This Is Your Brain On Words
What happens after you pick up a book, or pull up some text on your phone?
What occurs between the written words hitting your eyes and your brain understanding what they represent?
Scientists are trying to better understand how the brain processes written information—and how a primate brain that evolved to make sense of twisty branches and forking streams adapted to comprehend a written alphabet.
Researchers used electrodes implanted in the brains of patients being evaluated for epilepsy treatment to study what parts of the brain were involved when those patients read words and sentences. They found that two different parts of the brain are activated, and interact in different ways when you read a simple list of unrelated words, compared to when you encounter a series of words that builds up a more complex idea.
Dr. Nitin Tandon, a professor of neurosurgery at UTHealth Houston and one of the authors of a report on the work published in the Proceedings of the National Academy of Sciences, joins guest host Sophie Bushwick to talk about the study, and what scientists are learning about how the brain allows us to read.
Invest in quality science journalism by making a donation to Science Friday.
Dr. Nitin Tandon is a professor of Neurosurgery at UTHealth Houston in Houston, Texas.
SOPHIE BUSHWICK: This is Science Friday. I’m Sophie Bushwick. Here’s a question, what happens right after you pick up a book or pull up some text on your phone? What goes on between the words hitting your eyes and your brain understanding what they represent? And is that process different when you look at a story versus a series of unrelated words, like those in a grocery list?
Just how the brain processes written information isn’t fully understood, but work published this month in the proceedings of the National Academy of Sciences might offer some clues. Joining me now to talk about that is one of the authors of that report. Dr. Nitin Tandon is a professor of neurosurgery at UTHealth Houston Welcome to Science Friday.
NITIN TANDON: Thank you so much, Sophie, for having me.
SOPHIE BUSHWICK: Reading seems like a pretty straightforward thing. So why don’t we know how it works?
NITIN TANDON: Reading happens at such a rapid pace that we have not in the past had tools to study it effectively. People have used various brain imaging techniques, but they lack the time resolution or the space resolution to be able to understand it well. Reading is the fastest input system to our brain. So the average reader reads around 300 words a minute, and a speed reader can read as many as 1,500 words per minute. And as you can imagine, that’s an extremely rapid transmission of information.
Up to about 500 years ago, almost no one on our planet could read. And if you consider that here we are a primate brain capable of understanding the shapes of branches and twigs and forks in the river, how do we then take this enormously complex alphabetical system that exists today and build that upon these brains that haven’t done this in millennia? So we found that to be a particularly puzzling and important question to ask, and that has really been the focus of our research over the last decade or so.
SOPHIE BUSHWICK: And in your recent study, you worked with people with epilepsy. Why were they key to this research?
NITIN TANDON: So as we’ve just discussed, reading is a very rapid process. And to study a process of this speed, the best modality that we actually have is electrodes placed in the brain. In many individuals with epilepsy, the site where seizures start is not exactly well defined, and the standard is for us to implant tiny probes to understand and to create a network of where the seizures begin and how they propagate.
So the opportunity that we’ve been given in individuals who have electrodes implanted in their brain to pinpoint where seizures begin provides us data with very high spatial temporal characteristics that can enable us to really delve into the reading processes.
SOPHIE BUSHWICK: So using those existing electrodes, you could monitor what parts of the brain were activated during different tasks?
NITIN TANDON: Exactly. And so while these individuals are in our epilepsy monitoring unit waiting for seizures to happen, they have several days on hand that they’re sitting around and bored. And this is the opportunity that we use to give them various materials to read and study their brain activity as they read.
SOPHIE BUSHWICK: And what did you find?
NITIN TANDON: So we were interested specifically in understanding what the difference in the brain was that allowed us to memorize or to learn strings of words, put together strings of words that were meaningless words that still try to communicate some global concepts and then actual sentences. And we found that the regions that do this are very closely related to each other. One is in the temporal lobe, just above your left ear, and one is in the frontal lobe, just next to the area of the brain that allows us to speak.
And these two zones communicate with each other in different ways when single words are being read and being understood and in a different more complex way when sentence or phrase level information is being transmitted. And so specifically, when there’s a sentence that is building up on meaning, so I could say something like, “Jack took his black cat to the vet.” So Jack took his black cat is one phrase in that sentence. And at that point, when that phrase ends, there is a spike in activity that amalgamate that chunk of information together.
A lot of that has to do with word frequency. Meaning that words that have lower frequency, less common words, are the ones that on which the brain binds information together. And then these chunks of information get stitched together to create the meaning of the sentence.
SOPHIE BUSHWICK: Just a reminder that you’re listening to Science Friday from WNYC Studios. I’m Sophie Bushwick talking with Dr. Nitin Tandon about the brain and reading. What about silent reading versus reading something aloud. Does the brain respond differently?
NITIN TANDON: Yeah, it does. And so this entire experiment was people reading silently, but if people read out aloud, two things happen. The first is that you read slower because your speech production rate is only so fast. And the second is that your auditory systems are much, much more online as opposed to just your reading systems. This is, of course, how we all learn to read originally. And one of our goals here is to elaborate why some individuals cannot read well.
So dyslexia, or the inability to read, affects about 10% of all people. And they either cannot read it all or the reading is very slow. And what we are trying to do here is to create a map of how normal reading operates so that we can extrapolate what is likely going wrong in people who cannot read well.
SOPHIE BUSHWICK: Aside from the special patients with these electrodes, what do you need to make this type of experiment work?
NITIN TANDON: It takes a really special group and dedicated group of researchers, and I’m privileged to have postdoctoral Fellows like Oscar Woolnough and Elliot Murphy and a collaborator across the Atlantic in France, Stan Dahaene, who’ve all been critical for this work to happen. And of course, none of this could happen without funding. And the NIH BRAIN Initiative has really launched a whole bunch of fantastic explorations of the neurobiology of language and of other systems in the human brain all across the country. And I’d like to credit them as well.
SOPHIE BUSHWICK: What do you still want to know? What is the next question you want to explore about the way the brain processes reading?
NITIN TANDON: There are three or four interesting new directions we are going. And one is trying to see if we can teach adults new words. And one of our experiments is giving them exposure to very uncommon English words. English has about 350,000 words in it, but most people’s vocabulary is only a few tens of thousands of words. So we’re trying to see what happens when people learn new words and what that process looks like over an interval of about 5 to 7 days while they are with us in our epilepsy units waiting for seizures to happen.
Another interesting question for us is trying to understand if we can give people symbols that they’ve never been exposed to before and have them learn a whole new alphabet. And so for this, we are using Gaelic runes from the United Kingdom.
SOPHIE BUSHWICK: And does this tell us anything about how we made that leap you talked about from–
NITIN TANDON: Oh, yes.
SOPHIE BUSHWICK: –in order to read in the first place?
NITIN TANDON: Yeah. So what we’ve learned, and this is an earlier paper that we’ve published, is that the same regions of your brain that are very interested in find patterns, so for example, distinguishing between two human faces or two geometric shapes, are very, very closely aligned with and likely overlapping with the areas that enable us to read.
There are two patches of brain regions in the base of the brain next to the occipital lobe that are critical for this. One is called the lateral occipitotemporal region, or the classic visual word form area, and the other is the mid fusiform cortex. And the interaction between these two areas is critical for us to take a shape and understand it.
And so this is just that whether you have A that is written in italics or in Gothic font or is written on the side of a building 10 stories high, it is still the letter A. And so this invariant representation of a symbol lives in these two areas, and both of them appear to be quite critical for us to read letters.
SOPHIE BUSHWICK: Well, we wish you luck in decoding all of these brain areas and what they do. Dr. Nitin Tandon is a professor of neurosurgery at UTHealth Houston. Thank you for taking the time to talk with me today.
NITIN TANDON: Thank you so much, Sophie. It’s been a pleasure.