Mind maps: tracking language across the brain

Like a game of mental Battleships, neurolinguists work out how we process language by matching up brain injuries and the effects they have on speech. The method has been in use...
20 February 2017

Left side of the brain

The perisylvian cortex is located on the outside of the brain, with Broca's area at the front and Wernicke's area at the back.


Neurolinguists have a habit of naming key discoveries as though they are machines out of Wallace and Gromit, or perhaps mobile phones: the P600, the N400, the ELAN. Each of these refers to an electrical impulse in the brain - one flicker as raw noise makes its way through the brain and becomes speech. The field of neurolinguistics sets out to map this process, testing the theories that linguists come up with against the often-murky image of the brain that neuroscience can muster...

Paradoxically this entire field sprang from the study of people unable to process language. It began a rather inelegant science - doctors, unable to see inside patient's heads, could only observe the broken speech that brain injuries caused. Today, these damaged brains remain one of the few ways that neurological theories can be tested.

Ten patients, two theories

The study of language deficits - aphasiology - gained traction in the nineteenth century, beginning with Pierre Paul Broca, a French doctor who treated patients with garbled speech. Aphasiology had then, as it does now, two main footholds. In the first, “deficit analysis”, neurologists examine a patient’s speech and understanding. In expressive aphasia, a patient might speak entirely without verbs, or be limited to one-word sentences. Broca’s first patient was nicknamed “Tan" because it was the only syllable he could produce.

The second approach involves identifying the areas of the brain that might be at fault. Thanks to computer imaging, this “functional localisation” has become much easier since Broca’s time, when the only way to examine a brain was an autopsy. The new technology (and better treatment for patients with brain damage, who now live longer) makes it possible for scientists to locate a lesion within the brain, observe a patient’s speech patterns as they change, and eventually narrow down specific language functions to particular areas of the brain.

Broca and his most famous successor, Carl Wernicke, had in common a certain paucity of data. Wernicke based his 1874 description of receptive aphasia - difficulties understanding language, rather than producing it - on just two patients. Broca developed his most notable theory, that of left-hemisphere dominance in language, from observing a grand total of eight. Noticing, at the time of autopsy, that all of these patients had lesions on the left sides of their brains, Broca concluded that the left hemisphere was mainly responsible for language production. Since the left hemisphere was also associated with right-handedness, he predicted that in left-handed people, language would be produced in the right half of the brain instead.

This video of stroke survivor Sarah Scott, now an advocate for teenage stroke awareness, gives an example of Broca's aphasia.

Locating language within the brain

Perhaps the least likely contributor to aphasiology, especially to testing Broca’s theories, was the Second World War. Just as the First World War is infamous for the “shell shock” it inflicted on its soldiers (now known as Post-Traumatic Stress), the second left thousands of soldiers with traumatic brain injuries, many with suffering speech deficits. There was now enough data to collect statistics on brain damage - importantly, among people who spoke diverse languages - and plot it against the speech impairments they had. Sure enough, in most patients with expressive aphasia, the brain damage was in the left hemisphere. (Interestingly, as Broca predicted, many left-handed aphasia victims had the opposite response, developing aphasia after a right-hemisphere injury - but not all.)

Another somewhat heavy-handed way of finding a patient’s language-dominant hemisphere was pioneered by Juhn Atsushi Wada, a neurologist, in 1949. In the Wada test, a barbiturate – a drug that depresses the nervous system – is injected into one hemisphere of the brain, shutting it down. The patient is then tested for their response to speech, from which it becomes clear whether the hemisphere of their brain that has been shut down is in fact responsible for language production.

Hemispheric asymmetry

From any diagram, it is pretty clear that the brain is broadly symmetrical. The right and left hemispheres are rough mirror images, but this overall similarity masks minute differences in the size and shape of certain features on each side. Neuroscientists hold these variations responsible for the “lateralization of brain function,” or the tendency for one hemisphere to dominate in specific cognitive tasks.

Research into brain asymmetry gave rise to the “right brain, left brain” myth – the popular idea, expounded by many an online quiz, that every person had a dominant hemisphere. Right-brained people were said to be artistic and creative; left-brained people were supposed to be logical and mathematical. While patently untrue (in fact, research on lateralisation reveals that the two halves are equally important and dependent on one another) the myth was based partially in fact: the left side of the brain is usually the side associated with language and mathematical thinking.

Asymmetry raises two questions. First, why do some people develop language dominance in the right hemisphere, and some in the left? And second, since the human brain can vary so dramatically from person to person, how can scientists generalise about where language function is located?

Where does this leave linguists?

Given that it took 120 years of aphasiology to confirm Broca’s cerebral language dominance theory (with some corrections), it might seem unlikely many sub-tasks of language will be pinned down to one location anytime soon. Indeed, neurolinguists disagree over whether further pinpointing is even possible.

Most linguists, theoretical and otherwise, base their work on the idea of Universal Grammar – the inborn structure in the brain that allows humans to learn a language. Universal Grammar has several prerequisites, including a “language acquisition device” (the brain structures all children are born with that allow them to learn a language). According to advocates of Universal Grammar, most famously the linguist Noam Chomsky, humans must also have a mechanism for interpreting syntax, the way a sentence is structured to communicate meaning.

Studies of aphasia show that the syntax device, at least, is necessary. When given a sentence like “The fish ate the man,” aphasia patients often understand it as “the man ate the fish,” or don’t make sense of the sentence at all. This is known as agrammatism, and occurs in expressive aphasia as well as receptive. The patient uses a combination of the vocabulary they can glean (fish, ate, and man) and common sense (a fish is unlikely to be eating a man) to understand the sentence.

Some neuroscientists have suggested that these patients have damage to a syntax-specific area of their brain; others insist that the brain parses sentences in various areas. At any rate, it appears that the structures of Universal Grammar, neat enough on paper, are nowhere near as organised in the brain as they are in books.

Anatomy of a sentence

Wernicke’s and Broca’s areas are both located in an area called the perisylvian cortex, which aphasiologists hold broadly responsible for many aspects of language. However, while the perisylvian cortex is widely thought to be important in language processing, many neurolinguists do not think it can be divided further into linguistic areas.

A 2012 study led by Evelina Fedorenko used fMRI (functional Magnetic Resonance Imaging) to track blood flow to Broca’s area, and found that certain sections of it were used solely for language, but others were activated by non-language tasks as well. Several of these sections lay side-by-side. Yet Fedorenko later co-authored a paper proposing that in fact, it was impossible to accurately map the perisylvian region by function. The paper suggested that any attempt to divide the region into language-specific and “domain-general” relied too much on bad statistics: that is, that people’s brains vary so widely that no average location can accurately describe them.

None of these studies suggest that the perisylvian cortex is the only linguistic area of the brain: aphasiologists have found that damage to the thalamus, among other areas, is associated with severe problems in speaking and understanding. These results are difficult to verify, though, because a stroke that damages the thalamus is very likely to affect nearby areas of the brain, as well.

For this reason, aphasiology is an indispensable but blunt tool in neurolinguistics. Aphasic patients and their brains provide vital data for neuroscientists trying to understand the brain’s relationship with language. At the same time, the human brain, and the damage it sustains, is so diverse that it often remains a mystery.


Add a comment