To determine whether it is the magnitude of early postnatal catch-up growth (CUG) in response to fetal growth restriction (FGR) or the FGR itself that relates to cognitive outcome in a model of monochorionic twins discordant for fetal growth. This analysis is part of the LEMON study, a cohort study including all monochorionic twins with selective FGR aged 3 through 17 years. Growth measurements as documented by our primary care system were collected retrospectively. An age-appropriate neurodevelopmental test was performed generating a full-scale intelligence quotient (FSIQ). CUG at two years was calculated as (weight [kg] at two years - birth weight [kg]). We used a multivariable regression model investigating the association between FSIQ (outcome) and birth weight z-score, gestational age at birth and CUG at two years (predictors). Generalized estimating equations accounted for the fact that observations between co-twins are not independent. Median age at follow-up of the 46 included twin pairs was 11 (IQR 8-13) years. Birth weight z-score and gestational age at birth were significantly associated with FSIQ, with β-coefficients of 5.897 (95% CI 3.382-8.411), and 2.589 (95% CI 1.227-3.951), respectively (p<0.0001). Adjusted for birth weight z-score and gestational age, CUG in the first two years after birth was not significantly associated with FSIQ (β-coefficient 0.108 (95% CI -1.373-1.590), p=0.886). Our results, combining detailed growth measurements and neurodevelopmental follow-up in a discordant identical twin model, demonstrate that FGR itself rather than early postnatal CUG has negative consequences for cognitive development.