In their Review article earlier this year, Fedorenko, Ivanova & Regev (Fedorenko, E., Ivanova, A. A. & Regev, T. I. The language network as a natural kind within the broader landscape of the human brain. Nat. Rev. Neurosci. 25, 289–312 (2024))1 propose a functional separation between the core language network and other perceptual, motor and higher-level cognitive components of communication-related networks in the left hemisphere of the human brain. In the ‘Open questions and a way forward’1 section that ends their Review, the authors discuss the need for cross-species comparative research to disentangle how these brain networks came to support human language. Here, we suggest that the authors’ functional separation of a core language network and other components in the human brain is grounded in the evolution of two separate structural networks within primate brains.

Fedorenko and colleagues describe the core language network as left-lateralized, and involving the middle frontal gyrus (MFG), inferior frontal gyrus (IFG), superior temporal gyrus (STG) and middle temporal gyrus (MTG). Perceptual and motor systems for speech are defined as separate subsystems located in auditory cortex and speech perception areas in the STG and motor cortex and motor planning areas1, respectively. Importantly, these functionally defined key brain areas are known to be structurally connected via dorsally and ventrally located white-matter fibre tracts, which guarantee the information flow between areas. In humans, two separate dorsal pathways that provide structural connections have been identified for two distinct networks2,3 (Fig. 1). A perceptual–motor network connects Brodmann area (BA) 6 in the premotor cortex to the auditory cortex in the STG, and a core language network connects cytoarchitectonically defined BA44 in the IFG with the STG and MTG of the posterior temporal cortex via the arcuate fascicle. The former allows information flow from speech perception areas to production areas, and the latter allows linguistic information to flow within the core language network.

The thickness of the arrows indicates the strength of the fibre connection. The arrowheads indicate that the information flow is bidirectional. The numbers indicate the cytoarchitectonically defined Brodmann areas (BA). Perceptual–motor connections in the premotor cortex (BA6) to auditory cortex in the STG (blue) and between the IFG (BA44) and STG via the arcuate fascicle (AF) (red) are present in macaques, chimpanzees and humans. Only humans have the AF fibre tracts expanded to connect with the MTG (pink), an area within the core language network. Thus, this change to those white-matter fibre tracts might be a key evolutionary change in the development of human language.

Cross-species analyses revealed that these two dorsal structural networks have different evolutionary trajectories, both in the target regions of the fibre tracts and their laterality4,5,6 (Fig. 1). During evolution, the target region BA44 in the left IFG expanded in humans (as compared to chimpanzees) and has changed its function partly from a motor and communicative-related region to a language-related region4. Crucially, the arcuate fascicle that connects the IFG with the temporal cortex underwent major changes during primate evolution, and shows a lateralized extension beyond the STG reaching to the MTG only in humans7,8. In summary, the white matter connection from the premotor cortex to the STG as part of the perceptual–motor network is present in macaques, chimpanzees and humans5,7,8, whereas the connection from the IFG to the temporal cortex has extended to reach the MTG as part of the core language network that is present only in humans6,7,8, which suggests that this extension might be of particular relevance for human language (Fig. 1).

On the basis of these findings, we suggest that the theoretically predicted9,10 and empirically demonstrated1,2 functional separation between a core language network and a perceptual–motor network observed in humans finds its grounding in the structural evolution of the primate brain.

There is a reply to this letter by Fedorenko, E., Ivanova, A. A. & Regev, T. I. Nat. Rev. Neurosci. https://doi.org/10.1038/s41583-024-00899-7 (2024).