The authors have declared that no competing interests exist.
Conceived and designed the experiments: JSC YCS DHK JSK. Performed the experiments: WHJ JHJ CHC SWC. Analyzed the data: JSC WHJ CHC. Contributed reagents/materials/analysis tools: JYL JYH. Wrote the paper: JSC JSK.
Pathological gambling (PG) and obsessive-compulsive disorder (OCD) are conceptualized as a behavioral addiction, with a dependency on repetitive gambling behavior and rewarding effects following compulsive behavior, respectively. However, no neuroimaging studies to date have examined reward circuitry during the anticipation phase of reward in PG compared with in OCD while considering repetitive gambling and compulsion as addictive behaviors.
To elucidate the neural activities specific to the anticipation phase of reward, we performed event-related functional magnetic resonance imaging (fMRI) in young adults with PG and compared them with those in patients with OCD and healthy controls. Fifteen male patients with PG, 13 patients with OCD, and 15 healthy controls, group-matched for age, gender, and IQ, participated in a monetary incentive delay task during fMRI scanning. Neural activation in the ventromedial caudate nucleus during anticipation of both gain and loss decreased in patients with PG compared with that in patients with OCD and healthy controls. Additionally, reduced activation in the anterior insula during anticipation of loss was observed in patients with PG compared with that in patients with OCD which was intermediate between that in OCD and healthy controls (healthy controls < PG < OCD), and a significant positive correlation between activity in the anterior insula and South Oaks Gambling Screen score was found in patients with PG.
Decreased neural activity in the ventromedial caudate nucleus during anticipation may be a specific neurobiological feature for the pathophysiology of PG, distinguishing it from OCD and healthy controls. Correlation of anterior insular activity during loss anticipation with PG symptoms suggests that patients with PG fit the features of OCD associated with harm avoidance as PG symptoms deteriorate. Our findings have identified functional disparities and similarities between patients with PG and OCD related to the neural responses associated with reward anticipation.
Pathological gambling (PG) is a chronic disorder that occurs primarily in men and is characterized by a persistent pattern of continued gambling behavior despite adverse consequences. PG is considered a behavioral addiction
Addictive behaviors in PG and OCD are associated with dysfunctional process of the reward system. The reward processing is dependent on ventral striatal-orbitofrontal circuitry in brain
Considering repetitive gambling and compulsion as addictive behaviors, investigation of neurobiological features in PG and OCD may be helpful for understanding addictive behaviors. To date, however, no neuroimaging studies have investigated reward circuitry in PG compared with in OCD. Brain activity during the anticipation of reward may be important as it might influence subsequent choices and behaviors. Therefore, we applied the monetary incentive delay task
We included only male subjects in this study based on the findings showing predominant male prevalence in PG and sex differences in the neural response to reward stimuli
This study was conducted according to the principles expressed in the Declaration of Helsinki. The institutional review boards of the Kangbuk Samsung Medical Center and Seoul National University Hospital approved the study protocol, and all subjects signed an informed consent form prior to participation.
Fifteen male patients with PG (age, 27.93±3.59 years), 15 age-and IQ-matched healthy male controls (age, 26.60±4.29 years), and 13 male patients with OCD (age, 24.92±6.92 years) were enrolled. Previously, our group reported fMRI study on neural activities in reward system of OCD
Healthy controls were recruited from the local community and had no history of any psychiatric disorder. Exclusion criteria for all subjects were a history of significant head injury, alcohol or substance abuse, seizure disorder, and psychotic disorder.
All participants were medication naïve at the time of assessment. The Korean version of the Wechsler Adult Intelligence Scale was administered to all subjects to estimate IQ.
We used the monetary incentive task developed by Knutson et al.
Cues indicating potential gain, potential loss, and no monetary outcome are denoted by red, blue, and yellow circles, respectively. The amount of money at stake in each trial was 1000 Korean Won (approximately 1 US dollar). Three cues were pseudo-randomly presented with a fixation crosshair, separated by a jittered inter-trial interval of 5.17–9.85 s. Before MRI scanning, all participants completed an unpaid short practice session for each cue to minimize learning effects and provide an estimate of each individual's reaction time (RT) to standardize task difficulty. Mean RT during the practice session was used as duration of the target exposure in the first trial during MRI scanning. The target durations decreased by 30 ms on the next trial after a correct response and increased by 40 ms on the next trial after an incorrect response. As result, success rates in PG, OCD and healthy control were 57.27%, 54.48%, and 58.08%, respectively (
Variables | Control (N = 15) | PG (N = 15) | OCD (N = 13) | F | p |
|
|||||
Age (years) | 26.60 (4.29) | 27.93 (3.59) | 24.92 (6.92) | 1.25 | 0.30 |
Education (years) | 14.27 (1.39) | 14.80 (1.70) | 14.38 (3.07) | 0.26 | 077 |
IQ score | 114.47 (7.10) | 113.67 (9.96) | 108.54 (14.59) | 1.21 | 0.31 |
|
|||||
Age of Onset (years) | 25.67 (3.92) | 16.23 (5.66) | 26.89 | <0.01 |
|
Duration of illness (years) | 2.20 (1.29) | 8.69 (5.68) | 18.62 | <0.01 |
|
YBOCS |
|||||
|
16.13 (7.28) | 19.54 (5.94) | 1.80 | 0.19 | |
|
8.40 (3.46) | 10.31 (4.68) | 1.53 | 0.23 | |
|
7.73 (4.01) | 9.23 (4.55) | 0.86 | 0.36 | |
SOGS | 15.90 (1.73) | ||||
BDI | 3.00 (4.81) | 15.20 (12.32) | 15.08 (10.42) | 7.68 | <0.01 |
BAI | 3.67 (3.85) | 11.00 (13.40) | 14.08 (16.09) | 2.80 | 0.07 |
Smoking (cigarettes a day) | 9.33 (9.61) | 8.67 (9.90) | 8.46 (9.87) | 0.03 | 0.97 |
|
|||||
Total earnings (in Korean Won) | 28000.00 (2360.39) | 27266.67 (2120.20) | 28923.08 (3226.49) | 1.44 | 0.25 |
Reaction times (ms overall) | 222.51 (19.40) | 231.33 (26.26) | 215.48 (23.60) | 1.64 | 0.21 |
Hit rate (% overall) | 58.08 (1.36) | 57.27 (1.88) | 58.48 (1.71) | 1.93 | 0.16 |
Data are given as mean (SD).
PG, Pathological Gambling; OCD, Obsessive-compulsive disorder; IQ, Intelligence Quotient; BDI, Beck Depression Inventory; BAI, Beck Anxiety Inventory; YBOCS, Yale-Brown Obsessive Compulsive Scale; SOGS, South Oaks Gambling Screen.
One-way analysis of variance (ANOVA) was used.
YBOCS was administered for patients with OCD and PG-YBOCS was used for patients with PG.
MR scanning was performed on a 1.5 Tesla Scanner (Siemens, AVANTO, Munich, Germany). Twenty-five axial slices were obtained using a gradient-echo echo planar imaging (EPI) sequence (TE/TR, 52/2340 ms; FOV, 220×220 mm2; FA, 90°, 3.44×3.44×5 mm3, interleaved, no interslice gap); we collected an interleaved slice-acquisition pattern, approximately parallel to the anterior-posterior commissure plane to avoid signal contamination from adjacent slices. Functional images were obtained in two runs of 538.2 s each, resulting in 245 volumes per run for each subject. High resolution T1-weighted structural MRI images (MPRAGE, TR, 11.6 s; TE, 4.76 ms; FOV, 230 mm; FA, 15°; 208 slices, 0.45×0.45×0.9 mm3) were also acquired for each subject for anatomical reference. Stimuli were back-projected onto a translucent screen located at the subject's feet using an LCD projector and viewed through a periscope mirror attached to the head coil. The subjects responded to the target by pressing an MRI-compatible mouse button with the index finger of the right hand.
Data were analyzed using Statistical Parametric Mapping software (SPM8, Wellcome Department of Imaging Neuroscience, London, UK). The first three volumes of EPI images were discarded due to magnetization instability. The EPI images were corrected for differences in acquisition time, realigned, spatially normalized to the Montreal Neurology Institute reference brain (voxel size 3×3×3 mm3), smoothed using an isotropic 4-mm full-width half-maximum Gaussian kernel, and high-pass filtered with a 128-s cut-off. After images were preprocessed, the first-level analysis was performed on each subject by modeling the different conditions (cues indicating potential gain, potential loss, and no outcome) and convolved with a canonical hemodynamic response function according to the general linear model. Movement parameters were also included in the statistical model. The relevant contrasts were as follows: gain cue vs. no-outcome cue to assess the main effect of gain anticipation, and loss cue vs. no-outcome cue to assess the main effect of loss anticipation. A within-group analysis and a between-group analysis were performed for the contrast images for gain and loss anticipation. The results of the between-group activation are presented in
Regions | L/R | MNI coordinates | T/Z-value | voxels | ||
x | y | Z | ||||
|
||||||
Thalamus | R | 18 | −24 | 0 | 4.41/3.96 | 24 |
Fusiform | R | 39 | −72 | −18 | 4.56/4.07 | 16 |
Middle temporal gyrus | R | 57 | −51 | −12 | 3.59/3.32 | 8 |
Inferior temporal gyrus | R | 60 | −60 | −6 | 3.97/3.62 | 21 |
Middle occipital gyrus | R | 39 | −81 | 12 | 3.94/3/60 | 39 |
Lingual gyrus | L | −15 | −78 | −12 | 4.01/3.65 | 18 |
L | −21 | −60 | −9 | 3.80/3.49 | 7 | |
Cerebellum | L | −9 | −33 | −12 | 3.90/3.57 | 16 |
|
||||||
none | ||||||
|
||||||
none | ||||||
|
||||||
none | ||||||
|
||||||
Ventromedial caudate nucleus | 0 | 6 | −6 | 4.81/4.25 | 11 | |
Caudate | L | −12 | 27 | 15 | 4.73/4.19 | 27 |
R | 12 | 24 | 9 | 3.98/3.63 | 18 | |
Thalamus | R | 6 | −24 | 0 | 3.98/3.63 | 14 |
Posterior insula | L | −33 | −36 | 24 | 5.03/4.40 | 8 |
Middle frontal gyrus | L | −21 | 24 | 24 | 5.67/4.83 | 8 |
Superior occipital gyrus | R | 18 | −90 | 24 | 4.50/4.02 | 34 |
Lingual gyrus | L | −21 | −60 | −6 | 4.71/4.18 | 18 |
R | 21 | −72 | 0 | 4.55/4.06 | 21 | |
|
||||||
none |
All results
Regions | L/R | MNI coordinates | T/Z-value | voxels | ||
x | y | z | ||||
|
||||||
Ventomedial caudate nucleus | 0 | 3 | 0 | 4.33/3.90 | 8 | |
Temporal pole | L | −39 | 3 | −24 | 4.32/3.89 | 7 |
|
||||||
Rolandic operculum | L | −42 | −6 | 15 | 4.30/3.88 | 16 |
|
||||||
none | ||||||
|
||||||
Anterior insula | L | −18 | 30 | 9 | 4.26/3.85 | 10 |
Caudate | L | −15 | 15 | 24 | 4.71/4.17 | 11 |
Putamen | R | 30 | −6 | 21 | 4.82/4.25 | 8 |
|
||||||
Ventromedial caudate nucleus | 0 | 3 | 0 | 6.24/5.18 | 25 | |
Caudate | R | 12 | 15 | 18 | 4.80/4.24 | 9 |
|
||||||
Inferior parietal gyrus | L | −54 | −30 | 39 | 4.11/3.73 | 15 |
Superior frontal gyrus | L | −18 | 12 | 69 | 3.71/3.42 | 7 |
All results
In order to test our main hypotheses, we selected regions of interest (ROIs, 5-mm radius spheres) in the ventromedial caudate nucleus (part of the ventral striatum) and the anterior insula and calculated individual mean contrast values for each ROI using Marsbar software (
Comparisons of demographic and clinical variables among patients with PG, patients with OCD, and healthy controls were conducted using an ANOVA. Statistical analysis was two-tailed, and significance was set at
No significant differences in age, education, IQ or BAI score were observed among the three groups (
With regard to behavioral data, no significant differences were observed among the three groups for the mean hit rate (
During anticipation of potential gain, all groups showed significant activation in the ventral striatum, frontal cortex and occipital cortex. The extent of activation in the ventral striatum of patients with PG was less than those in healthy controls and patients with OCD. During anticipation of loss, no significant activation was observed in the ventral striatum in the PG group.
In the between-group analysis, patients with PG showed less activation in the ventromedial part of the caudate nucleus as well as middle frontal gyrus, and occipital cortex than did patients with OCD during anticipation of gain. Patients with PG showed reduced activation in the thalamus relative to patients with OCD and healthy controls. No statistical differences in the BOLD signal were observed between patients with OCD and healthy controls during anticipation of gain (
PG: Pathological Gambling; OCD, Obsessive-compulsive disorder; L, left; R, right.
Significant differences in brain activations between groups are marked (asterisk).
During anticipation of loss, patients with PG showed significantly less activation in the ventromedial caudate nucleus relative to that in patients with OCD and healthy controls. In contrast, increased activation in the anterior insula as well as in the putamen and caudate nucleus was found in patients with OCD relative to healthy controls (
We found significant positive correlations between activity in the anterior insula and SOGS score in patients with PG during the anticipation of loss (r = 0.64,
To our knowledge, this is the first event-related fMRI study to elucidate neural changes associated with reward anticipation in patients with PG distinguished from those with OCD and healthy controls. We found that activation in the ventromedial caudate nucleus during anticipation of both gain and loss decreased in patients with PG compared with that in patients with OCD and healthy controls. Additionally, reduced activation in the anterior insula during anticipation of loss was observed in patients with PG compared with that in patients with OCD, and a significant positive correlation between activity in the anterior insula and SOGS score was found in patients with PG.
As we expected, we found reduced activation of the ventromedial caudate nucleus during both anticipation of gain and anticipation of loss in patients with PG compared with that in healthy controls. The ventromedial caudate nucleus has been considered a part of the ventral striatum
There was an evidence for involvement of the ventral striatum in PG. Patients with PG showed less activation of the ventral striatum during simulated gambling
Contrary to our expectations, neural activity in the ventral striatum of PG was less than that in OCD, and patients with OCD and healthy controls showed comparable activation of the ventral striatum during anticipation of gain. It could be speculated that the OCD group in our study was heterogeneous in terms of symptom dimension. Figee et al.
Decreased activation of the anterior insula during anticipation of loss was observed in patients with PG relative to that in patients with OCD. The anterior insular activity in patients with PG was intermediate between that in patients with OCD and healthy controls, although anterior insular activity in patients with PG was not significantly different from that in healthy controls. The insula is an important region for emotion processing and its activation is associated with anticipation of adverse events
In the present study, a reduction in the thalamic activation was also found in patients with PG during anticipation of gain compared with those with OCD and healthy controls. The thalamus acts as an intermediary between the ventral striatum and medial prefrontal cortex
We found no significant correlations of neural responses in the ventral striatum with clinical variables including symptom severity, duration of illness, and age of onset. Furthermore, patients with PG were drug naïve and were in the early phase after disease onset, i.e., within 5 years after onset: therefore, reduced neural activities in the ventral striatum during anticipation of reward could be trait markers for the pathophysiology of PG, distinguishing it from OCD and healthy controls.
Our study had some limitations. The small sample size and the inclusion of only male participants may limit the generalizability of the results. This limitation was inevitable because our goal was to recruit a homogeneous sample to control for confounding factors such as medication and gender effects. Thus, we recruited patients with PG who had an illness duration of <5 years and were young adults in order to elicit earlier brain changes in the reward system of PG. Furthermore, we selected drug naïve patients with PG.
In summary, our results showed that neural activities in the ventromedial caudate nucleus during anticipation of both gain and loss decreased in patients with PG relative to those in patients with OCD and healthy controls, suggesting that this neural activity may be a trait marker for the pathophysiology of PG. During the anticipation of loss, anterior insular activation in patients with PG was intermediate between that in patients with OCD and healthy controls and was significantly correlated with PG symptom severity, suggesting that patients with PG fit the features of OCD as symptoms deteriorate. Our findings have identified functional disparities and similarities between patients with PG and OCD related to the neural responses associated with reward anticipation.