Grounding the neurobiology of language in first principles: The necessity of non-language-centric explanations for language comprehension

This article reviews a free-energy formulation that advances Helmholtz's agenda to find principles of brain function based on conservation laws and neuronal energy. It rests on advances in statistical physics, theoretical biology and machine learning to explain a remarkable range of facts about brain structure and function. We could have just scratched the surface of what this formulation offers; for example, it is becoming clear that the Bayesian brain is just one facet of the free-energy principle and that perception is an inevitable consequence of active exchange with the environment. Furthermore, one can see easily how constructs like memory, attention, value, reinforcement and salience might disclose their simple relationships within this framework.

The advent of functional neuroimaging has allowed tremendous advances in our understanding of brain-language relationships, in addition to generating substantial empirical data on this subject in the form of thousands of activation peak coordinates reported in a decade of language studies. We performed a large-scale meta-analysis of this literature, aimed at defining the composition of the phonological, semantic, and sentence processing networks in the frontal, temporal, and inferior parietal regions of the left cerebral hemisphere. For each of these language components, activation peaks issued from relevant component-specific contrasts were submitted to a spatial clustering algorithm, which gathered activation peaks on the basis of their relative distance in the MNI space. From a sample of 730 activation peaks extracted from 129 scientific reports selected among 260, we isolated 30 activation clusters, defining the functional fields constituting three distributed networks of frontal and temporal areas and revealing the functional organization of the left hemisphere for language. The functional role of each activation cluster is discussed based on the nature of the tasks in which it was involved. This meta-analysis sheds light on several contemporary issues, notably on the fine-scale functional architecture of the inferior frontal gyrus for phonological and semantic processing, the evidence for an elementary audio-motor loop involved in both comprehension and production of syllables including the primary auditory areas and the motor mouth area, evidence of areas of overlap between phonological and semantic processing, in particular at the location of the selective human voice area that was the seat of partial overlap of the three language components, the evidence of a cortical area in the pars opercularis of the inferior frontal gyrus dedicated to syntactic processing and in the posterior part of the superior temporal gyrus a region selectively activated by sentence and text processing, and the hypothesis that different working memory perception-actions loops are identifiable for the different language components. These results argue for large-scale architecture networks rather than modular organization of language in the left hemisphere.

Currently, production and comprehension are regarded as quite distinct in accounts of language processing. In rejecting this dichotomy, we instead assert that producing and understanding are interwoven, and that this interweaving is what enables people to predict themselves and each other. We start by noting that production and comprehension are forms of action and action perception. We then consider the evidence for interweaving in action, action perception, and joint action, and explain such evidence in terms of prediction. Specifically, we assume that actors construct forward models of their actions before they execute those actions, and that perceivers of others' actions covertly imitate those actions, then construct forward models of those actions. We use these accounts of action, action perception, and joint action to develop accounts of production, comprehension, and interactive language. Importantly, they incorporate well-defined levels of linguistic representation (such as semantics, syntax, and phonology). We show (a) how speakers and comprehenders use covert imitation and forward modeling to make predictions at these levels of representation, (b) how they interweave production and comprehension processes, and (c) how they use these predictions to monitor the upcoming utterances. We show how these accounts explain a range of behavioral and neuroscientific data on language processing and discuss some of the implications of our proposal.