Localizing age-related changes in brain structure between childhood and adolescence using statistical parametric mapping.
Volumetric studies have consistently shown reductions in cerebral gray matter volume between childhood and adolescence, with the most dramatic changes occurring in the more dorsal cortices of the frontal and parietal lobes. The purpose of this study was to examine the spatial location of these changes employing methods typical of functional imaging studies. T1-weighted structural MRI data (1.2 mm) were analyzed for nine normally developing children and nine normal adolescents. Validity and reliability of the tissue segmentation protocol were assessed as part of several preprocessing analyses prior to statistical parametric mapping (SPM). Using SPM96, a simple contrast of average gray matter differences between the two age groups revealed 57 significant clusters (SPM[Z] height threshold, P<0.001, extent threshold 50, uncorrected). The pattern and distribution of differences were consistent with earlier findings from the volumetric assessment of the same subjects. Specifically, more differences were observed in dorsal frontal and parietal regions with relatively few differences observed in cortices of the temporal and occipital lobes. Permutation tests were conducted to assess the overall significance of the gray matter differences and validity of the parametric maps. Twenty SPMs were created with subjects randomly assigned to groups. None of the random SPMs approached the number of significant clusters observed in the age difference SPM (mean number of significant clusters = 5.8). The age effects observed appear to result from regions that consistently segment as gray matter in the younger group and consistently segment as white matter in the older group. The utility of these methods for localizing relatively subtle structural changes that occur between childhood and adolescence has not previously been examined.