Grounding language processing on basic neurophysiological principles

In animal models the neural basis of cognitive and executive processes has been studied extensively at various hierarchical levels from microcircuits to distributed functional networks. This work already provides compelling evidence that diverse cognitive functions are based on similar basic neuronal mechanisms. More recent data suggest that even cognitive functions realized only in human brains rely on these canonical neuronal mechanisms. Here we argue that language, like other cognitive functions, depends on distributed computations in specialized cortical areas forming large-scale dynamic networks and examine to what extent empirical results support this view.