Claire Corlett

Fish Food, Fish Tanks, and More
Extraordinary Variations of the Human Mind: Simon Fisher: Language at the Extremes

Extraordinary Variations of the Human Mind: Simon Fisher: Language at the Extremes


– [Narrator] We are the paradoxical ape. Bipedal, naked, large-brained, long the master of fire,
tools, and language, but still trying to understand ourselves. Aware that death is inevitable, yet filled with optimism. We grow up slowly, we hand down knowledge. We empathize and deceive. We shape the future from our shared understanding of the past. CARTA brings together experts
from diverse disciplines to exchange insights on who
we are and how we got here. An exploration made
possible by the generosity of humans like you. (upbeat thoughtful music) – So why is a geneticist interested in something like language,
something that we learn? To answer this question,
I’m gonna hand over to somebody who’s much
more eloquent than I am, and it’s this fellow here, Charles Darwin. Almost 150 years ago he really summed up what’s amazing and
remarkable about language. He said, “Language is an
art, like brewing or baking. “It certainly is not a true instinct, “for every language has to be learnt. “It differs widely from all ordinary arts, “for man has an instinctive
tendency to speak, “as we see in the babble
of our young children, “while no child has an
instinctive tendency (audience laughs) “to brew, bake, or write.” This really beautifully captures something strange about language. And let’s dig into this a little bit more. By the time a child is
only a few years of age, they’ve already assembled a vocabulary of thousands of words. They can assemble these
words into a potentially limitless number of meaningful sentences using grammatical rules, and
these meaningful sentences can relate not only to the present, but also the past, the future,
even abstract concepts. And then something that we
take completely for granted, but is really extraordinary, is the ability that a child learns to take the thoughts that are
sitting inside their heads and convert them into streams of sound by the most incredible
feat of motor control, the articulatory functions moving the muscles that control
the face, the jaw, the larynx, in this rapid kind of dance of motor function that converts thoughts into streams of sound. And then another child’s
ears can do an amazing thing, which is to kind of
reverse engineer, decode, this stream of sound, reverse engineer it, figure out what the original thought was, and come to their own conclusions. (audience laughs) All these different things that happen in the first few years of life, a child can develop this suite of skills without needing any
kind of formal tuition. So in a way that seems
very, almost magical. And it’s a complicated
set of different things that go towards being able
to become a proficient speech and language user. So how do we explain this? And people have speculated
that there might be something important in our genes. Now importantly here, this is actually a great example, a perfect example, of interaction between genes and environment. And this is obvious to you
as soon as you think about the fact that a child who grows up in the Netherlands surrounded
by Dutch speakers will learn to speak Dutch. That same child growing up in Japan, surrounded by Japanese speakers, will learn to speak Japanese. And a child who is not exposed to language will never become a language user. So there’s this incredible
interaction between genetics and environment. But there’s something in
the genes that seems to predispose us to soak up language from the environment
around us in early life. Now not every child, unfortunately, when exposed to language,
becomes as proficient as this young lady here, for example. So there are some kids who
are exposed to language and have language-rich environments, but they fail to become
proficient language users. And in some cases there
might be a reason for this like they have a physical
problem, they’re deaf, or they have some kind of
general cognitive problem. And in some cases there are actually kids that fail to learn language
and it’s a mystery, and we don’t know why. It was noticed very early on when people started studying these kind of disorders that these cases clustered in families. So if you have a relative with a problem with development of speech and language, that dramatically increases your risk. Now this kind of familial
clustering could in fact relate to something
like shared environment, for example, in a family. But we know from studies
of twins that actually there’s a very very high heritability for these kind of speech
and language disorders, unexplained speech and language disorders. We know this by looking
at the concordance in the rates of language
disorders in identical twins as compared to non-identical twins. There’s been many many studies
that robustly show this. What I want to talk to you about in the remainder of the
talk is the idea that we should be able to go
further than just saying well, there’s genes involved, to actually being able to say can we take some people who have speech
and language problems and figure out what the actual genes are, the particular genes, that are important? One of the wonderful
things about being able to find the genes that are involved is that we can use these as kind of windows, molecular windows, into
the intermediate biology. And this is my favorite
slide that I always show. Everybody, I think, has
a version of this slide if they work on genes and brain behavior. There’s this huge gap between DNA and the behavioral outputs
that we’re interested in. And we’re not so naive to ignore that gap, but we think that we can use
the knowledge of the genetics to actually fill in, and
understand, each different level. So we can understand that
many genes that we study are important because
they code for proteins, and the proteins kind of all work together as molecular machinery in your cells to do all the functions of your cells. We know that genes and
proteins are important for different properties in
the way that neurons develop. So, for example, they might influence the proliferation of neurons
in early development. They might influence how neurons migrate to the final positions in the brain, how those neurons
differentiate and extend out kind of their axons to
connect up and wire up with other neurons in the brain. And then we might want to understand how neural circuits actually work. One of the most fascinating things about the interaction between
genes and neurocircuitry is that your genes are important for the functions of your neurocircuits, even in your adult brain,
because they help you to learn. Genes are actively working
to strengthen and weaken the links, the synapses, the links between your different brains
cells throughout life. And of course we have
these complex assemblages of neural circuits,
built by a combination of genes and experience, and
it’s those in the brain that are doing these kind
of complex behaviors. And the idea here is that if we can find something at this end of
the spectrum, the DNA level, and we know that it’s
linked at this other end to speech and language, we can use that as a tool for understanding
all these different levels. And there’s interactions and relationships between the levels. This is a gross oversimplification, but it’s a starting
point for thinking about what I’m gonna talk about. So I’m gonna give you an example where we’ve been able to identify a gene that’s important in speech
and language development. And the starting point for this study was this family here, the KE family. It’s three-generation family,
now a four-generation family, and in each of these
generations you can see these shaded individuals are affected with a severe speech and language disorder. And they’re growing up together with these unaffected individuals
who are non-shaded, who are growing up
together with their kids who develop a severe problem. The severe problem is that
they have difficulties learning to make those
coordinated sequences of speech. They do actually become speech users but there’s always a problem. I’m gonna now play you an
example, hopefully if this works, from one of the adult
members of the KE family, from the second generation, and she’s being asked by Kate Watkins, a PhD student at the time, she
now leads a group in Oxford, but at the time she was
a PhD student in London with Faraneh Vargha-Khadem,
studying this family. She asks the lady to repeat
different words five times. And she’s going to ask them to repeat the word catastrophe, so listen to this. – [Kate] Catastrophe. – [Woman] Cas tas, ca tas
to fee, ca trustef, mm. (woman groans as Kate laughs) – [Kate] Catastrophe. – [Woman] Catastof. (groans) – [Kate] It’s hard, isn’t it? – [Woman] It’s really a tongue twister. – So she’s having great trouble repeating the word catastrophe. That’s not a comment on
the political situation at the moment, but.
(audience laughs) I promised I wouldn’t be political. (audience laughs) Anyway, so this trait that
she has is something that, it’s called childhood apraxia
of speech but actually, as you can hear, it
persisted into adulthood. And even despite some
intensive speech therapy, people still suffer from these problems. The problem is that they have problems stringing together certain
sequences of sound. The problems get worse as
the utterances get longer and as they become more complex. This is shown by, this is
a study that Kate Watkins carried out on the KE family
where they’re being asked to repeat either simple
or complex nonsense words. And these are nonsense
words like contramponist and perplisterock, words
that now I practice every day in front of the mirror so I’m very good at them.
(audience laughs) But these people have
never encountered the words and they try to repeat them back, and they have difficulty doing so. What you can see on these graphs is, these are unaffected
members of the KE family, these are affected
members of the KE family, and this third line is
actually adults with Broca’s and other kinds of aphasias, brain damage, that yields
problems with speech praxis. Here, as you can see, the
material that gives them the most difficulty is when they have to repeat these nonsense words that are really long and complex. It’s not that they can’t
say certain sounds, it’s that they have
particular problems with stringing them together in the right way. Something to do with
the way that the brain is programming sequences of mouth and face movements during speech. Now they also have lots of other problems with all sorts of aspects of language. They have deficits not
just in spoken language, but also if you ask them to
write linguistic items as well. Their impairment’s not just in expression but also in reception of language, and there are tests of grammar
comprehension and production that they do worse on than
unaffected individuals. This is not a kind of general
intellectual disability. They don’t all have a
kind of IQ that’s in the intellectual disability range. But this family is not the
brightest family in the world so there are some unaffected
individuals in the family who have low non-verbal IQ, and there are affected individuals who
have normal non-verbal IQ and severe speech and language problems. So it doesn’t seem like
the non-verbal problems that some of the family
members suffer from, it doesn’t seem like they’re a core feature of the disorder. If you look at patterns of
non-verbal and verbal cognition, different kinds of
sub-tests, you find that they have deficits that
are much more severe in the verbal range, which
is what you would expect for a speech and language disorder. So working in the lab of Tony
Monaco, in the late 1990s, we started screening
the DNA from this family and we pinpointed a linked
region on chromosome seven. This was actually a different region from the region that
you’ve just heard about in the Williams syndrome talk. Eventually we zoomed in and spotted a single tiny change, one letter
of DNA, one base of DNA, which normally in every
unaffected person in this family and in every one of you
I can guarantee, was a G. In the affected members of the family, they have two copies of every gene. One copy had the G, one copy was normal. But one copy has an A instead of a G. This was private mutation
that had occurred in the grandma of that
family and then been passed on to half of the affected members. And the unaffected members
did not receive this variant. This was in a gene called FOXP2. We could already say something
about what that gene was. It’s a gene that regulates other genes, I’ll tell you about that in a minute. And all these genes like FOXP2, they have something called
a forkhead box domain. They make a protein that
has this special domain here that I’m showing you, and it consists of these three helices here,
one, two, and three, and these big loops here. And the KE family have
this mutation that changes one of the amino acids in the protein that they build for FOXP2,
and it’s this amino acid here. And it stops the protein
from doing its job properly, it stops the protein from binding to DNA. Since finding this
mutation in the KE family, we and others have found a bunch of other different mutations in FOXP2 in different cases around the world. This just shows you a
picture from a later study showing about eight or
nine different mutations that have been found, disturbing different
letters of the FOXP2 gene. In each case, they damage
one copy of the FOXP2 gene but leave another copy
of the FOXP2 gene intact. So these people have
kind of an insufficient amount of FOXP2 protein
that they’re making and they have speech
and language problems. Sometimes other problems in addition, but the kind of most common feature across all these patients is speech
and language disorder. We’ve been able to take
the different mutations that disrupt FOXP2 and
look at them in the lab. We can grow cells in the lab
and label FOXP2 in green. Here is the nucleus of cells, and here is normal FOXP2 being expressed in the nucleus of these cells
which is where it likes to be, and here are some examples
of mutations where the protein can no longer
get into the nucleus. We do other assays that show that FOXP2 is not working properly in these cells. So the idea then is that you can use FOXP2 to study all these intermediate levels from DNA to speech and language. And I refer you to these papers
to find out more about it. But I’m just gonna give
you a couple of examples, because I’m running out of time, of the kinds of things that
we’ve been studying with FOXP2. So FOXP2 is a regulatory gene that switches on and off other
genes, and this is how it works. It gets transcribed, it’s a messenger RNA, and this messenger RNA
is used as a template to build a protein also called FOXP2. This FOXP2 protein binds target genes and it binds to the promoters,
the regulatory regions at the front of each gene, and
it could either activate them or it could repress them,
it could silence them. And we think that a lot of
the time what FOXP2 is doing is to silence these genes, but it can also act as an activator. It seems to do this like a kind of genetic dimmer switch if you like,
it’s not all or nothing. We can then have a look, we
know that we have this gene, can we ask what are the other elements of this FOXP2 network that it belongs to? So we might ask what are the things that bind to the FOXP2 gene to switch it on? What are the signaling processes that interact with FOXP2 to
modulate its function? What are the other
factors that bind FOXP2? And what are the downstream targets that it’s switching on and off? We’ve identified quite a
few downstream targets, and many people have been
working on this over the years. What’s interesting is
that many of the targets that FOXP2 regulates are known to be important for neurodevelopment
because when they go wrong they cause neurodevelopmental
disorders like autism, schizophrenia,
epilepsy, and so on. and we’ve identified
pathways like sumoylation that modulate the way the
FOXP2 works in the cell, and also other proteins like TBR1 that interact with FOXP2
and are also implicated in neurodevelopmental syndromes. We can go further to ask what
is the influence of FOXP2 on the neuronal properties? One of the neuronal properties that FOXP2 appears to influence is
the outgrowth of neurites from cells, from neurons,
and these are the things that will eventually connect
up, become axons and dendrites, and connect up with other neurons. When FOXP2 is lost, it can lead to shorter neurites with reduced branching. You might ask then, if
FOXP2 is so important for things like the outgrowth of neurites, when we have a loss of
FOXP2 why don’t we have a global problem with the
whole of our brains? The reason for this is that FOXP2 is not actually switched on
itself all over the brain. It’s switched on in
certain subsets of cells. This is, if I took your brain
which I promise I won’t do, (audience laughs)
and I slice it in half, you can see this is a
kind of cross section. FOXP2 is expressed in the
deepest layers of the cortex, especially in motor
regions, it’s expressed in layer six and layer five. It’s also expressed in the basal ganglia, in the striatum, in caudate,
putamen, the thalamus, and in the cerebellum it’s expressed in a very specific cell type
called the Purkinje cells. What a lot of these circuits do, the cortico-basal ganglia and
cortico-cerebellar circuits is they’re important for you
to do motor skill learning, to learn to make motor sequences. Since FOXP2 is not a new human gene, it’s actually present in
all sorts of other organisms in evolutionary history, we
can study what FOXP2 does say by looking in a mouse. When we take a mouse, we
can study the way that these neurons that express
FOXP2, the ways they fire during motor skill learning. We can thus kind of get
insights into what’s happening. We found that in these different places in the striatum, also in
Purkinje cells in the cerebellum, that the firing of the neurons
during motor skill learning and motor skill tasks is different from when FOXP2 is mutated
from unaffected animals. So I’m gonna finish
there by just summing up. FOXP2 at the extremes,
what does this tell us about speech and language development? Heterozygous mutations
in FOXP2 cause a rare severe speech and language
disorder in humans. It’s a regulatory gene, so its
targets and its interactors give us entry points into neural pathways. It’s not a gene for speech
because as I’ve told you, versions of FOXP2 are
active in the brains of many different vertebrate species. The nice thing there
is that we can actually study what FOXP2 does
in these other species. By studying humans,
mice, and even songbirds which I didn’t have time to talk about, we can uncover the roles of the gene in these kind of circuits
like cortico-basal ganglia and cerebellar circuitry
that are known to be important for complex motor functions. A curious question would be, we know that when FOXP2 is damaged it impairs your speech-motor skills, so it
would be interesting to see whether, at the other end of the spectrum, for example in people
who are really good at beat boxing, or rapping, maybe they have (audience laughs)
interesting variants of FOXP2. So the last thing I’d like to say is that the fact that FOXP2 has
been around for a long time in evolutionary history,
and it’s been doing important things in the brain, means that whatever it’s doing in
speech and language in humans is not novel and de novo, but
it’s built on ancient stuff. So I think this suggests to us that our unique capacity for
acquiring spoken language is built on systems that
are evolutionarily ancient. And there I will stop,
thank you very much. (audience applauds) (upbeat thoughtful music)

Leave a Reply

Your email address will not be published. Required fields are marked *