language and the brain -- 5/21/25

21 May 2025

language and the brain -- 5/21/25

0 COMMENTS

Today's selection--from The Language Puzzle by Steven Mithen. Modern brain imaging has revealed that language is processed across a widespread neural network. This challenges earlier views that language was confined to just a few specific brain areas.

“Since the 1990s we have had a range of sophisticated techniques for identifying which parts of the brain are associated with which bodily and cognitive processes, including language. Most notable is fMRI (functional magnetic resonance imaging) scanning. This detects the locations of changes in blood flow in the brain of subjects when they perform designated tasks, with increased flows indicating brain activity. Other techniques include electroencephalography (EEG) and magnetoencephalography that rely on detecting electric signals in the brain and their magnetic fields respectively. Before the development of such imaging methods, we were reliant on discovering how the brain works only after parts of it broke down.

“During the mid-nineteenth century, the French anatomist Paul Broca undertook post-mortem studies of the brains of people who had suffered aphasia — the loss of all or some aspects of speech. He concluded their disorders had arisen from damage to an area of the frontal lobe that became known as Broca' s area. This is also known as Brodmann areas 44 and 45, Korbinian Brodmann being a German neuropsychiatrist who gave each area of the brain a specific number. Inspired by Broca's work, the German anatomist Carl Wernicke identified another area of the brain that impacted on language, one located within the temporal lobe. When damaged, comprehension rather than the production of speech becomes disrupted.

“Broca's and Wernicke's areas are both in the left hemisphere and were later found to be connected by a bundle of nerve fibres called the arcuate fasciculus. These dominated our understanding and further research into the neuroanatomy of language until the 1990s, encouraging a view that language is entirely located in the left hemisphere of the brain. Even before the advent of fMRI, however, the three-component system for language provided by Broca's area, Wernicke's area and the arcuate fasciculus had been questioned. Why was it that some patients with aphasia were found to have undisturbed Broca's areas? Conversely, why did those with damaged Broca's areas not necessarily suffer a loss of speech function?

“The answer was provided by fMRI and other brain-imaging techniques that showed language is distributed throughout the brain rather than being located in a small number of specific areas. Moreover, by developing a closer association with linguistic science, neuroscientists have begun to identify where specific aspects of language are processed. Broca's and Wernicke's areas remain important, with the former identified as critical to processing hierarchical phrase structures, such as embedded clauses. The cerebellum is recognised as playing key roles regarding language perception, processing and production, with a notable contribution to overall verbal fluency. Although the basal ganglia do not appear to be involved in basic linguistic functions, when damaged or diseased they can have a negative impact on language capability. The connection between Broca's area in the frontal lobe and Wernicke's area in the temporal lobe remains critical. But rather than the single-fibre tract of the arcuate fasciculus, two pathways are now recognised, a ventral pathway supporting how we connect the sounds of words to what they mean and a dorsal pathway responsible for speech production.

Language areas of the brain.

“The global network view of language is exemplified by semantic mapping. This is a method to identify where the brain stores and processes words or, more accurately, the concepts attached to the sounds designated as words. This research began by using fMRI to identify which areas of the cortex are activated when subjects hear different types of words, such as those relating to objects, actions and social narratives. It found these were processed at different locations in the brain, but predominantly in the left hemisphere. Further work derived a 'semantic atlas' for the brain. This stretches across both hemispheres, with some preference for narrative-related words on the right. Rather than being stored at single cortical locations, abstract concepts are represented by spatially distributed networks, thought to reflect the open-ended nature of such concepts. To quote Yizhen Zhang, the lead scientist, the takehome message from semantic mapping is that 'the human brain encodes a continuous semantic space' .This exemplifies the new understanding of how language in the brain relies on extensively distributed neural networks, extending into all four lobes, both hemispheres, the cerebellum, and most likely every anatomical part of the brain.

“Two other key lessons have been learned. The first is that no area of the brain appears to be entirely dedicated to language — they are all multifunctional. Broca's area, for instance, consists of at least ten sub-regions, supporting both language and non-language roles, including complex hand movements. While the cerebellum has a significant role in language processing, it also has a controlling role for balance, movement, motor skills and memory.

“Second, there is considerable variability in the neural networks for language found in different people. While this can have a genetic cause and lead to language dysfunction, some people develop different networks for no apparent reason and with no impact on their linguistic ability, such as having a greater prominence of their language-significant neural networks in the right hemisphere. The language that one speaks, signs or reads also influences the neural networks that develop: a study comparing Italian and English found that the words of these languages influence the extent to which areas of the temporal gyrus and frontal gyrus are activated when reading, probably because English words have a more complicated mapping of letters to sounds. Such studies have mainly been undertaken using spoken English or other languages of the Indo-European family, and hence share many linguistic features. We have no idea how the neuroanatomy of language would vary for a language that has a much greater reliance on word morphology than English, such as indigenous Australian and North American languages with their extremely long words. How would the brain respond to a language that lacks word order entirely?

“Only a few studies have been undertaken to explore the brains of people who speak more than one language. Bilingual and multilingual people have been found to have a greater quantity of grey matter in frontal regions of their brains and of white matter that provides connections between different brain regions. Rather than switching between languages, their brains keep all their languages activated at the same time, although how this occurs remains unknown.”