Structural and functional development of the orbitofrontal cortex during adolescence.

Abstract

This study was designed to assess the structural and functional development of the orbitofrontal cortex (OFC) in healthy adolescents (age 9 to 23, N = 129). Functional integrity was assessed using a probabilistic reversal learning task that is known to depend on the OFC. Repeated measures ANOVAs showed significant differences between age groups until approximately age 12. After age 12, performance leveled off, even on the more difficult probabilistic blocks. When the association between pubertal stage (as determined by self-report questionnaires) and reversal learning performance was examined, pubertal stage was found to be a stronger predictor of performance than age for most reversal learning variables. Pubertal stage also interacted with gender to predict performance on some reversal learning variables.

Structural development of the whole brain was assessed using vertex-based analysis of cortical thickness derived from T1-weighted MRI scans and using voxel-based morphometry of fractional anisotropy (FA) derived from diffusion tensor imaging scans. Widespread decreases in cortical thickness and increases in FA were observed during this age range, particularly in association cortex and in major white matter pathways connecting to the frontal, parietal, and temporal lobes. Most associations with age were best fit with a linear model, but there were a few clusters in which quadratic or cubic models improved the fit. Within the age range in which reversal learning performance was developing (9 - 12) there were fewer associations between the brain variables and age than in the older adolescents (13 - 23 years), perhaps due to power limitations in the younger group.

Perseverative errors on the reversal learning task were associated with cortical thickness in the superior frontal gyrus and the superior parietal cortex and with FA within anterior corpus callosum, as well as other pathways connecting to the frontal lobe. After controlling for age, only the association between FA and perseverative errors within the anterior corpus callosum remained significant, suggesting that most of the associations were developmental in nature. A gender by pubertal stage interaction was also observed in the relationship between perseverative errors and FA within this region. Results are discussed in relation to adolescent risk-taking behavior.