Neurogenomics of speech, language & reading disorders 30-01-2025
Session 1 – introduction to the language sciences, with an overview of biological
approaches for studying speech, language and reading. Characteristics, diagnostics
and prevalence of relevant neurodevelopmental disorders.
Why we study speech and language in the first place.
FOXP2 is an transcription factor we will talk a lot about.
In exam: a question about each of the sessions, marked 1 to 10 and then your point is the
mean of these points. You will only be asked about things mentioned in the sessions, not
that is only in the papers.
Papers for this session:
Deriziotis P, Fisher SE. Speech and Language: Translating the Genome. Trends Genet.
2017 33(9):642-656.
Fitch WT. The evolution of speech: a comparative review. Trends Cogn Sci. 2000
Jul;4(7):258-267.
Graham SA, Fisher SE. Decoding the genetics of speech and language. Curr Opin
Neurobiol. 2013 Feb;23(1):43-51
Introduction
Why should geneticists be interested in language and speech at all?
Darwin: “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; no child has an instinctive
tendency to brew, bake, or write.”
Captures general idea of an instinct to acquire a skill; intertwining of nature and nurture. You
have the instinct towards language but you have to be exposed to language as well to be
able to develop it.
Beyond the babbling of babies to experiment with the vocal system in the first few years of
life a lot of amazing things happen. A child is able to start to acquire the sound system and
able to acquire a vocabulary of thousands of words. They can also take these elements and
put them together using grammatical rules and put them together in limitless number of
different words and sentences.
Another thing that a child picks up in the first few years of life is moving the muscles of your
face, tongue, jaw and do a sort of dance and push air back and forwards which sends out a
signal that others can pick up. You get a stream of sound and others can reverse engineer
what the words are in that stream. We actually interpret the sounds and know what this
sounds stream means.
You have all of these elements (cognitice, physiological) that come together within the first
few years of life and they don’t have to be taught. You only need people talking around you.
Maybe there is something inside us that helps us to do this → genes?
How do these things work?
,You can take different words and sequence them. Noam Chomsky came many years ago
out with the concept that it is not the order of the words that is crucial, there is something
deeper about linguistic structure. You can take 5 words and it does not have to be
meaningful or grammatical. But if you flip them you get a phrase which is grammatical it still
doesn’t mean anything. Your brain doesn’t need a meaning to figure out the syntax. His idea
was that we had a language acquisition device that does syntax. He thinks that the fact that
your brain can do this is in your genetics.
The reason that you can feel the grammaticality of that sentence is not
just because you know about the order of words, but because you are
sensitive to the structural features of it. You can draw this structure.
Overall there is a sentence (S), there is a noun phrase (NP) and a verb
phrase (VP). The noun phrase has 3 elements, one is the noun (N).
Then you have the verb (V) and de adverb (Adv). The thing is you can
feel this structure. One of the key things that humans do in language is
merging pieces together, so he got the idea that there is an operation in your brain that can
take two linguistic structures and merch them together and then again put them in another
one. This can create this infinite long complicated sentences.
Another way to realize this is that it is not the word order that is
important in these examples. That is shown by this example where
you have the same sequence of words depending on the structure
underneath it can mean different things. You can have the same
order of words but with different structure underneath it and thus
another meaning.
That is just a tiny flavor of syntax. Those examples were of
English and there are a lot of different languages.
Chomsky’s claim was that even though the specifics of the
language might differ, there might be different
vocabularies, word orders, ways of marking syntax and so
on, they all have the same deep structure. They all rely on
the fact of merging these elements together. The idea is
that kids are discovering the specifics of their language but
they are helped by the fact that they have inbuild these
kind of principle ideas of how syntax works.
There are a lot of different ideas of how syntax works. Morphology is another part of
linguistics which is related to syntax, but it’s the idea that within words you can have parts
that can change their functions in syntax. For example plurals, I have one dog then with two
you have to put an s on the end. If we zoom in more there is phonology. This is the individual
sounds that are making the words. In all languages these different things come together to
make words.
Semantics is the idea that you want to convey a message, so the meaning. Pragmatics is
how the meanings are modified by context and speech. You can use the tones that you are
using to change the message.
,There is a lot of complicated stuff going on, but you don’t exactly teach the child…
The other thing that is fascinating beyond the actual linguistic content of language is how we
can put it out in the world. The most common form of language in the world is
spoken language. How does this work?
• Air exhaled from lungs drives oscillations of vocal ‘cords’ in larynx, rate of
oscillation determines pitch
• Acoustic energy passes through vocal tract (pharyngeal, oral & nasal
cavities)
and out to environment via nostrils/lips
• Filtering: formants modify sound, only some frequencies get through
• Formants determined by length/shape of vocal tract; modified by moving
articulators (tongue, lips, soft palate...)
• ‘beet’, ‘boot’, ‘bought’ & ‘bat’ differ in formant structure
It is a mysterious system. The idea is that to produce vocalizations, we have to have 2
things. The source for the sound to come out, the larynx and the frequency in which they are
vibrating determines the pitch of the sound that is coming out. We can whisper as well, so
then there is not that much of pitch. Something else is happening in our vocal tracks which is
that the source that produces the sound then the vibrations get modified by the vocal tract
which has a certain shape. What happens is that you have a source and a window and some
frequencies are letting trough but not all frequencies are getting through. This gives a kind of
formants pattern of frequencies of sound. You can change which formants there are, even if
the pitch stays exactly the same, if you change the shape of your vocal tract (tongue, lips)
that will change the formants that can get through. That is what makes formants sound
different. Anyone who tries to learn a second language where there are other formants than
in their own language, to learn this new is quite hard but children pick this up.
• Fine rapid movements of articulators (tongue, lips, jaw etc.) closely synchronised with
each other & vibrations of larynx
• e.g. ‘pat’ versus ‘bat’ depends on when larynx begins vibrating relative to vocal tract
movements
• Tiny difference in timing can change phonemes
This also implies some clever stuff that we do with the
coordination with our different muscles. Pat and bat are different
because of when the larynx starts vibrating, it’s 10 ms delay
before the larynx begins vibrating. The question is what happens
if you have a 50 ms delay → experiments.
Segmenting speech
We do amazing dances with our mouth to produce these phonemes that we can interpret.
There is a continuous space between having some ms delays. But we categorize in
language of the perception so we can cluster these phonemes in our heads and are able to
know the difference between ‘pat’ and ‘bad’.
What people realized when they tried to
figure out how does a child learn where a
word starts and ends. In written language,
there is a space between the words and
dots, etc. What do we do if we listen to a
continuous stream of sounds? They realized
, that if you look for gaps or regions, so if you slize it up at the points where you might think
there is a break. But when you look at the words, it doesn’t match at all. But now there is a
really good idea of the amplitude envelope within speech that we can use to break up words.
But it’s not that the children have to figure this stuff out. How are we able to do that?
Neural systems supporting language
We use the speech articulating system, we use the auditory system and in the brain there
are different neurosystems that are supporting language.
Sensory systems because we have to be able to
receive signals to be able to get the language input in
the first place. If you take a 4 day old newborn they
can already discriminate speech sounds into
categories and are sensitive to speech rhythms. They
can separate sound which is language versus
something else. They have been inside the womb for
9 months (fetal development influences?).
The motoric aspects we have to make a dance to
produce these sounds. Humans are good at making a diverse phenetic repertoire. We can
modify the shape of our vocal tract in vary flexible ways, we can produce a lot of
differentiation in our phenomes. There is a very large number of muscles that you are using
when you are speaking.
Cognitive elements are the combinatorial nature in language. We learn a number of words
and are able to put them all together in different maturances. We don’t know how important
this is in real language, but working memory becomes important.
These aspects are interlinked. A lot of this stuff is connected.
Decoding neural basis of language
Now we have tools where you can put someone in a brain scan and see what parts of the
brain are active when doing certain tasks, you can also look at the blood flow when doing a
task. People started studying the neurobasics of language in the 1800s, how?
They did brain lesion studies. The people started to think
of the brain in a modular way, that different parts of the
brain do different stuff. In the mid 1800s a physiologists
was the founder of this idea of mapping bits of brain
based on the kinds of disorders that people were having.
The most common thing is that people had
neurodegenerative disorders (stroke) that would damage
a certain part of the brain. A man was studied in his 30s
he started to loss the ability to speak and all he could
say was ‘tan’. When he died he was studied and you see
a lot of damage as a hole. They noticed that this was on
one side of the brain and in a certain part of the brain. You see a second patient which looks
a little bit different but the same kind of region. These are the two primary patients that led to
this definition of Brokes area, which is the left inferior frontal gyrus which is really important
for aspects of speech production and integration of linguistic elements.
In fact it’s not exactly this area.
Session 1 – introduction to the language sciences, with an overview of biological
approaches for studying speech, language and reading. Characteristics, diagnostics
and prevalence of relevant neurodevelopmental disorders.
Why we study speech and language in the first place.
FOXP2 is an transcription factor we will talk a lot about.
In exam: a question about each of the sessions, marked 1 to 10 and then your point is the
mean of these points. You will only be asked about things mentioned in the sessions, not
that is only in the papers.
Papers for this session:
Deriziotis P, Fisher SE. Speech and Language: Translating the Genome. Trends Genet.
2017 33(9):642-656.
Fitch WT. The evolution of speech: a comparative review. Trends Cogn Sci. 2000
Jul;4(7):258-267.
Graham SA, Fisher SE. Decoding the genetics of speech and language. Curr Opin
Neurobiol. 2013 Feb;23(1):43-51
Introduction
Why should geneticists be interested in language and speech at all?
Darwin: “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; no child has an instinctive
tendency to brew, bake, or write.”
Captures general idea of an instinct to acquire a skill; intertwining of nature and nurture. You
have the instinct towards language but you have to be exposed to language as well to be
able to develop it.
Beyond the babbling of babies to experiment with the vocal system in the first few years of
life a lot of amazing things happen. A child is able to start to acquire the sound system and
able to acquire a vocabulary of thousands of words. They can also take these elements and
put them together using grammatical rules and put them together in limitless number of
different words and sentences.
Another thing that a child picks up in the first few years of life is moving the muscles of your
face, tongue, jaw and do a sort of dance and push air back and forwards which sends out a
signal that others can pick up. You get a stream of sound and others can reverse engineer
what the words are in that stream. We actually interpret the sounds and know what this
sounds stream means.
You have all of these elements (cognitice, physiological) that come together within the first
few years of life and they don’t have to be taught. You only need people talking around you.
Maybe there is something inside us that helps us to do this → genes?
How do these things work?
,You can take different words and sequence them. Noam Chomsky came many years ago
out with the concept that it is not the order of the words that is crucial, there is something
deeper about linguistic structure. You can take 5 words and it does not have to be
meaningful or grammatical. But if you flip them you get a phrase which is grammatical it still
doesn’t mean anything. Your brain doesn’t need a meaning to figure out the syntax. His idea
was that we had a language acquisition device that does syntax. He thinks that the fact that
your brain can do this is in your genetics.
The reason that you can feel the grammaticality of that sentence is not
just because you know about the order of words, but because you are
sensitive to the structural features of it. You can draw this structure.
Overall there is a sentence (S), there is a noun phrase (NP) and a verb
phrase (VP). The noun phrase has 3 elements, one is the noun (N).
Then you have the verb (V) and de adverb (Adv). The thing is you can
feel this structure. One of the key things that humans do in language is
merging pieces together, so he got the idea that there is an operation in your brain that can
take two linguistic structures and merch them together and then again put them in another
one. This can create this infinite long complicated sentences.
Another way to realize this is that it is not the word order that is
important in these examples. That is shown by this example where
you have the same sequence of words depending on the structure
underneath it can mean different things. You can have the same
order of words but with different structure underneath it and thus
another meaning.
That is just a tiny flavor of syntax. Those examples were of
English and there are a lot of different languages.
Chomsky’s claim was that even though the specifics of the
language might differ, there might be different
vocabularies, word orders, ways of marking syntax and so
on, they all have the same deep structure. They all rely on
the fact of merging these elements together. The idea is
that kids are discovering the specifics of their language but
they are helped by the fact that they have inbuild these
kind of principle ideas of how syntax works.
There are a lot of different ideas of how syntax works. Morphology is another part of
linguistics which is related to syntax, but it’s the idea that within words you can have parts
that can change their functions in syntax. For example plurals, I have one dog then with two
you have to put an s on the end. If we zoom in more there is phonology. This is the individual
sounds that are making the words. In all languages these different things come together to
make words.
Semantics is the idea that you want to convey a message, so the meaning. Pragmatics is
how the meanings are modified by context and speech. You can use the tones that you are
using to change the message.
,There is a lot of complicated stuff going on, but you don’t exactly teach the child…
The other thing that is fascinating beyond the actual linguistic content of language is how we
can put it out in the world. The most common form of language in the world is
spoken language. How does this work?
• Air exhaled from lungs drives oscillations of vocal ‘cords’ in larynx, rate of
oscillation determines pitch
• Acoustic energy passes through vocal tract (pharyngeal, oral & nasal
cavities)
and out to environment via nostrils/lips
• Filtering: formants modify sound, only some frequencies get through
• Formants determined by length/shape of vocal tract; modified by moving
articulators (tongue, lips, soft palate...)
• ‘beet’, ‘boot’, ‘bought’ & ‘bat’ differ in formant structure
It is a mysterious system. The idea is that to produce vocalizations, we have to have 2
things. The source for the sound to come out, the larynx and the frequency in which they are
vibrating determines the pitch of the sound that is coming out. We can whisper as well, so
then there is not that much of pitch. Something else is happening in our vocal tracks which is
that the source that produces the sound then the vibrations get modified by the vocal tract
which has a certain shape. What happens is that you have a source and a window and some
frequencies are letting trough but not all frequencies are getting through. This gives a kind of
formants pattern of frequencies of sound. You can change which formants there are, even if
the pitch stays exactly the same, if you change the shape of your vocal tract (tongue, lips)
that will change the formants that can get through. That is what makes formants sound
different. Anyone who tries to learn a second language where there are other formants than
in their own language, to learn this new is quite hard but children pick this up.
• Fine rapid movements of articulators (tongue, lips, jaw etc.) closely synchronised with
each other & vibrations of larynx
• e.g. ‘pat’ versus ‘bat’ depends on when larynx begins vibrating relative to vocal tract
movements
• Tiny difference in timing can change phonemes
This also implies some clever stuff that we do with the
coordination with our different muscles. Pat and bat are different
because of when the larynx starts vibrating, it’s 10 ms delay
before the larynx begins vibrating. The question is what happens
if you have a 50 ms delay → experiments.
Segmenting speech
We do amazing dances with our mouth to produce these phonemes that we can interpret.
There is a continuous space between having some ms delays. But we categorize in
language of the perception so we can cluster these phonemes in our heads and are able to
know the difference between ‘pat’ and ‘bad’.
What people realized when they tried to
figure out how does a child learn where a
word starts and ends. In written language,
there is a space between the words and
dots, etc. What do we do if we listen to a
continuous stream of sounds? They realized
, that if you look for gaps or regions, so if you slize it up at the points where you might think
there is a break. But when you look at the words, it doesn’t match at all. But now there is a
really good idea of the amplitude envelope within speech that we can use to break up words.
But it’s not that the children have to figure this stuff out. How are we able to do that?
Neural systems supporting language
We use the speech articulating system, we use the auditory system and in the brain there
are different neurosystems that are supporting language.
Sensory systems because we have to be able to
receive signals to be able to get the language input in
the first place. If you take a 4 day old newborn they
can already discriminate speech sounds into
categories and are sensitive to speech rhythms. They
can separate sound which is language versus
something else. They have been inside the womb for
9 months (fetal development influences?).
The motoric aspects we have to make a dance to
produce these sounds. Humans are good at making a diverse phenetic repertoire. We can
modify the shape of our vocal tract in vary flexible ways, we can produce a lot of
differentiation in our phenomes. There is a very large number of muscles that you are using
when you are speaking.
Cognitive elements are the combinatorial nature in language. We learn a number of words
and are able to put them all together in different maturances. We don’t know how important
this is in real language, but working memory becomes important.
These aspects are interlinked. A lot of this stuff is connected.
Decoding neural basis of language
Now we have tools where you can put someone in a brain scan and see what parts of the
brain are active when doing certain tasks, you can also look at the blood flow when doing a
task. People started studying the neurobasics of language in the 1800s, how?
They did brain lesion studies. The people started to think
of the brain in a modular way, that different parts of the
brain do different stuff. In the mid 1800s a physiologists
was the founder of this idea of mapping bits of brain
based on the kinds of disorders that people were having.
The most common thing is that people had
neurodegenerative disorders (stroke) that would damage
a certain part of the brain. A man was studied in his 30s
he started to loss the ability to speak and all he could
say was ‘tan’. When he died he was studied and you see
a lot of damage as a hole. They noticed that this was on
one side of the brain and in a certain part of the brain. You see a second patient which looks
a little bit different but the same kind of region. These are the two primary patients that led to
this definition of Brokes area, which is the left inferior frontal gyrus which is really important
for aspects of speech production and integration of linguistic elements.
In fact it’s not exactly this area.
