Childhood appetitive traits are heritable behavioural phenotypes proposed to mediate genetic susceptibility to increased adiposity. Despite this hypothesis, the genetic architecture of these appetitive behaviours and their role in transmitting polygenic risk for obesity have been incompletely characterised. The present study aimed to map genome-wide associations with parent-reported eating behaviours in childhood, quantify overlap with known BMI loci, and estimate how much of genetic association with childhood BMI can be decomposed through specific appetitive traits.
The analysis used data from the Norwegian Mother, Father and Child Cohort Study (MoBa), analysing up to 31,018 eight-year-old children. Appetitive domains were derived from 18 items of the parent-reported Children's Eating Behaviour Questionnaire, collapsed into six appetitive domains. Genome-wide association studies (GWAS) were performed for these domains. A trio-based design (child and both parents) was used to decompose direct genetic effects transmitted to the child from indirect parental genetic effects that might operate via parental phenotype or reporting. The study reports summary-level results and Supplementary Information; individual-level MoBa data are not publicly available without application and relevant approvals.
Across the six appetitive domains, the study identified ten independent genome-wide significant loci associated with childhood eating behaviour. The strongest findings were concentrated primarily in domains of Food Responsiveness, Satiety Responsiveness, and Food Fussiness. The set of significant loci provides the largest reported GWAS survey of childhood eating behaviour to date in this cohort.
Of the ten genome-wide significant loci for childhood eating behaviour, eight lie at loci previously established for childhood or adult BMI, indicating substantial shared genetic architecture between appetitive traits and adiposity. Both Food Responsiveness and Satiety Responsiveness demonstrated phenotypic and genetic correlations with BMI trajectories spanning early childhood through adolescence, supporting the view that these appetite dimensions are relevant to weight development over time.
Statistical mediation analyses were conducted to estimate how much of the aggregated genetic association with BMI at age 8 could be decomposed through appetitive traits. The authors report that portions of the aggregated genetic association were mediated: 22.1% and 10.4% via the appetitive traits examined (the manuscript provides trait-level attribution). These mediation estimates were obtained under standard mediation assumptions and indicate that a substantial, though partial, fraction of polygenic adiposity risk is routed through specific eating behaviours.
Locus-specific patterns suggested distinct mechanistic pathways linking genotype to behaviour and adiposity. The FTO locus appeared to operate predominantly via Food Responsiveness, consistent with an effect on hedonic or food-cue–driven intake. In contrast, the ADCY3 locus showed patterns consistent with action via Satiety Responsiveness, implying mechanisms related to internal satiety signalling or appetite suppression. These locus-level distinctions support the notion of multiple behavioural routes from genotype to BMI.
Using the trio design, the study decomposed associations into effects attributable to the child's inherited alleles and indirect effects from parental genotype. Results indicated that both BMI and eating behaviour associations were predominantly explained by children's own inherited alleles, with minimal contribution from indirect parental adiposity effects. An exception was observed for parental influence on reporting: parental genetic liability influenced reporting of Satiety Responsiveness, suggesting parental genotype can affect measured scores via reporting bias or related parental behaviours.
The findings indicate that childhood appetitive traits capture a meaningful portion of genetic susceptibility to adiposity through distinct behavioural pathways. Because these effects are driven mainly by the child's genotype rather than indirect parental genetic effects, appetite traits represent biologically grounded targets for early preventive strategies directed at eating behaviour. The study notes that individual-level MoBa data are not publicly available due to consent and data protection requirements; researchers may apply for access through the Norwegian Institute of Public Health. The manuscript and Supplementary Information report summary-level results, and ethical approvals were obtained from the Regional Committees for Medical and Health Research Ethics (REK), Norway.