Avoidant/restrictive food intake disorder (ARFID) is a selective eating disorder and a chronic psychiatric condition. It affects between 0.3% to 4% of adults and between 0.3% to 15.5% of children. The disorder can lead to serious health complications, including loss of body weight, impaired growth, dependence on supplemental feeding, mental health disorders, and nutritional deficiencies. To date, no studies have explored the neural mechanism of ARFID while including healthy control (HC) subjects. A research study published in JAMA Network Open provided a comprehensive investigation of these mechanisms.
The objective of the study was to determine whether subjects with ARFID show disruptions in brain regions associated with appetite, disgust, and fear compared to HC individuals when viewing images of objects and food.
This case-control study was conducted to strengthen the reporting of observational studies in epidemiology (STROBE) guidelines. Subjects enrolled in this study were aged 10-23 years, among them, 59% of females were recruited between July 2016 and January 2021. All these participants completed the interviews, including eating disorder assessment (EDA-5), Kiddie schedule for affective disorders and schizophrenia (KSADS), Pica, ARFID, and rumination disorder interview (PARDI) and eating disorder examination (EDE) and validated functional magnetic resonance imaging (fMRI) food cue paradigm. The blood oxygen level-dependent (BOLD) activation was measured in the whole brain and regions of interest (ROIs), including the hypothalamus, anterior cingulate cortex (ACC), amygdala, and insula.
A total of 110 participants were included, consisting of 75 participants with partial or full ARFID (mean age = 16.2±3.8 years, 55% female) and 35 matched age with HC subjects (mean age = 17.3±4 years, 69% female). Results showed that ARFID individuals exhibited greater BOLD activation compared to HC participants in supplementary motor cortex with a mean difference of 0.81 (95% confidence interval [CI], 0.47 to 1.15), P = 0.04, ACC with mean difference of 0.48 (95% CI, 0.19 to 0.77), P = 0.009, right sensory association cortex with mean difference of 0.52 (95% CI, 0.28 to 0.76, P = 0.02) and left sensory association cortex with mean difference of 0.54 (95% CI, 0.29 to 0.79), P = 0.005.
The ARFID-fear group demonstrated greater activation in the amygdala compared to HC, with a mean difference of 0.49 (95% CI, 0.16 to 0.82) and a P value of 0.04. Moreover, within the ARFID-lack of interest group, there was a significant lack of interest in eating reported with lower activation in the hypothalamus (r = −0.38 [95% CI, −0.69 to −0.11]; P = 0.03).
The ARFID-sensory sensitivity group did not exhibit greater activation in the insula compared to HC individuals. In contrast to increased activation of somatosensory cortex (mean difference on left side, 0.60 [95% CI, 0.33-0.87]; P = 0.001; right side, 0.54 [95% CI, 0.29 to 0.80]; P = 0.03) and ACC (mean difference, 0.48 [95% CI, 0.22 to 0.74]; P = 0.005) was observed in sensitivity group compared to HC group.
This study’s limitations include modest sample size, sample heterogeneity due to wide age range, phenotype overlap, passive viewing task, psychoactive drug use, and limited to non-Hispanic White participants.
In conclusion, this study suggests that hyperactivation of the sensory association cortex, ACC, and supplementary motor cortex in response to visual food stimuli among study participants with ARFID. This highlights a potential neurobiological circuit underlying the disorder. ARFID phenotypes consistently show activation in food avoidance along with activation shown in the fear region of the ARFID-fear group and reduced activation shown in the appetite region of the ARFID-lack of interest group.
Reference: Thomas JJ, Holsen L, Van De Water AL, et al. Neural Response to Food Cues in Avoidant/Restrictive Food Intake Disorder. JAMA Netw Open. 2025;8(2):e2460101. doi:10.1001/jamanetworkopen.2024.60101
