Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 May 1.
Published in final edited form as: J Psychiatr Res. 2023 Mar 3;161:54–61. doi: 10.1016/j.jpsychires.2023.03.004

Aberrant Reward-Related Neural Activation: Dimensional Correlate of Binge-Eating Severity or Categorical Marker of Binge Eating?

Kelsey Hagan a,*, Nandini Datta b, May Stern b, Cara Bohon b,c
PMCID: PMC10149604  NIHMSID: NIHMS1881333  PMID: 36898327

Abstract

Binge eating (BE) is characterized by consuming an objectively large amount of food over a short period of time and experiencing loss of control over one’s eating. The neural underpinnings of monetary reward anticipation and their association with BE severity remain poorly understood. Fifty-nine women aged 18 to 35 (M = 25.67, SD = 5.11) with a range of average weekly BE frequency (M = 1.96, SD = 1.89, range = 0 – 7) completed the Monetary Incentive Delay Task during fMRI scanning. Mean percent signal change within the left and right nucleus accumbens (NAc) during anticipation of monetary gain (versus non-gain) was extracted from a priori-defined functional 5mm spheres and correlated with average weekly BE frequency. Exploratory voxel-wise whole-brain analyses examined the association between neural activation during anticipation of monetary reward and average weekly BE frequency. Body mass index and depression severity were covariates of non-interest in analyses. Mean percent signal change in the left and right NAc inversely correlated with average weekly BE frequency. Whole-brain analysis revealed no significant associations between neural activation during reward anticipation and average weekly BE frequency. In exploratory case-control analyses, mean percent signal change in the right NAc was significantly lower in women with BE (n = 41) versus women without BE (n = 18), but whole-brain analyses revealed no significant group differences in neural activation during reward anticipation. Decreased right NAc activity during monetary reward anticipation may distinguish women with and without BE.

Keywords: monetary incentive delay task, reward anticipation, bulimia nervosa, binge-eating disorder, binge eating, fMRI

Introduction

Binge eating (BE) is characterized by consuming an objectively large amount of food in a short time and experiencing loss of control over eating (American Psychiatric Association, 2013). BE is a diagnostic criterion of both bulimia nervosa and binge-eating disorder, which affect 1–3% of the population (Stice et al., 2013), and is associated with psychosocial impairment and economic burden (Mitchison et al., 2017; Tannous et al., 2021). First-line treatments for binge-type eating disorders posit that caloric restriction (Fairburn, 2008) or interpersonal dysfunction (Agras et al., 2000) are core mechanisms of BE and target these issues. However, more than 50% of patients with binge-type eating disorders remain symptomatic or relapse after receiving first-line treatments (Linardon, 2018; Linardon and Wade, 2018). This disappointing statistic suggests that additional mechanisms of BE should be considered to guide adaptations to existing interventions or inform novel therapeutics for this impairing behavior.

Alterations in reward anticipation have been mechanistically implicated in other forms of impulsive or compulsive behavior. Specifically, lower ventral striatal activation during monetary reward anticipation inversely correlates with behavioral frequency (severity) in substance use and gambling (Luijten et al., 2017; Peters et al., 2011; Reuter et al., 2005), suggesting this attenuated response during monetary reward anticipation may be transdiagnostically implicated in the severity of behavioral dyscontrol, which could extend to BE. Although research has suggested lower self-reported reward anticipation in BE (compared to healthy volunteers) (Dolan et al., 2022), the association between ventral striatal response to monetary reward anticipation and BE severity remains unclear.

Hypotheses of Reward Anticipation and Binge Eating

To date, theoretical accounts of reward processing have posited that general reward hyposensitivity or hypersensitivity may drive an individual to engage in impulsive or compulsive behavior, such as BE (Blum et al., 2000; Dawe and Loxton, 2004). However, data suggest reward-processing alterations in persons with BE are not uniform across reward types (disease-relevant versus other rewards) (Leenaerts et al., 2022), necessitating nuanced models of reward processing in BE. One such model is the maladaptive scaling hypothesis (Zald and Treadway, 2017), which posits that reward-related neural activation depends upon a reward’s subjective value relative to other rewards, such that rewards are evaluated against an anchor reward that has been assigned the highest subjective value. In BE, binge foods could downscale the value of other rewards, such as money. High subjective value of binge foods may be constructed from previously learned associations between BE and outcomes (or eating expectancies), particularly from a history of positive and negative reinforcement from BE, and correspond with BE severity. In turn, this high subjective value of BE may decrease expectancies about other rewards, leading to decreased neural activation during anticipation of other rewards, such as money, that would be inversely associated with BE severity.

Empirical Research on Reward Circuit Activation During Monetary Reward Anticipation in Binge Eating

The monetary incentive delay task (MIDT) (Knutson et al., 2000) is a neuroimaging paradigm designed to test neural activation in response to incentive cues for money. Research has found that the ventral striatum (encompassing the nucleus accumbens, NAc) encodes the subjective value of rewards and signals expectations about reward receipt (Oldham et al., 2018; Wilson et al., 2018). To our knowledge, two studies have administered the MIDT to persons with BE. Simon et al. (2016) found no differences in neural activation during monetary reward anticipation between treatment-seeking women with binge-type eating disorders (n = 56) and healthy volunteer women (n = 55). In contrast, Balodis et al. (2013) found adults with binge-eating disorder and obesity (n = 19) showed lower bilateral ventral striatal activation during monetary reward anticipation relative to adults with obesity only (n = 19); however, neural activation did not differ between adults with binge-eating disorder and healthy-weight control participants (n = 19). In a follow-up study, persons with binge-eating disorder received four months of treatment with sibutramine and/or cognitive-behavioral therapy (Balodis et al., 2014). Persons with binge-eating disorder who continued to BE following treatment showed lower ventral striatal activation during monetary reward anticipation than those who ceased BE (Balodis et al., 2014), suggesting that neural response during monetary reward anticipation may be a prognostic indicator of BE persistence versus remission.

Bodell et al. (2018) administered a monetary reward-guessing task with anticipation and outcome phases to community-recruited 16-year-old girls with BE (n = 28) and without BE (n = 92) and found no group differences in neural activation during monetary reward anticipation. However, Bodell et al. (2018) found reward anticipation-related activation of the right caudate was positively associated with BE severity in the combined sample cross-sectionally and at two-year follow-up, but this association did not hold when accounting for baseline depressive symptoms and receipt of public assistance. Combined, findings on neural activation during monetary reward anticipation in persons with BE have been mixed. Still, they suggest a potential role of altered ventral striatal activation during reward anticipation in BE.

The current literature on monetary reward anticipation in persons with BE has mostly relied on case-control designs and categorical diagnoses. Limitations of categorical diagnoses include within-disorder heterogeneity and poor reliability. Researchers have increasingly attributed a lack of progress in identifying reliable biomarkers of psychopathology to these limitations (Latzman and DeYoung, 2020). In addition to traditional case-control designs, researchers have begun to conduct neuroimaging analyses dimensionally, as this approach affords insights into correlates of illness severity. For instance, dimensional analyses in samples with other forms of behavioral dyscontrol, including gambling, smoking, and alcohol use, have found lower ventral striatal activation during monetary reward anticipation associated with greater frequency (severity) of these behaviors (Peters et al., 2011; Reuter et al., 2005). These results highlight the potential utility of conducting dimensional analyses using BE frequency and reward-processing data.

The Present Study

Here, we tested the hypothesis that BE frequency would inversely scale with ventral striatal activation during monetary reward anticipation in a transdiagnostic sample of young women who endorsed, on average, zero to multiple BE episodes per week over the past three months. Our hypothesis was informed by research showing lower monetary reward anticipation-related activation within the ventral striatum is associated with increased severity in other forms of behavioral dyscontrol.

Material and Method

Participants

Participants were recruited from clinics in the Stanford University Department of Psychiatry and Behavioral Sciences and the community. Inclusion criteria were female sex, aged 18 to 35 years, and with or without current BE. Exclusion criteria were MRI contraindications; neurological disorder or insult; substance use disorder or psychosis; body mass index (BMI) < 18.5 kg/m2; pregnancy; and psychotropic medication use (except stable dose of selective serotonin reuptake inhibitors). Participants with BMI < 18.5 kg/m2 were excluded to reduce heterogeneity in the neural reward processing that may result from malnutrition and underweight status (Bernardoni et al., 2020). The Stanford University Institutional Review Board approved the study procedures (Protocol #35204). All participants provided written informed consent before completing the study procedures.

Procedures

Telephone screening.

Interested individuals completed a telephone screen with a trained bachelors-level research coordinator to assess inclusion and exclusion criteria. Individuals who met eligibility criteria were invited to participate, and study procedures were described.

Study session.

In-person study procedures consisted of two appointments. During the first appointment, participants provided sociodemographic information, completed diagnostic interviews and self-report measures of eating disorder and general psychiatric symptoms, and provided anthropometric measurements via a wall-mounted stadiometer and a calibrated digital scale. During the second appointment, participants provided menstrual cycle information and completed the MIDT during fMRI scanning.

Measures

A trained bachelors-level research coordinator administered interviews. Inter-rater reliability data were not collected.

Eating disorder symptoms.

The Eating Disorders Examination (EDE) version 16 (Fairburn, 2008) is a semi-structured interview assessing eating disorder cognitions (e.g., overvaluation of shape and weight, etc.), eating disorder behavior frequency (e.g., binge eating, purging, etc), and can be used to determine eating disorder diagnosis. The EDE cognitive subscales and behavioral frequency items have demonstrated evidence of validity and reliability (including inter-rater reliability) (Berg et al., 2015). We used the EDE to ascertain average weekly BE frequency and eating disorder diagnosis.

Other psychiatric symptoms.

The Mini International Neuropsychiatric Interview (MINI) (Lecrubier et al., 1997) was administered to assess psychiatric comorbidity. The MINI has demonstrated evidence for good psychometric properties (Lecrubier et al., 1997).

Participants completed the Beck Depression Inventory-II (BDI-II) (Beck et al., 1987), a dimensional assessment of depression severity over the past two weeks. Internal consistency of the total BDI-II score was excellent (α = 0.90).

MIDT.

Participants completed the MIDT during fMRI scanning to assess the impact of monetary incentive cues on neural activity (Knutson et al., 2000). Before completing the MIDT in-scanner, participants practiced the task outside of the scanner. Reaction times from the practice MIDT were used to individually calibrate task difficulty, such that each participant would “win” two-thirds of MIDT trials in-scanner. The MIDT used here consisted of a 10-minute run comprised of 80 trials. Participants could win or lose $0 or $5 in each trial.

In the fMRI environment, participants began the MIDT with $10.00. Each MIDT trial consisted of three phases: anticipatory, probe, and outcome. In the anticipatory phase, participants viewed a cue that signaled the coming reward: circle with two horizontal lines for gain (+$5), empty circle for non-gain (+$0), square with two horizontal lines for loss (−$5), or empty square for non-loss (−$0). The cue was presented for 250ms, followed by a jittered fixation target. Next, a target probe appeared, and participants were instructed to respond with a button press. Target probe presentation duration was individually calibrated (based on out-of-scanner practice MIDT reaction times). Participants received performance feedback in the outcome phase. Feedback included the amount of money earned in the trial and cumulative winnings over the task. See Figure 1 for a sample MIDT trial. Neural activation evoked during the MIDT has shown evidence of fair to excellent test-retest reliability (ICCs = .50 to .84) (Elliott et al., 2020).

Figure 1.

Figure 1

Open in a new tab

Schema of the Monetary Incentive Delay Task (MIDT)

Note. The target cue presentation accuracy is determined based on reaction times collected during the pre-scanner MIDT practice.

fMRI Analyses

Details on fMRI data acquisition, processing, and preprocessing – which was completed using FMRIPrep version 1.2.3 (Esteban et al., 2021) – are presented in the Supplementary Material. fMRI data were analyzed using FSL (FMRIB Software Library) version 6.0.0 (Jenkinson et al., 2012). Other statistical analyses were conducted in R version 3.6.3.

At the individual level of analysis, each MIDT event was convolved with a double-gamma hemodynamic response function (to model BOLD response) and entered into a general linear model (GLM) using FSL’s FEAT (FMRI Expert Analysis Tool; www.fmrib.ox.ac.ku/fsl) software. The temporal derivative of each MIDT event was also included in the GLM. Participant volumes with relative motion outliers greater than 1.5mm – indicated by DVARS and framewise-displacement measures – were dropped from analysis by including a single timepoint regressor in the analysis (Siegel et al., 2014). If participant motion consisted of more than relative motion outliers, participant data were not used; however, no participants were excluded for motion. Contrasts included: 1) anticipation of gain versus non-gain (reward anticipation); 2) anticipation of loss versus anticipation of non-loss (loss anticipation); 3) outcome of gain for “hits” versus “misses” (reward outcome); and 4) outcome of loss for “hits” versus “misses” (loss outcome). “Hits” were responses to the target cue during the probe phase, whereas “misses” were non-responses to the target cue during the probe phase.

Region-of-Interest (ROI) Analyses.

Mean percent signal change was extracted from functional 5mm spheres of the left and right NAc for the reward anticipation contrast. Functional NAc spheres were created using Montreal Neurological Institute (MNI) coordinates from Balodis et al. (2013), who used the MIDT in adults with BE: left NAc MNI: x = −12, y = 10, z = 2; right NAc MNI: x = 10, y = 8, z = 2. The NAc was selected because it is robustly associated with monetary reward anticipation (Oldham et al., 2018).

The association of mean percent signal change of the left and right NAc for the reward anticipation contrast and BE severity was examined using Spearman partial correlations, covarying for BMI and BDI-II total scores (depression severity). BMI and depression severity were selected as covariates due to their associations with reward processing (Dolan et al., 2022; Gill et al., 2021).

Visual inspection of scatterplots suggested that the correlation was driven by participants without BE. We repeated analyses, removing participants without BE, and found no association between mean percent signal change in the NAc for reward anticipation and BE severity. We conducted post hoc exploratory group analyses to compare mean percent signal change for reward anticipation within the left and right NAc between women with and without BE.

Exploratory Whole-Brain Analyses.

A voxel-wise, whole-brain, single-group analysis with BE severity as a covariate of interest was conducted to search neural activation unconstrained to ROIs and test whether neural response for all four contrasts correlated with BE severity. Whole-brain analyses were conducted using mixed-effects modeling via FSL’s FLAME 1 (FMRIB’s local analysis of mixed-effects) to account for between-person variability. A whole-brain corrected, family-wise error cluster significance threshold of p < .05 and Z > 3.1 was applied (Eklund et al., 2016). BMI and depression severity were covariates of non-interest.

A post hoc voxel-wise, whole-brain, group-level analysis was conducted to compare neural response for reward anticipation, loss anticipation, reward outcome, and loss outcome contrasts between women with and without BE, with BMI and depression severity as covariates of non-interest.

We did not have hypotheses about loss anticipation and outcome contrasts as our hypotheses centered on the reward anticipation contrast.

Results

Participants were 59 postmenarchal women: 41 with BE and 18 without BE (see Table 1). The sample size is based on participants whose fMRI data passed quality control and were included in MIDT analyses. Data from one participant (without BE) were excluded for ventriculomegaly.

Table 1.

Participant Sociodemographic and Clinical Characteristics

BE
(n = 41)		 non-BE
(n = 18)
Characteristic M (SD) or n M (SD) or n Statistical Comparison
Age 25.17 (5.25) 26.61 (3.96) t(57) = −1.04, p = .303
Race χ2 (5) = 3.24, p = .662
 American Indian or Alaskan Native 1 1
 Asian 14 5
 Black 0 1
 Native Hawaiian or Other Pacific Islander 1 0
 White 23 10
 Other 2 1
Ethnicity Fisher’s test p = 1.00
 Hispanic/Latine 0 0
 Not Hispanic/Latine 41 18
BMI 27.13 (8.64) 23.74 (5.33) t(57) = 1.53, p = .131
EDE variablesa
 Global score 2.69 (1.08) .40 (.35) t(57) = 8.79, p <.001
 Average weekly binge-eating frequency 2.63 (1.67) - -
 OBE frequency 31.34 (20.17) - -
 SBE frequency 0.37 (1.92) - -
 Self-induced vomiting frequency 2.88 (9.77) - -
 Laxative misuse frequency 0.51 (1.65) - -
 Diuretic misuse frequency - - -
 Diet pill misuse frequency 0.02 (0.16) - -
 Driven exercise frequency 16.44 (21.99) - -
Eating disorder diagnosis
 BN 18 -
 OSFED: subthreshold BN 3 -
 BED 15 -
 OSFED: subthreshold BED 5 -
BDI-II score 13.03 (8.91) 2.71 (6.65) t(55) = 4.29, p <.001
Psychiatric comorbidities
 Agoraphobia 6 (14.63%) -
 Dysthymia 2 (4.88%) -
 Generalized Anxiety Disorder 17 (41.46%) -
 Major Depressive Disorder 16 (39.02%) -
 Obsessive Compulsive Disorder 3 (7.32%) -
 PTSD 2 (4.88%) -
 Social Phobia 5 (12.20%) -
Psychotropic medication use 7 (17.07%) -
Menstrual cycle phaseb χ2 (3) = 7.24, p = .065
 Follicular phase (0–14 days since start of period) 12 (29.27%) 7 (38.89%)
 Luteal phase (>14 days since start of period) 7 (17.07%) 6 (33.33%)
 Amenorrhea (missed ≥ 3 periods)c 2 (4.88%) 0 (0%)
 No period due to hormonal contraceptive use 10 (24.39%) 0 (0%)
Hormonal contraceptive used χ2 (1) = .092, p = .761
 Yes 19 (46.34%) 8 (44.44%)
 No 20 (48.78%) 7 (38.89%)
Open in a new tab

Note. BE = binge eating; BED = binge eating disorder; BDI-II = Beck Depression Inventory; BMI = body mass index; BN = bulimia nervosa; EDE = Eating Disorders Examination; OBEs = past-three-month objective binge-eating episode frequency; OSFED = other specified feeding or eating disorder; PSTD = post-traumatic stress disorder; SBEs = past-three-month subjective binge eating episode frequency; SD = standard deviation.

a

Frequency of EDE behavioral variables is the number of times the behavior occurred during the three months before the assessment date.

b

Menstrual cycle data is missing for ten women with binge eating and five women without binge eating.

c

Amenorrhea was defined as the loss of the menstrual cycle for 3 months when a participant was not using a hormonal contraceptive.

d

Hormonal contraceptive use data were missing for two women with binge eating and three women without binge eating.

Behavioral Performance on the MIDT

Women with and without BE did not significantly differ on amount of money won, hit rate, or reaction time (see Table 2).

Table 2.

Summary of MIDT Behavioral Performance

Whole Sample BEa non-BEb
Variable M (SD) M (SD) M (SD) Statistical Comparisonc
Total monetary reward (in USD) $35.48 (8.64) $36.25 ($7.66) $34.41 ($9.82) t(55) = .76, p = .450
Hit rated (%) F(4, 47) = 1.08, p = .380
 Gain $0 trials 66.03 (13.60) 65.89 (14.53) 66.35 (11.45)
 Gain $5 trials 65.79 (9.06) 64.72 (8.31) 68.42 (10.53)
 Lose $0 trials 63.75 (14.65) 63.11 (12.60) 65.33 (19.22)
 Lose $5 trials 66.92 (13.40) 68.51 (9.19) 63.00 (20.33)
Reaction time (in ms) F(4,47) = .92, p = .459
 Gain $0 trials 177.40 (49.65) 178.06 (52.68) 175.79 (42.90)
 Gain $5 trials 165.44 (35.33) 162.89 (29.39) 171.72 (47.62)
 Lose $0 trials 168.96 (40.18) 169.37 (34.96) 167.93 (52.31)
 Lose $5 trials 167.93 (36.67) 172.62 (29.40) 155.31 (49.47)
Open in a new tab

Note. BE = binge eating; USD = United States dollars.

a

Behavioral data were missing for four women with binge eating.

b

Behavioral data were missing for four women without binge eating.

c

Comparing women with and without binge eating.

d

Hit rate is pressing the button during the target phase.

ROI Analysis

A priori dimensional analysis.

BE severity was significantly inversely correlated with mean percent signal change in the left and right NAc for reward anticipation, covarying for BMI and depression severity (see Figure 2). Results were similar without covariates). Visual inspection of scatterplots suggested significant effects were driven by participants with zero average weekly BE episodes. We repeated analyses, excluding participants with zero average weekly BE episodes, and found no association between BE severity and mean percent signal change for reward anticipation, covarying for BMI and depression severity, in the left (rs = −.120, p = .061) and right (rs = −.061, p = .150) NAc. Results were similar without covariates.

Figure 2.

Figure 2

Open in a new tab

Correlation of Reward Anticipation-Related Nucleus Accumbens Activation and Average Weekly Binge-Eating Frequency in the Whole Sample

Notes. Figure 2a shows the correlation between average weekly binge-eating episodes and left nucleus accumbens (NAc) activation for the reward anticipation contrast. Figure 2b shows the correlation between weekly binge-eating episodes and right NAc activation for the reward anticipation contrast. Figure 2c shows the functional 5mm NAc spheres in navy blue. MNI coordinates are: left NAc x= −12, y = 10, z = 2; and right NAc x = 10, y = 8, z = 2.

Post hoc group comparisons.

Mean percent signal change in the right NAc for reward anticipation was significantly lower in women with BE (M = −.095, SD = .597) compared to women without BE (M = .362, SD = .468), F(1,53) = 6.610, p = .013, η2 = .110 (medium to large effect; 95% CI = .004 – .262). However, left NAc activation during this contrast did not significantly differ between women with BE (M = .033, SD = .449) and women without BE (M = .305, SD = .317), F(1,53) = 2.638, p = .110, η2 = .046 (small to medium effect size; 95% CI = .0 – .180) (see Figure 3). These analyses covaried for BMI and depression severity; results were similar without covariates.

Figure 3.

Figure 3

Open in a new tab

Comparison of Mean Percent Signal Change During Anticipation of Monetary Reward (vs. No Monetary Reward) in the Nucleus Accumbens Between Women with and without Binge Eating

Notes. This figure is a bar chart of mean percent signal change for the reward anticipation in women with binge eating (teal color) compared to women without binge eating (red color) in the left and right NAc, using a priori-defined functional 5mm spheres. Each point represents one participant’s data, and the bars are standard error bars. BE = women with binge eating; non-BE = women without binge eating; NAc = nucleus accumbens.

Exploratory Whole-Brain Analysis

A priori dimensional analysis.

There were no significant associations between neural activation for reward anticipation, loss anticipation, reward outcome, and loss outcome contrasts and BE severity. Results were similar without covariates.

Post hoc group comparisons.

There were no significant group differences in neural activation for reward anticipation, loss anticipation, reward outcome, and loss outcome contrasts between women with and without BE. Results were similar without covariates.

Discussion

Current first-line treatments for BE posit that caloric restriction and interpersonal dysfunction are BE mechanisms and intervene upon these targets. Yet, these treatments do not work for half of those with BE, suggesting other putative mechanisms, such as reward processing, should be considered. To this end, informed by previous research showing an inverse association between ventral striatal (NAc) activation during monetary reward anticipation and behavioral frequency (severity) in other impulsive or compulsive behaviors (e.g., substance use, gambling, etc.), we tested the hypothesis that NAc activation during monetary reward anticipation would inversely scale with BE severity in a transdiagnostic sample of young women. Consistent with our hypothesis, we found an inverse association between activation of the left and right NAc during reward anticipation and BE severity. However, scatterplots from these dimensional analyses suggested this association might be driven by women who did not endorse BE. Indeed, post hoc analyses revealed significantly lower right – but not left – NAc activation during monetary reward anticipation in women with BE compared to those without BE. Among young women with BE, NAc activation during monetary reward anticipation did not correlate with BE severity. Thus, reward anticipation-related right NAc activation may differentiate women with BE from those without BE but does not appear to be a dimensional metric of BE severity.

Our finding suggesting lower right NAc activation during monetary reward processing in women with versus without BE is consistent with the maladaptive scaling hypothesis, which posits that psychopathology-related reinforcers (e.g., binge foods, etc.) become more rewarding and other reinforcers (e.g., money, etc.) become less rewarding over the course of illness. The maladaptive scaling hypothesis is dimensional and posits reward processing alterations scale with illness severity. As such, our dimensional (whole sample) findings suggesting an inverse association between left and right NAc activation during monetary reward anticipation and BE also fit with the maladaptive scaling hypothesis. Notably, we did not measure neural activation during food reward anticipation, precluding comparison of neural response to food and monetary reward anticipation. To date, only one study has directly compared neural response to food and monetary reward anticipation in persons with BE; they found increased reward-related neural activation during food – but not monetary – reward anticipation relative to persons without BE (Simon et al., 2016). Research examining food reward anticipation similarly documented increased neural activation during food reward anticipation in BE (Schienle et al., 2009; Uher et al., 2004). Additional research is needed to clarify anticipation-related neural response to food and other rewards in persons with BE. A recently proposed mechanistic staging model of reward processing alterations in binge-type eating disorders provides a framework to guide such research (Bodell and Racine, 2022). This model hypothesizes that reward processing of food and other reinforcers changes with illness stage, such that neural response to food reward anticipation is greater in recent-onset and established BE relative to controls, whereas neural response to monetary reward is lower than food reward in established BE.

To date, findings on neural response to monetary reward anticipation in BE have been mixed, with research suggesting lower (Balodis et al., 2013) or no difference in (Bodell et al., 2018; Simon et al., 2015) ventral striatal activation in persons with versus without BE. The mechanistic staging model of reward processing alterations in binge-type eating disorders could help to explain mixed findings on neural activation during monetary reward anticipation in BE. Although prior work has not considered illness duration, research has shown a strong positive association between age and illness duration in eating disorders (Meule et al., 2022). Our sample was restricted to young adults, whereas Simon et al. (2015) and Baldois et al. (2013) recruited adults with a wide age range, and Bodell et al. (2018) recruited 16-year-old adolescents with short illness durations. Moreover, neural processing of monetary reward changes as a function of developmental stage (Somerville et al., 2010) and healthy aging (Vink et al., 2015).

Other methodological differences may explain mixed findings. First, we created functional 5mm spheres of the NAc based on peak coordinates reported by Balodis et al. (2013), who administered the MIDT to persons with and without binge-eating disorder. Like this study, we found lower reward anticipation-related NAc activation within individuals with versus without BE. In contrast, Simon et al. (2015) used an anatomical ROI of the entire striatum (which encompassed the NAc, caudate, and putamen), and Bodell et al. (2018) created a functional 20mm sphere of the striatum with different coordinates than we used. A large striatal ROI encompassing ventral and dorsal subregions could have contributed to differences in results, given their dissociable functions. The ventral striatum has been implicated in reward anticipation whereas the dorsal striatum has been implicated in decision-making based on the experience of previous reward outcomes (Knutson et al., 2009). The functional ROI used in our study was smaller and specific to the NAc, potentially permitting greater specificity. Additionally, we recruited our sample from both the clinic and the community, whereas previous samples were treatment-seeking only or recruited exclusively from the community. Finally, the reward anticipation contrast was specified differently across studies. Whereas we and other studies using the MIDT (Balodis et al., 2013; Simon et al., 2016) specified reward anticipation as the contrast of neural response to a gain versus non-gain cue, Bodell et al. (2018) specified the contrast as neural response to a winning cue versus a fixation cross. Though this assumption is untested, neural response to a fixation cross and non-gain cue are likely not comparable, as non-gain may elicit disappointment from non-gain (Balodis and Potenza, 2015; Oldham et al., 2018). Thus, differences in sample age (and perhaps illness duration), ROI definition, treatment status, and contrast specifications may partly account for mixed findings.

Lateralization of Findings

Results suggest group differences in reward anticipation lateralized to the right NAc. Seminal MIDT research suggests lateralization of response to the right NAc during reward anticipation (Knutson et al., 2001), but meta-analytic findings do not suggest lateralization (Oldham et al., 2018). Of the previous research examining reward anticipation in BE, Bodell et al. (2018) found a lateralized response of the right striatum, Balodis et al. (2013) found a bilateral ventral striatal response, and Simon et al. (2016) did not find significant activation of the striatum. Recent meta-analytic work suggests response to reward anticipation is lateralized to the right striatum in persons with an eating disorder (Yu and Desrivières, 2023).

Limitations

Our findings should be considered in the context of some limitations. First, data collection began before the publication of recommendations for human neuroimaging research in eating disorders (Frank et al., 2018). Although our approach was consistent with many recommendations, we did not collect data on food and fluid intake before neuroimaging procedures, previous eating disorder diagnoses, or illness duration. Relatedly, although we collected menstrual cycle information and hormonal contraceptive use due to the known effects of these variables on reward processing (Dreher et al., 2007), menstrual cycle data were missing for ~25% of participants, precluding our ability to covary for menstrual cycle status. We conducted descriptive analyses using menstrual cycle information, and participants with and without BE did not significantly differ concerning cycle phase, amenorrhea, irregular periods, or hormonal contraceptive use. Third, our sample size was modest and comprised of young women, and findings may not generalize to larger, more heterogeneous (e.g., gender, age, etc.) samples with BE. Fourth, although unintentional, all women without BE did not have a non-eating disorder psychiatric diagnosis, whereas many women with BE had a comorbid psychiatric diagnosis. Thus, comorbidity could have impacted our results. Finally, our study design was cross-sectional, precluding the ability to test whether neural response to monetary reward anticipation prospectively predicted BE persistence or remission.

Clinical Implications and Future Directions

If there is reduced anticipation of general reinforcers, such as money, relative to food rewards in BE, novel treatments designed to target attenuated response to general reinforcers may hold promise. Indeed, pilot randomized trial data support the effectiveness of addressing this reward “imbalance” in BE (Juarascio et al., 2022). Additionally, positive affect treatment for anorexia nervosa is a neuroscience-informed treatment designed to enhance response to other reinforcers (Haynos et al., 2021). Adapting this intervention for binge-type eating disorders may hold promise for enhancing treatment outcomes. Future research is needed to investigate the extent to which reward anticipation-related NAc activation is associated with illness duration and maladaptive scaling, as proposed by Bodell and Racine’s (2022) staging model of binge-type eating disorders.

Supplementary Material

1

Acknowledgments:

The authors thank Z. Ayotola Onipede, Fabiola Valenzuela, and Hannah Welch for their study coordination and data collection efforts.

Funding:

This work was supported by the National Institute of Mental Health (K23 MH106794 to CB; T32 MH096678 supporting KH). The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the NIMH.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Competing Interests

The authors have no financial disclosures nor conflicts of interest to declare.

References

  1. Agras WS, Walsh T, Fairburn CG, Wilson GT, Kraemer HC, 2000. A multicenter comparison of cognitive-behavioral therapy and interpersonal psychotherapy for bulimia nervosa. Arch. Gen. Psychiatry 57, 459–466. 10.1001/archpsyc.57.5.459 [DOI] [PubMed] [Google Scholar]
  2. American Psychiatric Association, 2013. Diagnostic and statistical manual of mental disorders : DSM-5, 5th ed. American Psychiatric Association. [Google Scholar]
  3. Balodis IM, Grilo CM, Kober H, Worhunsky PD, White MA, Stevens MC, Pearlson GD, Potenza MN, 2014. A pilot study linking reduced fronto–Striatal recruitment during reward processing to persistent bingeing following treatment for binge-eating disorder. International Journal of Eating Disorders 47, 376–384. 10.1002/eat.22204 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Balodis IM, Kober H, Worhunsky PD, White MA, Stevens MC, Pearlson GD, Sinha R, Grilo CM, Potenza MN, 2013. Monetary Reward Processing in Obese Individuals With and Without Binge Eating Disorder. Biological Psychiatry, Food Addiction? 73, 877–886. 10.1016/j.biopsych.2013.01.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Balodis IM, Potenza MN, 2015. Anticipatory reward processing in addicted populations: a focus on the monetary incentive delay task. Biol Psychiatry 77, 434–444. 10.1016/j.biopsych.2014.08.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Beck AT, Brown GC, Steer RA, 1987. Beck Depression Inventory-II (BDI-II). The Psychological Corporation, San Antonio, TX. [Google Scholar]
  7. Berg KC, Peterson CB, Frazier P, Crow SJ, 2015. Psychometric evaluation of the Eating Disorder Examination and Eating Disorder Examination-Questionnaire: A systematic review of the literature. International Journal of Eating Disorders 45, 428–438. 10.1002/eat.20931 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bernardoni F, Bernhardt N, Pooseh S, King JA, Geisler D, Ritschel F, Boehm I, Seidel M, Roessner V, Smolka MN, Ehrlich S, 2020. Metabolic state and value-based decision-making in acute and recovered female patients with anorexia nervosa. J Psychiatry Neurosci 45, 253–261. 10.1503/jpn.190031 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Blum K, Braverman ER, Holder JM, Lubar JF, Monastra VJ, Miller D, Lubar JO, Chen TJ, Comings DE, 2000. Reward deficiency syndrome: a biogenetic model for the diagnosis and treatment of impulsive, addictive, and compulsive behaviors. J Psychoactive Drugs 32 Suppl, i–iv, 1–112. 10.1080/02791072.2000.10736099 [DOI] [PubMed] [Google Scholar]
  10. Bodell LP, Racine SE, 2022. A mechanistic staging model of reward processing alterations in individuals with binge-type eating disorders. Int J Eat Disord. 10.1002/eat.23875 [DOI] [PubMed] [Google Scholar]
  11. Bodell LP, Wildes JE, Goldschmidt AB, Lepage R, Keenan KE, Guyer AE, Hipwell AE, Stepp SD, Forbes EE, 2018. Associations Between Neural Reward Processing and Binge Eating Among Adolescent Girls. J Adolesc Health 62, 107–113. 10.1016/j.jadohealth.2017.08.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Dawe S, Loxton NJ, 2004. The role of impulsivity in the development of substance use and eating disorders. Neurosci Biobehav Rev 28, 343–351. 10.1016/j.neubiorev.2004.03.007 [DOI] [PubMed] [Google Scholar]
  13. Dolan SC, Reilly EE, Brown TA, Shott ME, Frank GKW, 2022. Anticipatory and consummatory pleasure in eating disorders. J Eat Disord 10, 161. 10.1186/s40337-022-00692-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dreher J-C, Schmidt PJ, Kohn P, Furman D, Rubinow D, Berman KF, 2007. Menstrual cycle phase modulates reward-related neural function in women. Proceedings of the National Academy of Sciences 104, 2465–2470. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Eklund A, Nichols TE, Knutsson H, 2016. Cluster failure: Why fMRI inferences for spatial extent have inflated false-positive rates. Proc. Natl. Acad. Sci. U.S.A 113, 7900–7905. 10.1073/pnas.1602413113 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Elliott ML, Knodt AR, Ireland D, Morris ML, Poulton R, Ramrakha S, Sison ML, Moffitt TE, Caspi A, Hariri AR, 2020. What Is the Test-Retest Reliability of Common Task-Functional MRI Measures? New Empirical Evidence and a Meta-Analysis. Psychol Sci 31, 792–806. 10.1177/0956797620916786 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Esteban O, Markiewicz CJ, Goncalves M, DuPre E, Kent JD, Salo T, Ciric R, Pinsard B, Blair RW, Poldrack RA, Gorgolewski KJ, 2021. fMRIPrep: a robust preprocessing pipeline for functional MRI. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Fairburn CG, 2008. Cognitive Behavior Therapy and Eating Disorders. The Guilford Press, New York, NY. [Google Scholar]
  19. Frank GK, Favaro A, Marsh R, Ehrlich S, Lawson EA, 2018. Toward valid and reliable brain imaging results in eating disorders. International Journal of Eating Disorders 51, 250–261. 10.1002/eat.22829 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gill H, Gill B, Lipsitz O, Rodrigues NB, Cha DS, El-Halabi S, Mansur RB, Rosenblat JD, Cooper DH, Lee Y, Nasri F, McIntyre RS, 2021. The impact of overweight/obesity on monetary reward processing: A systematic review. Journal of Psychiatric Research 137, 456–464. 10.1016/j.jpsychires.2021.03.029 [DOI] [PubMed] [Google Scholar]
  21. Haynos AF, Anderson LM, Askew AJ, Craske MG, Peterson CB, 2021. Adapting a neuroscience-informed intervention to alter reward mechanisms of anorexia nervosa: a novel direction for future research. Journal of Eating Disorders 9, 63. 10.1186/s40337-021-00417-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Jenkinson M, Beckmann CF, Behrens TEJ, Woolrich MW, Smith SM, 2012. FSL. Neuroimage 62, 782–790. 10.1016/j.neuroimage.2011.09.015 [DOI] [PubMed] [Google Scholar]
  23. Juarascio AS, Presseller EK, Wilkinson ML, Kelkar A, Srivastava P, Chen JY, Dengler J, Manasse SM, Medaglia J, 2022. Correcting the reward imbalance in binge eating: A pilot randomized trial of reward re-training treatment. Appetite 176, 106103. 10.1016/j.appet.2022.106103 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Knutson B, Adams CM, Fong GW, Hommer D, 2001. Anticipation of increasing monetary reward selectively recruits nucleus accumbens. J Neurosci 21, RC159. 10.1523/JNEUROSCI.21-16-j0002.2001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Knutson B, Delgado MR, Phillips PEM, 2009. Chapter 25 - Representation of Subjective Value in the Striatum, in: Glimcher PW, Camerer CF, Fehr E, Poldrack RA (Eds.), Neuroeconomics. Academic Press, London, pp. 389–406. 10.1016/B978-0-12-374176-9.00025-7 [DOI] [Google Scholar]
  26. Knutson B, Westdorp A, Kaiser E, Hommer D, 2000. FMRI Visualization of Brain Activity during a Monetary Incentive Delay Task. NeuroImage 12, 20–27. 10.1006/nimg.2000.0593 [DOI] [PubMed] [Google Scholar]
  27. Latzman RD, DeYoung CG, 2020. Using empirically-derived dimensional phenotypes to accelerate clinical neuroscience: the Hierarchical Taxonomy of Psychopathology (HiTOP) framework. Neuropsychopharmacol. 45, 1083–1085. 10.1038/s41386-020-0639-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Lecrubier Y, Sheehan DV, Weiller E, Amorim P, Bonora I, Sheehan KH, Janavs J, Dunbar GC, 1997. The Mini International Neuropsychiatric Interview (MINI). A short diagnostic structured interview: reliability and validity according to the CIDI. European psychiatry 12, 224–231. [Google Scholar]
  29. Leenaerts N, Jongen D, Ceccarini J, Van Oudenhove L, Vrieze E, 2022. The neurobiological reward system and binge eating: A critical systematic review of neuroimaging studies. Int J Eat Disord 55, 1421–1458. 10.1002/eat.23776 [DOI] [PubMed] [Google Scholar]
  30. Linardon J, 2018. Rates of abstinence following psychological or behavioral treatments for binge-eating disorder: Meta-analysis. Int J Eat Disord 51, 785–797. 10.1002/eat.22897 [DOI] [PubMed] [Google Scholar]
  31. Linardon J, Wade TD, 2018. How many individuals achieve symptom abstinence following psychological treatments for bulimia nervosa? A meta-analytic review. Int J Eat Disord 51, 287–294. 10.1002/eat.22838 [DOI] [PubMed] [Google Scholar]
  32. Luijten M, Schellekens AF, Kühn S, Machielse MWJ, Sescousse G, 2017. Disruption of Reward Processing in Addiction : An Image-Based Meta-analysis of Functional Magnetic Resonance Imaging Studies. JAMA Psychiatry 74, 387–398. 10.1001/jamapsychiatry.2016.3084 [DOI] [PubMed] [Google Scholar]
  33. Meule A, Kolar DR, Rauh E, Voderholzer U, 2022. Comparing illness duration and age as predictors of treatment outcome in female inpatients with anorexia nervosa. Eating Disorders 0, 1–11. 10.1080/10640266.2022.2114586 [DOI] [PubMed] [Google Scholar]
  34. Mitchison D, Touyz S, González-Chica DA, Stocks N, Hay P, 2017. How abnormal is binge eating? 18-Year time trends in population prevalence and burden. Acta Psychiatr Scand 136, 147–155. 10.1111/acps.12735 [DOI] [PubMed] [Google Scholar]
  35. Oldham S, Murawski C, Fornito A, Youssef G, Yücel M, Lorenzetti V, 2018. The anticipation and outcome phases of reward and loss processing: A neuroimaging meta-analysis of the monetary incentive delay task. Hum Brain Mapp 39, 3398–3418. 10.1002/hbm.24184 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Peters J, Bromberg U, Schneider S, Brassen S, Menz M, Banaschewski T, Conrod PJ, Flor H, Gallinat J, Garavan H, Heinz A, Itterman B, Lathrop M, Martinot J-L, Paus T, Poline J-B, Robbins TW, Rietschel M, Smolka M, Ströhle A, Struve M, Loth E, Schumann G, Büchel C, 2011. Lower Ventral Striatal Activation During Reward Anticipation in Adolescent Smokers. AJP 168, 540–549. 10.1176/appi.ajp.2010.10071024 [DOI] [PubMed] [Google Scholar]
  37. Reuter J, Raedler T, Rose M, Hand I, Gläscher J, Büchel C, 2005. Pathological gambling is linked to reduced activation of the mesolimbic reward system. Nat Neurosci 8, 147–148. 10.1038/nn1378 [DOI] [PubMed] [Google Scholar]
  38. Schienle A, Schäfer A, Hermann A, Vaitl D, 2009. Binge-eating disorder: reward sensitivity and brain activation to images of food. Biol Psychiatry 65, 654–661. 10.1016/j.biopsych.2008.09.028 [DOI] [PubMed] [Google Scholar]
  39. Siegel JS, Power JD, Dubis JW, Vogel AC, Church JA, Schlaggar BL, Petersen SE, 2014. Statistical improvements in functional magnetic resonance imaging analyses produced by censoring high-motion data points. Human Brain Mapping 35, 1981–1996. 10.1002/hbm.22307 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Simon JJ, Skunde M, Walther S, Bendszus M, Herzog W, Friederich H-C, 2016. Neural signature of food reward processing in bulimic-type eating disorders. Soc Cogn Affect Neurosci 11, 1393–1401. 10.1093/scan/nsw049 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Simon JJ, Skunde M, Wu M, Schnell K, Herpertz SC, Bendszus M, Herzog W, Friederich H-C, 2015. Neural dissociation of food- and money-related reward processing using an abstract incentive delay task. Soc Cogn Affect Neurosci 10, 1113–1120. 10.1093/scan/nsu162 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Somerville LH, Jones RM, Casey BJ, 2010. A time of change: behavioral and neural correlates of adolescent sensitivity to appetitive and aversive environmental cues. Brain Cogn 72, 124–133. 10.1016/j.bandc.2009.07.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Stice E, Marti CN, Rohde P, 2013. Prevalence, incidence, impairment, and course of the proposed DSM-5 eating disorder diagnoses in an 8-year prospective community study of young women. J Abnorm Psychol 122, 445–457. 10.1037/a0030679 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Tannous WK, Hay P, Girosi F, Heriseanu AI, Ahmed MU, Touyz S, 2021. The economic cost of bulimia nervosa and binge eating disorder: a population-based study. Psychological Medicine 1–15. 10.1017/S0033291721000775 [DOI] [PubMed] [Google Scholar]
  45. Uher R, Murphy T, Brammer MJ, Dalgleish T, Phillips ML, Ng VW, Andrew CM, Williams SCR, Campbell IC, Treasure J, 2004. Medial Prefrontal Cortex Activity Associated With Symptom Provocation in Eating Disorders. AJP 161, 1238–1246. 10.1176/appi.ajp.161.7.1238 [DOI] [PubMed] [Google Scholar]
  46. Vink M, Kleerekooper I, van den Wildenberg WPM, Kahn RS, 2015. Impact of aging on frontostriatal reward processing. Human Brain Mapping 36, 2305–2317. 10.1002/hbm.22771 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Wilson RP, Colizzi M, Bossong MG, Allen P, Kempton M, Abe N, Barros-Loscertales AR, Bayer J, Beck A, Bjork J, Boecker R, Bustamante JC, Choi JS, Delmonte S, Dillon D, Figee M, Garavan H, Hagele C, Hermans EJ, ICCAM Consortium, Ikeda Y, Kappel V, Kaufmann C, Lamm C, Lammertz SE, Li Y, Murphy A, Nestor L, Pecina M, Pfabigan D, Pizzagalli D, Rademacher L, Admon R, Sommer T, Stark R, Suzuki H, Van Amelsvoort T, Van Hell E, Vink M, Votinov M, Wotruba D, Bhattacharyya S, MTAC, 2018. The Neural Substrate of Reward Anticipation in Health: A Meta-Analysis of fMRI Findings in the Monetary Incentive Delay Task. Neuropsychol Rev 28, 496–506. 10.1007/s11065-018-9385-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Yu X, Desrivières S, 2023. Altered anticipatory brain responses in eating disorders: A neuroimaging meta-analysis. Eur Eat Disord Rev. 10.1002/erv.2967 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Zald DH, Treadway MT, 2017. Reward Processing, Neuroeconomics, and Psychopathology. Annual Review of Clinical Psychology 13, 471–495. 10.1146/annurev-clinpsy-032816-044957 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1881333-supplement-1.docx (28.4KB, docx)

RESOURCES