A Change Vector Method to Study Behavioral Development




Cooper, David L.

Journal Title

Journal ISSN

Volume Title



Broca‘s Area was the first region in the human cortex to be tied definitively to a specific behavior—language. However, structural, cytological and molecular peculiarities identified in Broca‘s Area are not unique to humans, and thus language appears to have emerged from other traits that were advantageous in the evolution of primates in general, such as fine motor control for gestures and vocalizations, and the so-called mirror system. One potential source of insight into the emergence of language is to study the correlation of brain structures with behavioral function. This work capitalizes on the existence of a unique resource to undertake that study: eight detailed cytological studies of the developing human cortex from birth through six years of age, accomplished by JL Conel from 1939 to 1967. Conel‘s atlases provide a consistent methodology applied to 37 cortical areas at each of the observations ages for neurons and 42 cortical areas for myelinated fibers, which further enables a quantitative comparison of change patterns during human cortical development. The change vector method that was developed to conduct that investigation normalizes measures for the change steps that occur in the seven change intervals that occur in the Conel data, measures the statistical significance of any of those change steps, and permits the direct comparison of change trajectories using k-means cluster analysis. This analysis reveals significantly correlated synchronized changes at different ages, linked to specific ―core‖ area/layer addresses that imply a clock-like coordination mechanism that appears to support sensorimotor developmental functions at the appropriate age. Neither functional cores nor statistically significant change steps emerge when a similar analysis is applied to verbal behavior, where such a clock-like mechanism is unlikely. In general for the cortex, and for language behavior related to Broca‘s Area in particular, the analysis supports innate architectural mechanisms that facilitate specific address level accommodation to external activity, whereas external behavioral evidence, as from language change, merely supports rapid evolution of the behavior itself to accommodate those same cortical mechanisms. That is, language evolves to enable the speaker to speak, just as writing evolves to enable the child to read. This dissertation research has complied with all George Mason University standards for the ethical conduct of research and for the appropriate use of human and animal subjects.



Neuron, K-means cluster analysis, Cortex, Maximal clique, Brodmann Area, Cortical layer