The emergence of language in the human mind and brain—Insights from the neurobiology of language, thought and action.

Abstract

The capacity for language has evolved remarkably quickly in recent human history. Its advent likely coincided with a range of cognitive innovations not found elsewhere at this level of complexity in the rest of the animal kingdom. This late yet near-simultaneous florescence of higher language and cognition is difficult to account for in terms of strictly modular neurocognitive systems, each with its own dedicated function and evolutionary trajectory. Nor does it legitimize the neurocognitive study of language in isolation from other systems of human thought and action. In the wake of emergentist approaches to key human cognitive abilities, the present article considers language as the differentiated product of multiple neural networks dedicated to qualitatively distinct cognitive functions-including (at least) the cortical systems of general semantic cognition and control and the sensorimotor systems supporting language production. A model is proposed to account for how these systems congregate to produce language, featuring a dualstream architecture of the semantic interface into item-based and item-independent semantic knowledge on the one hand, and a notion of the sensorimotor interface as a key component for the temporal tracking and verbal rehearsal of task-relevant information on the other. Avenues are also offered for enriching this architecture in future versions of the model. Finally, it is proposed that language is an "optimal" combination of these neurocognitive systems, enabling fast and cost-effective transfer of information at the systems level. This last point underpins evidence for the privileged status of language as a tool for adaptive thought and behavior, as well as some important features of brain evolution, development, and functional organization.