fMRI Study of Normal Language Development in Children


In the same treatise in which he states that his historical findings "compel us to formally advance the principle of functional localization as an irrefutable rule", Korbinian Brodmann clarifies that "One cannot think of – human perception (of sound for example) – taking place in any other way than through an infinitely complex and involved interaction and cooperation of numerous elementary activities, with the simultaneous functioning of just as many cortical zones, and probably of the whole cortex…" (Brodmann, 1909 [1, p. 255]). Modern neuroimaging findings now demonstrate how well Brodmann understood his data and support the notion of brain functions as served by a distributed network.

Functional magnetic resonance imaging (fMRI) has emerged as an important non-invasive methodology in 21st century studies of language networks and functional connectivity in the human brain. Although its application in studies of child language development have been somewhat more challenging, during the first five years of this proposal, we addressed these challenges by developing child-friendly approaches for studying language processing in children and made technical advances that assure valid data is obtained and analyzed from this population. We have demonstrated that the cerebral distribution of language function changes with age, exhibiting patterns that are influenced by normal development as well as neurological disorders.

This large data set (n>300, age 5-18 years) provides the necessary power for the development and application of new statistical approaches to identify chronologically-linked regions of activation through Independent Component Analysis and model their functional connections through Linear Structural Equation Modeling techniques. This work will advance the understanding of brain-language relations beyond the static models of adult language to the dynamics represented by the period of language acquisition. This data will also serve as a frame of reference for the continuing longitudinal study of normal language development to cover a period of 11 years in children 5-18 years of age. These studies and methods hold the promise of revealing the ontogeny of both localization and connectionist models foreshadowed a century ago by Brodmann, in the unique context of the developing brain.