Table 2 Characteristics and key findings of included studies using fMRI as the primary method.

1. Amianto et al. (2013) [57]

Cerebellum, 12: 623-631.

AN (n=12)

BN (n=12)

HC (n=10)

AN: 20(4) BN: 23(5) HC: 24(3)

100%

One resting-state fMRI.

Lifetime history of psychosis, schizophrenia, schizoaffective disorder, delusional disorder, bipolar I/II disorder, psychotic depression, organic mood disorder; severe medical illness; severe underweight that could not be managed as an outpatient; use of psychotropic medication; neurological disease

AN vs BN & HC: grey matter reduction

AN & BN vs HC: hyperconnectivity of the cerebellar network to the parietal cortex; increased bilateral connectivity of cerebellar ICN with temporal poles

BN vs AN & HCs: GMV reduction in the CN

2. Balodis et al. (2013) [23]

Biological Psychiatry, 73:877-886.

Obese BED (n=19)

Obese non-BED (n=19)

HCs (n=19)

BED 43.7 (12.7)

OB 38.3 (7.5)

HC 34.8 (10.7)

BED 73.7%

OB 52.6%

HC 52.6%

fMRI completed while completing MIDT (monetary incentive delay task)

In the obese non-BED or HC group: past history of, or current binge eating or other eating disorder diagnosis

Anticipation processing:

BED vs OB: decreased ventrostriatal (VS) and striatal activity;

OB vs HCs: increased VS activity

Outcome processing:

BED vs OB & HCs: diminished activity in PFC and Insular

3. Bohon & Stice (2011) [58]

International Journal of Eating Disorders, 44(7): 585-595.

BN sub-threshold* (1xBE & comp/wk) (n=11)

BN (n=2)

HCs (n=13)

* (Sub-BN = 1 x binge episode/week – sub-threshold for DSM-IV criteria, however this frequency meets the new DSM-V criteria for BN)

Not reported per group.

For all participants: 20.3 (1.87)

100%

fMRI examining reward circuitry during actual (choc milkshake) and anticipated (tasteless solution) food intake

Any Axis I disorder; food allergy to milkshake / taste aversion to chocolate milkshake

In the BN group: psychoactive medications other than SSRIs (sertraline & fluoxetine)

BN vs HCs: less activation in right precentral gyrus in both anticipatory and consumatory conditions; less activation in right anterior insula while anticipating the milkshake; and less activation in the left middle frontal gyrus, right posterior insula, left thalamus in response to milkshake

4. Brooks et al. (2011) [42]

PLoS One, 6(7):e22259.

BN (n=8)

AN-R (n=11)

AN-BP (n=7)

HCs (n=24)

BN: 25 (7.1)

AN: 26 (6.8)

HC: 26 (9.5)

100%

fMRI while asking participants to imagine eating the foods shown in photographs (72 colour photos of high and low energy, sweet & savoury foods; 72 photos of non-food items

Left handedness; caffeine / alcohol within specified times preceding the fMRI; history of head trauma, hearing or visual impairment, neurological disease

In the ED groups: psychotropic medications other than SSRIs

In response to food vs. non-food images:

BN vs HCs and AN: greater activation in visual cortex, right dorsolateral prefrontal cortex, right insular cortex and precentral gyrus

BN vs. HC: deactivation in bilateral superior temporal gyrus, insular cortex, visual cortex

BN vs AN: reduced activation in parietal lobe, dorsal posterior cingulate cortex

BN vs AN-R: increased activation in bilateral inferior temporal lobe, left visual cortex, posterior cingulate & left inferior parietal lobe and deactivation in right precentral gyrus

BN vs AN-BP: greater activation in the left cerebellum, left parahippocampal gyrus, left posterior cingulate cortex, right supplementary motor area, and deactivation in left inferior temporal gyrus

5. Celone et al. (2011) [25]

NeuroImage, 56: 1749-1757.

‘Sub-threshold’* BN (n=18)

HCs (n=19)

* (Sub-BN = 1 x binge episode/week – sub-threshold for DSM-IV criteria, however this frequency meets the new DSM-V criteria for BN)

Sub-BN: 20.67 (2.10)

HC: 20.42 (1.95)

100%

fMRI during Weather Prediction Task (WPT), a probabilistic learning paradigm.

Previous or current neurological or medical disease; learning disability; substance abuse; history of significantly low body weight (<85% of ideal body weight); past or current AN

No behavioural differences in performance

Results demonstrate processing inefficiencies in the fronto-striatal system in BN. BN women demonstrated increased overall category learning-related activity in the right caudate nucleus and bilateral dorsolateral PFC and decreased suppression of the category learning related BOLD signal in the anterior cingulate cortex. The direction of the BOLD signal changes within the fronto-striatal system differs from the initial hypothesis

6. Cyr et al. (2016) [43]

Journal of the American Academy of Child and Adolescent Psychiatry, 55(11): 963-972.

BN (n=27)

HCs (n=27)

BN: 16.6 (1.5)

HC:16.3 (2.1)

100%

fMRI BOLD response during reward based spatial learning task (virtual learning)

History of neurological illness; past seizures; head trauma with loss of consciousness (LOC); mental retardation; pervasive developmental disorder; current Axis I disorder (other than depressive / anxiety disorder for clinical group)

BN vs HCs: engaged the right anterior hippocampus when receiving unexpected rewards

HCs vs BN: engaged the right IFC when searching spatially and the right anterior hippocampus when receiving expected rewards

Overall the data suggest abnormal functioning of the anterior hippocampus and fronto-striatal circuits during reward-based spatial learning

Clinical correlates: Severity of BN was significantly associated with activation of the right anterior hippocampus during reward processing

7. Lee et al. (2017) [44]

Neuroscience Letters, accepted manuscript 26/04/17

BN (n= 13)

BED (n=12)

HC (n=14)

BN: 23.7 (2.2)

BED: 23.6 (2.6)

HC: 23.3 (2.2)

100%

fMRI performed while participants completed the Stroop match-to-sample task, in which participant attention is controlled by an interaction between bottom-up sensory processing and top-down cognitive processing driven mainly by the prefrontal cortex. The task was modified to include food and non-food conditions.

BMI < 17.5; current or past psychiatric disorder; traumatic brain injury; neurological illness; current or past use of psychiatric medications

BN vs HC: lower accuracy indicating impaired cognitive control over interference. Higher activation in the premotor cortex and dorsal striatum in response to food images

BED vs HC: higher activation in the ventral striatum in response to food images

8. Marsh et al. (2009) [45]

Archives of General Psychiatry, 66(1): 51-63.

BN (n=20)

HCs (n=20)

BN: 25.7(7.0)

HC: 26.35(5.7)

100%

fMRI used to examine BOLD during performance on a Simon spatial incompatibility task (SSIT). Two groups compared on patterns of brain activation.

History of neurological illness; past seizures; head trauma with LOC; mental retardation, pervasive developmental delay

In the BN group: current Axis I disorder excluding major depression

BN vs HC: responded significantly more impulsively and made a greater number of errors on the SSIT

BN group: The number of objective binge episodes correlated inversely with the significantly increased activation of the right medial prefrontal, temporal and inferior parietal cortices

HC vs BN: greater activation in the anterior cingulate cortex during incorrect responses and activated the striatum more when responding incorrectly

9. Marsh et al. (2011) [46]

American Journal of Psychiatry, 168(11): 1210-1220.

BN (n=18)

HCs (n=18)

BN: 18.4 (2.1)

HC: 17.3 (2.4)

100%

fMRI used to examine BOLD during performance on a Simon spatial incompatibility task. Two groups compared on patterns of brain activation.

History of neurological illness; past seizures; head trauma with LOC; mental retardation, pervasive developmental delay

In the BN group: current Axis I disorder excluding major depression

BN and HCs performed comparably however during correct responses in conflict trials the frontostriatal circuits failed to activate to the same degree in the BN group

BN vs HC: demonstrated abnormal patterns of activation in the frontostriatal and ‘default mode’ systems; specifically they did not have the same magnitude of activity in the frontostriatal circuits known to underlie self-regulatory control, including the right inferior frontal gyrus, dorsolateral PFC, and putamen

10. Marsh et al. (2015) [47]

Biological Psychiatry, 77: 616-623.

BN adolescent <19yo (n=16)

BN adult (n=16)

HCs (n=34)

Not reported for either group; only that there were adolescents and adults in both BN & HCs.

100%

fMRI completed to compare morphological characteristics of their cerebral surface

History of neurological illness; past seizures; head trauma with LOC; mental retardation, pervasive developmental delay

In the BN group: current Axis I disorder excluding major depression

BN vs HCs: Significant reduction of local volumes on brain surface found in the frontal and temperoparietal areas (bilateral middle frontal and precentral gyri; right postcentral gyrus and lateral superior, and lateral superior and inferior frontal gyri of the left hemisphere). Reductions were also found in temperoparietal regions including bilateral inferior temporal gyri, right superior parietal gyrus and cuneus, bilateral posterior cingulate cortices, left precuneus and fusiform gyrus.

Enlargements detected in the bilateral middle/inferior occipital and lingual gyri and right inferior parietal lobule in the BN group.

Significant inverse associations reported between cerebral surface morphology and objective binge and vomiting episodes in the bilateral IFG, PreCG and PoCG.

11. Miyake, Okamoto, Onada, Kurosaki et al., (2010) [59]

Neuroimaging, 181: 183-192.

BN (n=11)

AN-R (n=11)

AN-BP (n=11)

HCs (n=11)

BN: 24.5 (5.8)

AN-R: 22.2 (4.1)

AN-BP: 28.3 (4.5)

HC: 26.5 (5.5)

100%

fMRI with emotional decision task with distorted body images (varying degrees of ‘thinness and fatness’ of own and healthy female body photo)

Presence of Axis I or II disorder other than ED; left handedness

In the BN and HC groups: history of AN

In AN-R, AN-BP and HCs, but not BN, the amygdala was significantly activated in response to own ‘fat-image’

AN-BP and HCs vs BN and AN-R: medial PFC significantly activated

AN-R vs other ED groups: amygdala significantly more activated in response to ‘fat’ image of another woman

12. Miyake, Okamoto, Onada, Shirao et al. (2010) [49]

Neuroimage, 50: 1333-1339.

BN (n=12)

AN-R (n=12)

AN-BP (n=12)

HCs (n=12)

BN: 25.0 (6.9)

AN-R: 27.0 (9.0)

AN-BP: 27.2 (4.8)

HC: 25.4 (5.8)

100%

fMRI while completing emotional word decision making task, examining processing of words (negative body image words e.g.obesity; and neutral words).

Presence of Axis I or II disorder other than ED; left handedness

Note - one BN participant had a history of AN

Negative body image words condition:AN-R & AN-BP vs BN & HC: right amygdala significantly more activated

BN & AN-BP vs HC: left medial PFC significantly more activated

AN-R & AN-BP vs HCs: left inferior parietal lobule significantly more activated

Overall: results indicated that distorted cognition of negative body image words in eating disorder patients were related to enhanced activation in amygdala and mPFC. However there were no group differences in nonspecific negative emotion words

13. Mohr et al. (2011) [50]

NeuroImage, 56: 1822-1831.

BN (n=15)

HCs (n=15)

BN: 24.8 (3.2)

HC: 25.5 (4.5)

100%

History of substance abuse; schizophrenia and psychotic symptoms; bipolar disorder; neurological illness; closed head injury; left handedness

The activation pattern in the insula reflected satisfaction ratings of BN and HCs

HCs vs BN: in terms of general differences in body image processing (not specifically during satisfaction / perception conditions), the MFG & right posterior parietal cortex demonstrated significantly greater activation (potentially reflecting a reduced spatial manipulation capacity)

HC vs BN: during body size estimation/perception task, the MFG was significantly more activated in HCs than BN, and the MFG was recruited significantly more in the perception vs satisfaction task

HCs vs BN: posterior temporal-occipital cortex was sensitive for body image distortion (this ‘type of modulation’ was not observed in BN)

HC & BN: The amount of bilateral insula activity reflected the pattern of satisfaction rating task. In BN a linear trend occurred with a decline in insula and MFG activity from thinner to fatter images, although results weren’t as clear in HCs

14. Pringle et al. (2011) [51]

Neuropsychologia

49:3272-3278.

BN: 24.55 (4.97)

HC: 27.38 (5.44)

100%

fMRI to examine self-referent emotional processing, where patients had to endorse 60 personality characteristic words as ‘me’ or ‘not me’ in rapid event related design.

Left handedness; medication

BN vs HCs: rating of negative personality descriptors was associated with reduced activity in the parietal, occipital and limbic areas, including the amygdala

15. Schienle et al. (2009) [52]

Biological Psychiatry, 65: 654-661.

BED (n=17)

BN-P (n=14)

HCs normal weight (n=19)

Controls - overweight (C-OW) (n=17)

BED: 26.4 (6.4)

BN-P: 23.1 (3.8)

HC-N: 22.3 (2.6)

HC-O: 25.0 (4.7)

100%

Medication; clinically relevant depression; left handedness

All participants demonstrated increased activation in the OFC, ACC and insula highlighting a basic appetitive response pattern. No group differences in disgust-inducing images

BED vs. all other groups: enhanced reward sensitivity, stronger medial OFC activity while viewing food images

BN vs. all other groups: greater ACC activation and insula activation while viewing food images

16. Seitz et al. (2016) [18]

European Child and Adolescent Psychiatry, 1: S185-203.

BN (n=20)

HCs (n=20)

BN: 18.71 (2.53)

HC: 17.90 (1.35)

100%

History of psychosis; substance abuse; IQ <80

BN vs. HCs:

• Higher ADHD scores, especially inattention.

• Hyperactivation in the parieto-occipital regions and reduced deactivation of the precuneus, part of the default-mode-network areas during ‘alerting’

• Posterior cingulate activation during alerting correlated with severity of BN symptoms

• Exploratory correlation analyses found significant associations between neural activity in the ‘alerting’ condition in the bilateral middle cingulate and global eating disorder symptoms; significant inverse correlation between activity in the temperoparietal junction and ADHD symptoms; and activity in the right parahippocampus was inversely correlated with impulsivity scores

17. Skunde et al. (2016) [26]

Journal of Psychiatry and Neuroscience, 41(5): E69-E78.

BN (n=28)

HCs (n=29)

BN: 27.54 (10.52)

HC: 27.25 (6.68)

100%

fMRI while completing a general and food-specific (participants selected 8 of their favourite food images from a set of 85 high-calorie foods prior to completing the task) no-go task (the no-go task is a sub-task of the go-no-go task and measures behavioural inhibition).

Biploar disorder; psychosis; history of head injury; neurologic disorder; diabetes mellitus; nicotine / drug / alcohol abuse; lifetime diagnosis of BPD

In HCs: current psychotropic meditation

In BN: medication other than antidepresants

BN vs. HCs: reduced activation in the right sensorimotor area (postcentral gyrus, precentral gyrus) and right dorsal striatum (caudate nucleus, putamen)

HC vs BN (high frequency BEs only): stronger activation in the right postcentral gyrus, right caudate nucleus and right putamen

18. Spangler et al. (2012) [53]

International Journal of Eating Disorders, 45(1): 17-25

BN (n=12)

HCs (n=12)

BN: Age range reported only (18-38)

HC: (18-30)

100%

fMRI while looking at computer- generated images of ‘thin’ (BMI= 18) or ‘fat’ (BMI=31) bodies (and control condition: scrambled image). Participants instructed to imagine someone is comparing your body to the body of the woman in the picture.

In BN: medication other than antidepressants

BN: no significant difference found in brain activation while looking at thin vs fat images

BN vs. HC: mPFC activation was significantly greater while viewing ‘fat’ images, with increased activity in the regions associated with emotional processing. No differences between groups in the thin condition

In the mPFC, the peak location of activation for BN patients was in the pgACC rather than dorsal mPFC, as it was for HCs

19. Uher et al. 2004 [54]

American Journal of Psychiatry, 161(7): 1238-1246

BN (n=10)

AN (n=16)

HCs (n=19)

BN: 29.80 (8.80)

AN: 26.93 (12.14)

HC: 26.68 (8.34)

100%

fMRI completed while being presented with photographs of savoury and sweet foods; non-food items; emotionally aversive photographs and neutral stimuli.

Axis I disorders other than ED; neurological or psychiatric illness aside from ED; psychotropic medication other than antidepressants

BN vs HCs: greater occipital and cerebellar activity

BN vs AN & HCs: decreased activation in the anterior and lateral PFC in response to food images (associated with suppressing unwanted behaviours)

AN & BN vs HC: significantly increased medial PFC reaction to food images

20. Uher et al. 2005 [55]

Biological Psychiatry, 58(12):990-997

BN (n=9)

AN (n=13)

HCs (n=19)

BN: 29.6 (9.3)

AN: 25.4 (10.2)

HC: 26.6 (8.6)

100%

fMRI to examine cerebral correlates of body image activity when participants looking at line drawings of underweight (BMI = <17.5), normal weight (20<BMI<25), and overweight (BMI 27.5) female bodies vs. control images (line drawings of houses)

Psychosis; alcohol or drug dependence; neurological or psychiatric illness aside from ED; psychotropic medication other than antidepressants

No regions of significantly increased activations in either eating disorder group, compared to the control subjects

Across AN, BN & HCs, the lateral fusiform gyrus, inferior parietal cortex and lateral PFC were activated in response to body shapes vs. control condition

21. Vocks et al. 2010 [56]

Journal of Psychiatry and Neuroscience, 35(3): 163-176.

AN (n=13: 8 AN-R and 6 AN-BP)

BN (n=15)

HC (n=27)

AN: 29.08 (9.79)

BN: 28.4 (7.07)

HCs: 26.74 (7.6)

100%

fMRI while participants looked at 16 standardised photographs of their own body and another woman’s body (BMI 19), taken while wearing a bikini.

Left handedness; personality disorder

AN & BN vs. HCs: while viewing photographs of their own body, eating disorder patients showed weakened activity in the left inferior parietal lobule

AN vs. BN & HCs: higher amygdala activity while looking at photographs of another woman’s body

AN vs. BN & HCs: significantly greater activation in the bilateral superior temporal gyrus