Prospective Cohort Finds 10% Higher Early Ultra-Processed Food Intake Tied to 2% Smaller Reward-Related Brain Volumes at Age 6

The Children's Hospital Los Angeles team tracked diets via repeated 24-hour recalls and food-frequency questionnaires from infancy onward, then performed structural MRI at age 6. Higher ultra-processed food exposure correlated with reduced volumes in the nucleus accumbens, caudate, and putamen after adjustment for total energy intake, maternal education, and BMI. No differences emerged on standardized cognitive testing, indicating structural changes may precede functional deficits.

This pattern aligns with prior observational data from the ABCD study showing ultra-processed food consumption linked to altered frontostriatal connectivity in older children, and with rodent models demonstrating high-glycemic processed diets blunt dopamine signaling during critical windows of striatal maturation. The exclusive Latino/Hispanic sample raises questions about generalizability but also highlights a population with elevated ultra-processed food intake and rising pediatric obesity rates.

Because the design is observational, residual confounding by overall diet quality or socioeconomic factors cannot be excluded; randomized feeding trials starting in infancy are now required to test causality and determine whether reducing ultra-processed foods alters trajectories of brain volume or later behavioral outcomes.

Evidence calibration: This single prospective cohort with serial diet measures and imaging can demonstrate temporal associations but cannot establish causation or long-term clinical impact; replication in diverse cohorts with pre-registered analysis plans and extended neurodevelopmental follow-up to adolescence is the necessary next step.