The corpus callosum (CC) is the primary connection between cerebral hemispheres, allowing for the interhemispheric integration of sensory, motor, and cognitive processes. The CC is composed of densely packed, myelinated axons and is commonly partitioned into regions based on anatomical and functional connectivity with cortical regions. It contains interhemispheric projections that terminate in cortical layer IV (Schmahmann & Pandya, 2006
). The structure and function of the CC has been studied extensively in callosotomy patients, through post-mortem studies, and through the use of in vivo brain imaging. Despite the vast literature regarding its structure and function, little has been reported on the development of the CC during childhood and adolescence. Although available reports are somewhat inconsistent, both cross-sectional and longitudinal studies (described below) suggest that the CC continues to mature structurally from infancy to adulthood. Whether these structural changes support behavioral changes has not been extensively studied.
Conventional magnetic resonance imaging (MRI) methods have yielded morphological information about the CC in relation to age, gender and pathology. In a large study of children ages 2–15 and adults ages 16–79 years, Allen et al. (1991)
showed through midsagittal tracings of the CC that its area increased with age, reached a plateau in the group of children then decreased steadily with advancing age in the adults. The study also revealed gender differences in the morphology of the CC, with females exhibiting a bulbous shaped splenium compared to what was described as a more tubular shaped splenium in males. Pujol et al. (1993)
conducted a longitudinal study of individuals 11–61 years old and found increases in CC area up to roughly the third decade of life. Males and females did not differ in total CC area, however males did show an increased rate of CC development compared to females. Regional distinctions in CC development have also been reported. In a large longitudinal study of 139 individuals ages 4–18, Giedd et al. (1999)
examined CC size and found evidence of CC development through adolescence, even after correcting for total cerebral volume. The study showed a significant age-related increase in area of the posterior portion of the CC, especially in the splenium. The authors interpreted this pattern as possibly being indicative of an anterior-to-posterior gradient in CC development, where anterior regions were presumed to have reached adult sizes earlier in development. Similarly, Thompson et al. (2000)
showed sharp increases in area of the posterior portion of the CC (isthmus and splenium) with age in normally developing children and adolescents from ages 6–15 years. This group also reported that children ages 3–6 years showed marked increases in anterior CC area, which, in part, corroborates the findings and hypotheses of Giedd et al. (1999)
. In contrast, Paus et al. (1999)
studied individuals ages 4–17 using modified segmentation output from structural MRI scans and did not report any age-related differences in CC white matter density.
Conventional imaging methods can assess developmental trends in brain tissue macrostructure during childhood and adolescence, but they cannot provide specific information about microstructure, such as axonal organization and orientation. Diffusion tensor imaging (DTI) is an in vivo approach to examining white matter microstructure that has demonstrated sensitivity to both developmental and degenerative age-related changes in tissue integrity (Barnea-Goraly et al., 2005
; Bonekamp et al., 2007
; Sullivan & Pfefferbaum, 2006
; Wozniak & Lim, 2006
). DTI provides information about the magnitude and direction of water diffusion within tissue (Basser & Pierpaoli, 1996
). Myelin, a lipid and protein rich axonal covering, restricts water diffusion in white matter. Intracellular water within myelinated axons diffuses in a more directional manner compared to diffusion in unmyelinated axons. Mean diffusivity (MD) and fractional anisotropy (FA) are two commonly derived scalar measures of DTI. MD describes the degree of diffusion in all directions, whereas FA describes the directional portion of diffusion. High MD corresponds to relatively unimpeded water diffusion and indicates regions of low tissue organization, while high FA corresponds to preferential diffusion along one direction, indicating a high level of tissue organization. Developmental studies utilizing DTI have shown age-related changes in microstructural development in the CC using both region of interest (ROI) and voxel-based analyses (Ashtari et al., 2007
; Barnea-Goraly et al., 2005
; Bonekamp et al., 2007
; Li & Noseworthy, 2002
; Snook et al., 2007
). Li & Noseworthy (2002)
revealed increases in FA and volume in the splenium with age in a sample of healthy 10–40 year-olds. Also, histograms that were generated suggested that development peaked at some point in the second decade of life and then declined with increasing age. In a study of forty individuals ages 5–19, Bonekamp et al. (2007)
showed age-related decreases in the apparent diffusion coefficient of the splenium, but not the genu. Barnea-Goraly (2005)
reported age-related increases in FA and white matter density in the body of the CC using voxel based analyses. More recently, Ashtari et al. (2007)
studied twenty-four healthy males, ages 10–20 and found that the older males had significantly higher FA in the splenium compared to younger males.
Although the imaging methods discussed thus far reveal a great deal about the CC’s structural maturation, they lack direct information about how these structural refinements relate to the behavioral changes that accompany normal development. Numerous neuropsychological methods have been developed to assess the level of function in various brain regions, including the CC. A fundamental role of the CC, interhemispheric transfer of information, can be assessed in a number of different ways, from simple finger tapping exercises to complex visual hemifield stimulation. It has been suggested that there is both a clear increase in CC utilization throughout development and also increased utilization in children only, based on these behavioral measures (Banich et al., 2000
; Marion et al., 2003
). It has not been customary to examine the relationship between task performance and brain microstructure in normal development, however some studies have examined these associations in older individuals (Baird et al., 2005
; Johansen-Berg et al., 2007
; Roebuck et al., 2002
; Sullivan et al., 2001
). For example, Sullivan et al. (2001)
reported correlations between FA and alternating finger tapping performance in a cross-sectional analysis of healthy adults. The group also reported regression analyses in which task performance was predicted by FA and age. In a study of ten healthy adults, Johansen-Berg et al. (2007)
found relationships between bimanual coordination and FA of the CC midbody.
The current study uses DTI to investigate white matter changes with age using a cross sectional cohort of preadolescents, adolescents, and young adults. Regions of interest (ROI) were used to evaluate changes in CC white matter microstructure with age and whether these changes were associated with improved motor skills. Gender differences were examined in both task performance and DTI measures. Finally, regression analyses were performed to determine whether task performance could be predicted by age, gender and DTI values that showed age-related maturational changes. It was hypothesized that FA in multiple regions of the CC would correlate positively with age and that finger tapping performance would also improve with increased age. We also expected FA to correlate with bimanual finger tapping performance. We anticipated that MD would show similar, inverse relationships with age and task performance.