR, magnetic resonance; BP, binding
potential; PET, positron emission tomography (image); ROI, region of inter-
est; CER, cerebellum.
Score and Dorsal Caudate [11C]raclopride binding potential (BP) in recov-
ered anorexic subjects; p value Bonferroni corrected.
910 BIOL PSYCHIATRY 2005;58:908–912
G.K. Frank et al
weight and had normal blood glucose and ketone bodies (data
The finding of increased [11C]raclopride BP in the antero-
ventral striatum, a region that has been associated with the
modulation of reward, reinforcement, and addiction, is intrigu-
ing. Neuroimaging studies suggested a role for DA in the
pathophysiology of substance abuse and associated craving. For
example, some studies show that individuals with drug addic-
tions have reduced D2/D3 receptor binding (Wang et al 2004).
Some relationship between addiction and eating disorder has
been postulated (Marrazzi et al 1990). In fact, individuals with
bulimia nervosa have high rates of substance abuse, whereas
AN-R have low rates of substance and alcohol disorders (Bulik et
al 2004; Lilenfeld et al 1998). Whether increased D2/D3 receptor
availability protects individuals from substance abuse is highly
speculative, but this should be investigated in future studies.
Dopamine activity might not directly produce reward, but
rather, seems to modulate neuronal activity to “achieve an
optimal stimulus response” (Cooper et al 2003) and might help
organisms with making appropriate choices (Redgrave et al
1999). Volkow et al (2004) has postulated that decreased DA
function in addicted subjects results in decreased sensitivity to
non-drug stimuli, including natural reinforcers. We hypothesize
that the D2/D3 receptor system is overactive in AN, and thus,
might not respond appropriately to salient stimuli. Individuals
with AN have long been noted to be anhedonic and ascetic, able
to sustain self-denial of food as well as most comforts and
pleasures in life. Moreover, ill AN often show ego-syntonic denial
and resistance and ignore feedback about their precarious state
of health, and it is possible that overactive D2/D3 receptors in the
antero-ventral striatum might make anorexic subjects vulnerable
to developing the ill state.
No significant relationship between antero-ventral D2/D3
receptor binding and behavior was found in either group. For the
REC AN subjects, however, dorsal caudate [11C]raclopride bind-
ing was significantly related to Harm Avoidance after correction
for multiple comparisons. Although there were not significant
differences in Harm Avoidance scores between groups in this
small sample, a considerable literature suggests that Harm Avoid-
ance is elevated in individuals who are ill and recovered from AN
(Kaye et al 2004; Wagner et al, unpublished data) and is thus a
trait-related behavior. Harm Avoidance has been found to cor-
relate with caudate DA activity in Parkinson’s disease patients
(Kaasinen et al 2001). In addition, striatal DA activity has been
linked to trait anxiety in healthy subjects (Laakso et al 2003).
In terms of limitations, this pilot study combined restricting-
and binge-purge type REC AN to increase sample size. We had
found that reduced CSF HVA only occurred in REC AN-R in the
past (Kaye et al 1999). In this small sample, the subtypes were
statistically not different, although [11C]raclopride BP for the
restricting type subjects tended to be higher than for the binge-
purging subjects. This finding, however, could be simply a
question of power. Future studies with larger sample sizes will be
needed to test the hypothesis that increased D2/D3 BP occurs on
a continuum, with REC AN-R having higher values than REC
AN-BP, and REC AN-BP having higher values than CW. Thus, it
remains to be determined whether there are subgroup differ-
ences. In addition, this study cannot address whether the in-
creased receptor binding is due to increased density of the
D2/D3 receptors or whether there is increased binding caused by
reduced intrasynaptic DA and reduced competition with the
radioligand. Future studies are necessary to address this issue.
Also, this DA finding in REC AN might suggest that this is a
trait-related disturbance; it cannot be excluded, however, that
this could also be a persistent scar from the illness, because
altered DA activity during critical developmental periods might
have persistent effects on DA receptor interactions (Johnson and
Bruno 1990). Still, this finding could help explain, in part,
treatment resistance in AN.
In conclusion, this study suggests that increased D2/D3
receptor binding occurs in REC AN subjects. This finding might
help explain why individuals with AN are able to lose weight,
resist eating, overexercise, are protected from substance abuse,
and are insensitive to normal rewards.
This work was supported by National Institute of Mental Health
grants MH46001, MH42984, K05-MD01894, T32-MH18399 and
the Price Foundation. We thank Eva Gerardi and Katherine
Plotnicov for editorial assistance with manuscript preparation.
We are indebted to the participating subjects for their contribu-
tion of time and effort in support of this study.
Alexander GE, Crutcher MD, DeLong MR (1990): Basal ganglia-thalamocor-
tical circuits: Parallel substrates for motor, oculomotor, “prefrontal” and
Anderluh MB, Tchanturia K, Rabe-Hesketh S, Treasure J (2003): Childhood
orders: Defining a broader eating disorder phenotype. Am J Psychiatry
Bergen A, Yeager M, Welch R, Haque K, Ganjei K, van den Bree M, et al (in
press): Association of multiple DRD2 polymorphisms with anorexia ner-
Brewerton TD (2004): 9th annual meeting of the Eating Disorders Research
Bulik CM, Sullivan PF, Fear JL, PR. J (1997): Eating disorders and ante-
cedent anxiety disorders: A controlled study. Acta Psychiatr Scand
Alcohol use disorder comorbidity in eating disorders: A multicenter
Carr KD, Tsimberg Y, Berman Y, Yamamoto N (2003): Evidence of increased
dopamine receptor signaling in food-restricted rats. Neuroscience 119:
Cassano GB, Miniati M, Pini S, Rotondo A, Banti S, Borri C, et al (2003):
Six-month open trial of haloperidol as an adjunctive treatment for an-
orexia nervosa: A preliminary report. Int J Eat Disord 33:172–177.
Cloninger C, Przybeck T, Svarkic D, Wetzel R (1994): The Temperament and
Character Inventory (TCI): A Guide to Its Development and Use. St. Louis,
Missouri: Center for Psychobiology of Personality, Washington Univer-
Cooper J, Bloom F, Roth R (2003): The Biochemical Basis of Neuropharmacol-
ogy, 8th ed. Oxford, United Kingdom: Oxford University Press.
Amphetamine-induced dopamine release in human ventral striatum
correlates with euphoria. Biol Psychiatry 49:81–96.
Frank GK, Bailer UF, Henry S, Wagner A, Kaye WH (2004): Neuroimaging
studies in eating disorders. CNS Spectr 9:539–548.
Frank GK, Kaye WH, Meltzer CC, Price JC, Greer P, McConaha C, et al (2002):
Frost R, Marten P, Lahart C, Rosenblate R (1990): The dimensions of perfec-
Galkina OV, Podgornaya EK (1996): Regional brain patterns of dopamine,
metabolites and D2 receptors in memory. Pharmacol Biochem Behav
G.K. Frank et al
BIOL PSYCHIATRY 2005;58:908–912 911
Garner D (1991): Eating Disorder Inventory-2 Professional Manual. Odessa, Download full-text
Florida: Psychological Assessment Resources.
Halford JC, Cooper GD, Dovey TM (2004): The pharmacology of human
Johnson BJ, Bruno JP (1990): D1 and D2 receptor contributions to ingestive
and locomotor behavior are altered after dopamine depletions in neo-
natal rats. Neurosci Lett 118:120–123.
dopamine transporter binding in male adolescents with attention-defi-
ers and motor hyperactivity. Biol Psychiatry 57:229–238.
Kaasinen V, Nurmi E, Bergman J, Eskola O, Solin O, Sonninen P, et al (2001):
Personality traits and brain dopaminergic function in Parkinson’s dis-
In: Nestler EJ, Charney DS, editors. Neurobiology of Mental Illness. New
York: Oxford Press, 1112–1128.
Kaye WH, Frank GK, McConaha C (1999): Altered dopamine activity after
recovery from restricting-type anorexia nervosa. Neuropsychopharma-
Kuczenski R, Segal DS (1997): Effects of methylphenidate on extracellular
dopamine, serotonin, and norepinephrine: Comparison with amphet-
amine. J Neurochem 68:2032–2037.
Laakso A, Wallius E, Kajander J, Bergman J, Eskola O, Solin O, et al (2003):
Personality traits and striatal dopamine synthesis capacity in healthy
receptor studies. Neuroimage 4:153–158.
(1998): A controlled family study of anorexia nervosa and bulimia ner-
vosa: Psychiatric disorders in first-degree relatives and effects of pro-
Logan J, Fowler JS, Volkow ND, Ding YS, Wang GJ, Alexoff DL (2001): A
strategy for removing the bias in the graphical analysis method. J Cereb
Lou HC, Rosa P, Pryds O, Karrebaek H, Lunding J, Cumming P, et al (2004):
al (1990): Atypical endogenous opioid systems in mice in relation to
an auto-addiction opioid model of anorexia nervosa. Life Sci 47:
Maruya H, Watanabe Y, Okita M, Lawlor GF, Utsumi H, Niitsuma T (2003):
of dopamine and hyperactivity induced by Bay K 8644 in rats. Neuro-
Meltzer CC, Kinahan PE, Greer PJ, Nichols TE, Comtat C, Cantwell MN, et al
(1999): Comparative evaluation of MR-based partial-volume correction
schemes for PET. J Nucl Med 40:2053–2065.
Redgrave P, Prescott TJ, Gurney K (1999): The basal ganglia: A vertebrate
solution to the selection problem? Neuroscience 89:1009–1023.
Srinivasagam NM, Kaye WH, Plotnicov KH, Greeno C, Weltzin TE, Rao R
(1995): Persistent perfectionism, symmetry, and exactness after long-
term recovery from anorexia nervosa. Am J Psychiatry 152:1630–1634.
al (2003): Effects of catecholamine depletion on D2 receptor binding,
mood, and attentiveness in humans: A replication study. Pharmacol
ment and addiction in humans: Results from imaging studies. Behav
Volkow ND, Fowler JS, Wang GJ, Swanson JM (2004): Dopamine in drug
cations. Mol Psychiatry 9:557–569.
Wang GJ, Volkow ND, Thanos PK, Fowler JS (2004): Similarity between obe-
sity and drug addiction as assessed by neurofunctional imaging: A con-
912 BIOL PSYCHIATRY 2005;58:908–912
G.K. Frank et al