Brain structure and speech impairments

Some people, without doubt, have the gift of the gab. They can almost speak in poetry, it seems. Others, on the other hand, struggle and find language a chore. Hitherto, we've tended to give individuals affected like this labels such as "dyslexia"; and indeed dyslexia can cause some language difficulties. What we haven't had was objective data on why this is happening to some people. Until now, at least. Because by using newer brain imaging techniques that can quantify the numbers of cells and their white matter axonal connections in different parts of the brain, as she explains to Chris Smith, Saloni Krishnan, at Royal Holloway, University of London, has found reproducible differences in the density of the myelin that insulates nerve connections, in regions of the brain concerned with executing things in the correct sequences and at the right times. These are key aspects, of course, of stringing a sentence together…

Saloni - I'm really fascinated by spoken language and how our brains kind of evolved to have spoken language, because I'm really interested in children who struggle to speak; to learn the words in their native language; they might struggle with aspects of the grammar; they might struggle to kind of tell you a story. Many of these children you probably think of as, say, dyslexic, as half of the children with dyslexia actually end up having language disorders as well. And I'm really interested in kind of what's going on in their brains. Surprisingly, actually, we know quite a lot about what happens and say autistic brains, or the brains of people with ADHD, but despite the kind of fundamental importance of communication, we actually know very little about what might be the kind of systems we need to kind of configure a talking brain, if you will.

Chris - So how did you attack trying to understand more about the process of development? Because there are several things sort of wired together in this, aren't there? There's the kind of how the human brain puts itself together and then how it acquires information from the environment to do things. And that's really what language involves...

Saloni - When we talk about language, in the adult brain, we kind of know what systems we use to speak language. There's been beautiful work from a number of labs that has really given us insight into how we process speech around us, how we actually speak and so on. And one thing that really pops out there is that the left hemisphere particularly areas in like the frontal bit of the brain and the temporal bit of the brain are very important. One thing that's really interesting though is that if children have a stroke really early on in life, like as they're being born, we don't find that they have the same kind of debilitating language difficulties that an adult would have. There's this kind of paradox that sort of set up, which is, we know these systems are really important in adulthood, but they don't seem to have the same importance or the same configuration in infancy or childhood. Kind of relating to that, we have this big population, so about 7% of children or about two children in every classroom, who fit that profile of children who have spoken language problems that I was talking about earlier. And given that when we just look at brain scans from these children - like a regular MRI scan - we actually can't see any differences with the naked eyes. So it's not like they've got a little bit of their brain missing or there's something really obvious. So whatever's happening is happening at a much more subtle level and that's the kind of thing we're trying to understand.

Chris - And what did you do?

Saloni - We actually used this really exciting new technique. So most people have probably seen a picture of an MRI scan and if you get an MRI scan of the brain, it's, it's fundamentally really, really exciting. But these brain scans generally tend to be quite simplistic in that you can can see grey matter, which is sort of neuronal bodies the, the kind of bits that we think are doing the kind of calculations and computations that we need. And then there's white matter which is like axons of the neurons. So kind of making up the connections between different parts. However, the kind of meaning of those numbers is like it's not meaningful. You're just saying, "is that a bit more grey, or is that a bit more white?" and that's how we've been making conclusions. There's been a new technique that kind of really flips this on its head, which basically means that you can put someone in an MRI scanner in London, you can put them in an MRI scanner in Oxford and you could actually get the same numbers. And we call this quantitative MRI. And the reason this is really powerful and exciting is that it means that instead of just talking about a bit more grey and a bit more white, we're actually talking about quantitative hard numbers. So we recently used this protocol to study children who have this profile of spoken language impairments and compare them to children who don't have any spoken language impairments. And one of these things that the quantitative scans can really tell us about is myelin. And what myelin tends to do is to make information transfer between neurons much faster. And so, generally, as you grow older, you get kind of more myelinated neurons. And here we were kind of able to look at, in children with language disorder and those without, what were the differences in kind of myelination across their brains,

Chris - What were the differences and what did that reveal?

Saloni - We've had a hypothesis for a little while that children with language problems, they really struggle with making things habitual, or sequencing things. Because, inherently, what you have to do in order to learn language is actually sequence sounds together. So, for example, let's take a word you might have never heard before, like "geckizeissacad", and you suddenly have to hold onto those sounds. And the order of those sounds is also particularly important. And as you do it over time, you just say right there are these patterns in language and they become almost kind of default. And you do this of course with grammar as well where you might have things like, you know "walk, walked" and you kind of have all of these configuration and rules. And again, the sequence there is very, very important. So we've had this idea that perhaps in children with language disorders, the parts of the brain that might be really responsible for learning those kind of habits or sequences might be different. Using this new technique, what we were able to show is that myelin in these regions that might have to deal with kind of sequencing, habit formation, and so on seems to be reduced in children with language disorder.

Chris - Do you think that that's persistent, or is there an opportunity to intervene?

Saloni - Well, we don't know yet. And I think one of the difficult things about these kind of correlational studies almost where we say it, this group has this and this group doesn't have it, is we also don't know if this is a cause or a consequence, right? So it could be - because the children we worked with were between 10 to 15 - so it could be that a consequence of having a language problem for a long time is that you don't develop the same kind of myelination profile. And if it's a consequence then it's not necessarily a target for intervention. However, if it's a cause it could be an interesting target for intervention. So basically we need more research to fine tune that.

Chris - Overall, what do you think the implications of this are? It's a very useful observation. It gives us some, some numbers now as well. But what can we do with this data? What's the next logical step for this?

Saloni - From a research standpoint? I think I've kind of hinted that this already we need to establish if this is a cause or a consequence. And the way we could do that is by having longitudinal studies where we would follow up children to say like, what's happening when you had these low levels of myelin, and we saw you say three or four years later what happened to you? Another way we could establish this is by, you know, doing training studies. So for example, if we gave people lots of language training, what might happen and then we've got some really interesting ideas about kind of genetics and how that also might be influencing these kind of myelination patterns.