Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
#7 - Vitoria Piai - How we process and create language
15 Plays5 years ago
In this conversation we talk about the science of language. How does language work? What is language and how does it impact us every day? We discuss how language makes us uniquely human and where do the borders lie if there are any. We also talk about the overlaps of language and memory. Our guest is Dr. Vitoria Piai. Vitória is a senior researcher at the Donders Centre for Cognition and the Radboud University Medical Centre. Her research focuses on language function in healthy and neurological populations, such as stroke, brain tumor, epilepsy, and neurodegenerative disorders such as dementia or Parkinson’s. She pays special attention to the intersection of language and other functions, such as executive control, (semantic) memory, and motor control in the case of speaking.
Recommended
Transcript
Introduction to the Podcast
00:00:00
Speaker
Hey, what's up, everyone? This is Ajmal Savary and welcome to another podcast episode. Today we talk to Vitoria P.I. She studies the brain mechanisms of language. We discuss how language makes us uniquely human and also how it overlaps with memory. Enjoy.
00:00:30
Speaker
Hey everyone, and welcome to another podcast episode. If you're new here, my name's Ajmal. I'm a neuroscientist and entrepreneur. On this podcast, we explore the links between science, technology, business, and the impact they have on all of us. Our guest today is Dr. Vittoria Piai. Vittoria is a senior researcher at the Donder Center for Cognition and the Radbot University Medical Center.
Dr. Vittoria Piai's Research Focus
00:00:54
Speaker
Her research focuses on language function healthy and neurological populations such as stroke, brain tumor, epilepsy and neurodegenerative disorders such as dementia or Parkinson's. She pays special attention to the intersection of language and other functions such as executive control, memory, like semantic memory and motor control in the case of speaking.
00:01:16
Speaker
She uses a bidirectional approach, where she uses models from cognitive neuroscience to better understand language function in the neurological populations, while at the same time using observations of the breakdown of language in those populations as unique insights to improve the models from cognitive neuroscience.
Awards and Recognitions
00:01:32
Speaker
For her work, she has received multiple grants, amongst which are the Rubicon grant, the Dutch Vini grant, and the Junior Principal Investigator premium.
00:01:41
Speaker
Her awards range from getting the early career award three times from three different research societies, the top review award and the award for the top 10% of review contributions to the field of psychology.
The Science of Language
00:01:54
Speaker
The list goes on and on. In this conversation, we talk about the science of language. How does language work? What is language and how does it impact us every day? We discuss how language makes us uniquely human and where do the borders lie if there are any.
00:02:09
Speaker
We also talk about the overlaps of language and memory. Okay, enough background. Let's get into it, shall we? Welcome, everyone. Today, I'm joined by Vitoria. We haven't seen each other for a while. Actually, that's not true. Three weeks ago. Yeah, and before that, we hadn't seen each other for a while. Yeah, because you've been in the United States, and before that,
00:02:35
Speaker
We met actually here at the Donors Institute during my master's. Right. It was already that time. Yeah. That's some time ago. 2009, right? Yeah. 2009. Yeah. Yeah. That's how we first met. Yes. And well, today is supposed to be about the science of language. Wow. Or the specific focus that this conversation will take. We'll see. Yeah. We'll see. And you know,
Vittoria's Language Learning Journey
00:03:03
Speaker
When I did this little bit of research, I realized there was one thing about you that I never asked you because you speak a shitload of languages. And I never asked you, how come?
00:03:22
Speaker
Yeah, so I don't think it was a very deliberate thing from the start. So growing up, going to school in Brazil, you get English and Spanish. It's kind of regular and it isn't really good if you do it at school. But I've always been interested in learning.
00:03:39
Speaker
Do you get that standard from primary school already? I don't remember really. I think from maybe when we are... Oh, I could be saying something that doesn't make more sense anymore because it's been a long time, but maybe from when you were 12 or something. Oh, well, that's pretty early. Yeah, I guess. And it's usually not a very good level unless you go to more privileged schools, I guess.
00:04:04
Speaker
Um, but I went to a private, uh, school for English as well and Spanish in school. And then because we have the Italian cities and shipping, my family, I thought it didn't make much sense at all to have an Italian passport and not speak Italian. So when I was about 16, I started taking Italian lessons.
00:04:25
Speaker
So that's how the Italian came on the list. So it's not really that I said, no, I have to learn more languages. And then when I was 18, I moved to Belgium and there I learned French. And then eventually I moved to the Netherlands and then I was already in Belgium learning Dutch as well.
00:04:44
Speaker
And later, because my husband is Swedish, I learned Swedish. So I think the only languages that I really kind of forced myself are forced, but that I really deliberately made the choice, I will learn this language, was with Italian and Polish.
00:05:00
Speaker
You also know Polish. Some, yeah, not anymore as I used to, but yeah, I still speak a tiny bit. Yeah. And I found it really fascinating. Okay. I didn't even know you spoke Italian. I had no idea. I knew Swedish was on the list. Brazilian, of course, Portuguese. I assumed you speak Spanish, but I didn't know that either. Whoa.
00:05:23
Speaker
But we said you speak lots, but you had no idea. Well, I knew the number. The number, but not what they were. Exactly. But then it's not that impressive, right? When you realize it's, oh, you know, four of them are Romance languages. Oh, come on. And then English, Swedish and Dutch. Yeah, they are.
00:05:41
Speaker
so similar anyways and then all of us to me it's not impressive because these are all very you know related languages so it's not like Polish is a real challenge because vocabulary wise there is very little I can use
Strategies for Language Learning
00:05:54
Speaker
from the other languages grammar wise it's quite complex so that to me is a real challenge these other languages I don't really count as being something very special or something yeah
00:06:04
Speaker
Well, that must be just you, because I barely know people who speak just one Roman language would be like, oh, just adding another one is fine. That's not a challenge. It is for them. But I think that's great. I noticed this as well when I was learning Dutch.
00:06:24
Speaker
that at some point I could guess my way through the vocabulary of the language, just getting a feel for, okay, this word, I have to know what this word is. I don't know what it is in Dutch. Would this be maybe German origin or English origin? And then I just got very good at figuring that out. And if nobody corrected me, I just like, oh, I guess that was it.
00:06:48
Speaker
They probably understood me. Yeah, they understood me and they just keep going. That's, that's how I brute forced my Dutch. But in a way it's very nice, right? Cause it does tell you that there are some intrinsic rules in there that you are learning implicitly and then you're just using them yourselves, right? So at some point you just learned that that's how words are formed.
00:07:08
Speaker
probably there's some rule through time that, you know, German and Dutch, once they shared that common, uh, proto Germanic language, and then they came about to be as they are because of some rules or some changes over time. That's probably something that you were able to pick up on and use now as well. But in all honesty, I consciously have no idea what's happening. I'm just,
00:07:36
Speaker
To all I know, I'm just guessing my way through it. And I just realized the likelihood of me guessing right is increasing, but the underlying mechanism, no idea. Just really no idea. I noticed this with, because I speak Farsi as well. And while I was learning Dutch, the Farsi was helping me a lot in my Dutch pronunciation. Because as a German, if you were to use German pronunciation in Dutch,
00:08:06
Speaker
It sounds terrible. So to all Germans, you really need to fix that. You sound terrible in Dutch.
00:08:15
Speaker
some effort, come on, I really don't like German accent in foreign language. Might just be me, others might not care so much. But the Farsi really helped me too, especially the sound was very easy for me to create.
00:08:36
Speaker
But what was very difficult was the large amount of consonant combinations Dutch has, that the German doesn't have, where it is really this, this was hard. Really German, I thought German had, well, I think Dutch has quite some, a certain level of complexity that is even further than the German, but I think German also has some of those consonant clusters there. Yeah, exactly. I think it was just that I wasn't used to it. It's specifically this, this...
00:09:03
Speaker
Yeah. The German doesn't have that. No, exactly. And just moving from the front of your mouth to the back, and then that took me quite some time. Yeah, that's true. Yeah. That's true. And Farsi is quite regular. In Portuguese, for example, most of our syllables are just going to be one consonant, one vowel, one consonant, one vowel.
00:09:25
Speaker
So it's very kind of simple for most of the vocabulary. And that I think it's also why some, when people from Brazil speak English, they will add certain sounds to the end of words, you know, because our syllables have to follow that particular rule. I see. And then you just add that to, so that it conforms to what you're used to doing your mother tongue. So it stays in the flow, basically. Yeah. That's interesting. I was always wondering why that, why some people do that. I never,
00:09:54
Speaker
I never thought it's because of that. I just thought maybe it's just a cultural thing. No, no, no. It's definitely your mother tongue, like rules from your mother tongue that you're just, I don't know for what reason if people really realize, but you just apply them when producing a foreign language as well. But about the Farsi, I can't answer your question. I have no idea. I don't know. I never learned how to read or write. I can speak it. And if you were to ask me something specific,
00:10:22
Speaker
I would have to reverse engineer it to give you an answer, and most of the time I do that, but then it takes me like 10 minutes. So let's not do that now. Let's not do that now. I can consult Wikipedia later. Yes, maybe that's better. But wow, I think you're speaking so many languages is impressive.
00:10:43
Speaker
I understand though where you're coming from, specifically with a generalization aspect of it. And is it that what fascinates you about language so much? Is the underlying complexity which is shared? No, actually not. So when I was doing my studies,
00:11:01
Speaker
What attracted me at first to studying languages was really, I think, rules and how some of these rules are the same across languages or how languages evolve and they follow certain patterns, all of these things.
00:11:18
Speaker
But then I quite soon noticed when studying linguistics that I kept asking the question, but how do people actually
Shift from Linguistics to Psycholinguistics
00:11:25
Speaker
do it? You know, you have all of these rules, but as we are speaking, we are applying all of these things and they are just coming out so fast and so naturally. How do speakers actually do it? How do we actually find the word we want to produce? So I noticed that that was the question. Those were the questions that really fascinated me, how it is that I come to a word and I get that word out.
00:11:47
Speaker
And that's what made me go more towards what we call psycholinguistics rather than linguistics, right? Linguistics is more descriptive. What are the rules that allow you to combine words or combine sounds, et cetera, or put words in a sentence?
00:12:05
Speaker
where psycholinguistics is more about what is really going on in the head of the speaker, the psychology behind language, how we use it, how we comprehend it, et cetera. So those are the questions that fascinate me more so than the rules. Even though I find the rules really cool, I think ultimately it's about how we use the rules themselves or the words. Yeah. No, it's fascinating that you say that because I just remember my linguistics classes in horror.
00:12:37
Speaker
I just, I loved language back then and I thought this would be, oh, this would be cool. And then they just dried, for me, they dried the entire, the greatness of the topic out. And I was just sitting there thinking, oh my God, I don't think I like anything about language anymore until I by accident stumbled into a computational linguistics class. And then I could see how you could actually apply all of that thinking.
00:13:06
Speaker
But at the beginning, I was just thinking, why are you doing to this topic? Was it things like this is morphology, and these are the rules for morphology? Yeah, exactly. You had the morphology, you had the semantic trees, then you had the particles, and then you had how does this fit into that together?
00:13:28
Speaker
And which I can understand that you need some type of approach to study it. I understand that now, but back then I would have really liked for the lecturer to say that rather than to just start talking and not say what this is for. Why is it done this way? So unfortunately my
00:13:52
Speaker
passion for studying language went away. It died at the beginning, unfortunately. I have a good friend, Christian, and he was always into languages.
Subconscious Language Production
00:14:07
Speaker
And he was also a lot into computational models. So we have been looking into these things. We've been debating a lot. And I think it's interesting that you say that you
00:14:19
Speaker
Did you found the psychological aspect about this more interesting? Because I think you're right. You have all these rules, but how, so just thinking about this now myself, how come even now in this conversation, I'm not explicitly thinking about following any rules. I'm just, I'm just speaking. I'm not searching for words. I'm not doing any of those things. And yeah.
00:14:47
Speaker
How does that work? Yeah, I don't know. That's one of the big questions. It's something I guess in some ways we are doing all of that all the time, right? It's just that we are really good at applying those rules.
00:15:03
Speaker
as we need them and for certain things also it's used so much that at some point it may be that some things are stored in memory already as bigger chunks and you just use that rather than always computing everything as you need it so that's we also know that for many things they're overly used you don't need to always be applying the rules over and over again but you rather have chunks of information already stored in that particular way
00:15:31
Speaker
Do you mean like sentences or? Yeah, it could even be, yeah, kind of these more, you know, stereotypical sentences. Like one of the ways you can see this is really fascinating actually. So you have people who have brain damage and then they acquire a language disorder due to that. So they can't, some people won't be able to speak at all.
00:15:54
Speaker
anymore, they may be able to produce, I don't know, five or six words, but they can sing happy birthday and they can curse and they can produce these stereotypical. The cursing would be me. What did you say? The cursing would be me. I'm sure I could still curse.
00:16:14
Speaker
Yeah, because these things, maybe the cursing has also all of the emotional... Attachment. Yeah, rage, that makes them be slightly different.
Language Disorders and Chunked Memory
00:16:24
Speaker
But you do find these stereotypical utterances, they can still be produced, even though they cannot produce any sentence.
00:16:31
Speaker
really requires the combination of different words into a sentence, for example, right? But they can, if they are singing Happy Birthday, they can sing the entire thing. But then they cannot produce any one of those words in isolation if they would have to, if they would want to. So this does tell us something about how certain things are really stored as a chunk entirely.
00:16:54
Speaker
This reminds me of, that should say that now, it reminds me of this one patient that was also studied here at the Department of Parkinson's Research. When you have Parkinson's at a later stage, you have all these tremors and you have problems initiating movement, controlling movement. But that person, as soon as he was sitting on the bicycle,
00:17:22
Speaker
No problem. You could just cycle through like incomplete flow and with quite some movement complexity, which normal cyclists can't even do. And the, the, yeah, the physicians were, yeah, dumbfounded literally is like, what the fuck?
00:17:40
Speaker
Yeah. And do you think there's some overlap in this type of mechanism? Because it seems low level, right? I'm not saying that Parkinson's and languages is the same.
00:17:53
Speaker
Yeah, but I would say, I don't know, for biking and walking, in both cases, it's a certain pattern that is over-learned, right? Yeah, that's true. So it could be rather than the case, I don't know, was it a static bicycle? No, no, it was really on that they went to the parking because his wife said he's fine when he's cycling. And no one believed it. And then they went on the parking lot, and then he was on the bicycle. They helped him onto the bicycle, and then he started.
00:18:23
Speaker
And they just couldn't believe it. Yeah. Yeah. Wow. That's amazing. Yeah. I was also, I just, so they had a video of this. I just, I thought this was fake. Yeah. That's how it looked like because that's not what you'd normally hear of Parkinson's patients. Right. No, no, no. There's also this. And so I don't, I don't know. I don't think in this case it's really more because I would say walking is just as automatic and over-learned to think about the rules. But maybe there is something with about the emotional.
00:18:52
Speaker
Part of it makes it much more pleasant to do and then somehow it interacts, I don't know, but interacts with your motor system. I don't know. No, I have no idea. There's this video of a patient who has a very severe type of aphasia, so this language disorder, due to a stroke, and he's having a conversation.
00:19:17
Speaker
The person asks him a question like, oh, so whatever your children. And then he's saying tono, tono, tono, tono, tono, tono, tono, tono, tono, tono, because that's the only word he can say. But there's like prosages intact. He's really having a whole conversation only saying tono, tono, tono, tono, tono.
00:19:33
Speaker
And then the person asks him like, how many children do you have? And then he shows four. And then she says, can you count that up for me? And he's like, tono, tono, tono, one, two, three, four, five, six, seven, eight, nine, 10, 11, 12, tono, tono, tono, tono, tono, tono, tono, tono, tono, tono, tono.
00:19:56
Speaker
And every time I show that video to students, everyone is like, what is that? And that's it's really this thing like, you know, one to ten, we've done that so many times. We know that so well. Yeah, you're probably not having to do that from, you know, it's just just a learned thing really that you can.
00:20:13
Speaker
produce but once you get to the 12 that is a bit harder maybe like it's not as over learned he gets stuck and goes back to his his kind of stereotypical production right yeah whoa yeah that's really amazing but then thinking about what you just said just from a from a language processing perspective if we break it down into just language production
Challenges in Language Production
00:20:40
Speaker
It has nothing to do with his motor skills of being able to produce those sounds. Yeah, exactly. Because he, he can. Yeah, exactly. Yeah. He can move his mouth. He can produce all of the sounds. Yeah. So there are no, no, let's say with a, with a speaking apparatus, there are no impairments there, but what, so then it's a correct me if I'm wrong, right? I'm just trying to reverse engineer this now. Uh, so then it would be.
00:21:13
Speaker
Well, I mean, he also understands what he's being asked. Right. Most of the time, yeah. Okay. So it's also not a comprehension problem. No. In many cases, it's not really. So they do comprehend. For some more complex types of constructions, they show some deficits, but otherwise it's really comprehension is pretty much intact. It's really producing words, really. It's about
00:21:37
Speaker
So it's not the motor parts of it, but really getting something going from a concept to a particular label, being able to have the sounds of how you should sound that. That's where it goes wrong. Whoa. Yeah. But okay.
00:21:55
Speaker
Well, hold on. That means, though, he does know the words. I mean, if he's able to understand you also saying how many children do you have, he has access to, let's just call it the.
00:22:10
Speaker
the lexicon of the language you call lexicon or the corpus lexicon lexicon is also fine okay so if i make a letter please correct me so here's access to that but now it's literally just the linking of the the word he knows in the lexicon to
00:22:39
Speaker
The motor? No, no. Well, so it can be at various levels. Or not various, not many, but there's one first step. Let's say you go from a concept to a particular abstract word. It's not yet the word as we would produce it in terms of how it sounds. So it's a more abstract thing. It's very far removed from a motor.
00:23:07
Speaker
So that's the first step. You have to go from a concept to that particular abstract representation. That's at least what theories assume. And it can be already at that point that it goes wrong. The concept is intact, but you can't get to a particular label for it.
00:23:25
Speaker
And then there's the second step, which is going from this label to putting it into what we call the word form or a phonological representation, something that has sound to it. So it's not yet motor commands, but it has information about what the sounds are in that word. That's another step where we know it can go wrong, that they do have an abstract word, but they can't exactly have this, they don't have access to the sounds of that word.
00:23:54
Speaker
If you ask these people, is it cup? So they can't find, I don't know, microphone. They can't find the word for microphone at that point. Sometimes if you give them the option, you say, is it a table that you're trying to say? They will say no. Is it cup? No. Is it microphone? Yes. So they do know, they do have the knowledge about that word, but once they have to produce that themselves, they for some reason cannot.
00:24:22
Speaker
Okay. So it can be then the process from the concept to the label or label to the word form.
00:24:32
Speaker
or word form to the actual physical production of that sound, which they don't have problems with. Right, exactly. Then it's usually... You usually see it, I think, in other cases. But then it's really more of a motor thing, right? So you really notice in the way that it's being produced, that it's either inconsistently produced over multiple utterances,
00:24:55
Speaker
So, if you would say microphone throughout the day a couple of times, every time the error would be somewhere else in the articulators. Or it's a general problem with particular articulators, so you find, you know, that all of sounds that require a certain strength of that particular muscle, that they are all showing that type of deficit. But these are more of the motor types of deficits that are not exactly
00:25:19
Speaker
seen as more of these, let's say, more truly or truly language. I don't like that term so much. But if you would see producing words from a more motor perspective or in more psycholinguistics or more retrieving things from memory, then some of the deficits will really be way more on the motor side of things. And these people who
00:25:43
Speaker
cannot easily find words. It's usually not at the motor level, but really more at the deeper level, going from representations in memory or going from a concept to a label to a sound. And when it comes to the concepts that you now talked about, is the source for knowing these concepts for language comprehension the same as for language
Concepts in Language Comprehension and Production
00:26:08
Speaker
production? Yeah, I think, yeah.
00:26:10
Speaker
I think that's something that is not debated. The concepts themselves, they are regardless of if you're using them to produce yourself or if you're hearing a word and going to the meaning of that word. I mean, if we look at the brain as being hopefully as efficient as it can be, why store something twice? Makes perfect sense. Right. And otherwise also brain patients suffering from brain damage should then also
00:26:37
Speaker
show similar, well, not similar, but very particular differentiation with their performance, if that was the case. Right. Yeah. And you do see in cases of what's called semantic dementia, right? What is that? So it is a type of dementia, but it particularly affects
00:27:03
Speaker
like conceptual knowledge so people will over time they will lose information about concepts and that holds if you if they now produce or comprehend words or even if you give them you know objects that they don't need any word they don't need to use any language for they will still show deficits in understanding which an animal is for example so everything at some point becomes a cat or an animal
00:27:30
Speaker
Even in their drawings, you can see that they are losing knowledge about the concepts themselves rather than just the labels. And does it... Oh, that's interesting. And does that semantic dementia move upward hierarchy? So for example, what I mean is, so you have a cat and then you have... I don't know...
00:27:55
Speaker
mammal, animal, you know what I mean? Yeah, I see what you mean. It becomes more prototypical. I'm not sure if, because I think a mammal
00:28:06
Speaker
is to an even a less used level of in the hierarchy that like we don't talk about so often mammals as such, right? Yeah, that's true. So I think you if you start generalizing you either, I'm not so familiar with the behavior of these patients, but I think when there's generalization, there's all either
00:28:27
Speaker
It goes towards prototypicality, so everything becomes a dog because it's more prototypical, or everything becomes an animal because it's a higher level label. But I don't think you would move up and say this is an eukaryote or something, right? I don't think that. And prototypical is also then, I guess, patient specific.
00:28:49
Speaker
I'm sure there are some general prototypical aspects, but if you're, for example, a patient that's a scientist, is that possible? I do think that familiarity with certain things would, I would guess, would shape
00:29:08
Speaker
what remains intact, right? So I do think that's the case. I'm just not very familiar with reports about that. I had to remember of this case. It was an aphasia, so someone with a language disorder due to a stroke. He was an accountant before the stroke.
00:29:27
Speaker
And I was observing his behavior, and he was reading aloud text. This happened over multiple occasions. Almost every time he had he read aloud a number, even though it was just, you know, he's reading text that says, you know, in that house, there are three people.
00:29:47
Speaker
When he would read the three, he would say three million dollars. He would always add a million or a thousand dollars. He would always add thousand dollars to it. So he would say like in that house, there are three thousand dollars people. And I thought that was just so amazing because, you know, for him, numbers.
00:30:07
Speaker
probably from his profession, numbers do come together quite a lot with $1,000 attached to it, right? So for him, that's a way more familiar, way more frequent type of utterance than it is for me.
00:30:20
Speaker
So I do think that you probably find these biases in their behavior as well, that, you know, familiarity does, or the frequency of certain things are also person dependent. That you would find that being reflected in their behavior for sure. Well, that's crazy. Wow, it would be nice to have $300,000.
00:30:43
Speaker
Exactly. If you're always stroking in the thousands, you're doing very well. Exactly. You definitely have the right clientele. Let's say that.
Language Impairments as Stroke Indicators
00:30:51
Speaker
You mentioned stroke. One thing I was... I actually don't know why I never asked you this.
00:31:01
Speaker
And my family also, some people suffer from stroke. And I've learned very early on that one of the earliest signs that something stroke related is happening is language impairment. That's one of the very first things that happens.
00:31:19
Speaker
How come? What's happened? Is that related to where the stroke's happening? Yeah, I think it's even, right? Because most, the majority of the people are left hemisphere dominant for language, right? So if you're having a left hemisphere stroke, then there are, I think, three signs, right? Which is the dropping of the mouth, weakness or paralysis of the side of the body, and speech.
00:31:48
Speaker
something with your speech. These are signs that will, in particular, the speech sign will be there in case the stroke is in the left hemisphere.
00:32:01
Speaker
Or in the hemisphere, there's dominant for language, in people, which in most cases is the left one. So I remember even seeing, I think it wasn't that long ago even, reading in a paper that actually right hemisphere strokes, they are less often noticed, diagnosed at that particular critical, more acute point because the speech isn't impaired. So people don't notice.
00:32:26
Speaker
You would have the motor saying, but because the three main signs, one of them includes speech, I think it ends up being under detected in some cases because of it not being a strong, it's not present there when it's a right hemisphere stroke in many cases. So people will actually end up not paying that much attention to it, possibly also being a stroke.
00:32:52
Speaker
Okay. Oh, wow. I did not know that. I thought it was universal. It's good, good, good to know that it's not because if it's already two from the three, that's enough. Okay. Is there, is there a specific reason that people get left, uh, hemispheric language dominant or right? Does it have to do with handedness?
00:33:13
Speaker
No, I think there's a lot of research on that. And no, we have not been able to show that that's the case. So I think from at least some very, very old, I can't say that, but some older studies.
00:33:26
Speaker
They're already, from early on, it was clear that even, I think the numbers at that point were 95% of the right-handers are left hemisphere dominant, but still 82% or 79% of the left-handed people were also left hemisphere dominant. So that dominance thing goes beyond just handedness. But I think there's a lot of debate in the literature why that's the case.
00:33:55
Speaker
to be like a genetic marker? I think that so there's someone on campus even at the MPI looking at that and I don't know what the latest research is saying about it but I think it's not that clear you know that there's for some reason you ended up being dominant on one side for one thing and therefore language goes to the other hemisphere it's really not the case so yeah so I don't know exactly how it works but it's usually the the answer is it's
Unexpected Research Findings
00:34:23
Speaker
more complicated than we thought. Yeah, but that's good, isn't it? I mean, it's I just remember it's when you have a result, you do a study and the result is as you expected. It's boring, right? Exactly. It's really like when something gets out that you didn't expect at all,
00:34:41
Speaker
then it's really, oh, that's interesting. I was wrong. Yeah, that's interesting. Yeah. Many people would hate that. They want your data to conform to your theories. But I tell this to students a lot when they come to me like, no, it's not working out. We are not finding what we thought. Oh, and they are really upset about it. And I'm like, oh, this is so exciting. And they're like, what? How can you say this? Like, now I don't know what you're writing my discussion. It's like, no, but it's the exciting thing.
00:35:06
Speaker
But that's, that's what the discussion is for. It's not, it's not, uh, we thought this and that's what happened. You have a great day. No, no. It's, we thought this, that's not what happened. What could this be? Yeah, exactly. This is awesome. Yeah, exactly. Yeah. Well, this is, uh, but then wait, then I was blindsided by all the language studies I took part in because they always excluded people that were left handed. Yeah. But then it makes no sense.
00:35:33
Speaker
Well, it does in a way because you run a slightly bigger risk of including people who are right hemisphere dominant for a language if you have left-handed people.
00:35:44
Speaker
Ah, okay. It's just not black and white. No, exactly. So you may be excluding tons of people who still have left hemisphere dominance because you're excluding the left-handed people, but they are not that many anyways. And so in that sense, it's okay.
00:36:04
Speaker
Coming back to a point you said earlier, so I have to apologize to in advance, I might be jumping back and forth. No, I love that. It might get annoying for the listeners, so I apologize also. I hope they already know that I keep doing this. You mentioned
00:36:20
Speaker
language and memory. And I remember when I, while I was doing my PhD here, we, um, we had this, what was this called? These Friday PhD meetings a long time ago. And I remember you were presenting. All right. And, um, that time already, it just, it just pops into my mind. Okay. This is actually not written down anywhere here. Um, and you were,
00:36:48
Speaker
you were looking at language and you were including memory, you were including motor, you were including all of these other, attention, research, all of those things. And I mean, you know, I was just focused on motor and I was sitting there and I was thinking, well,
00:37:07
Speaker
how else could you look at language? I mean, shouldn't you include all of those things? But there were also a lot of people there that really did not like what you were saying because they don't look at language from all of those perspectives.
00:37:27
Speaker
Why? Am I missing something? Am I being too motor focused or too naive? Maybe there's something, I'm not giving them the benefit of the doubt. I don't know. So I've always had that impression as well that, or let me put it that way. I think from frame or maybe even due to tradition over time, people came to study things
00:37:55
Speaker
You know, my domain is motor control, so I study motor control. My domain is, you know, memory, long-term memory, so I study mostly episodic memory. Or I do cognitive control, and then the other people do language. But if you think about it, to me, as a domain, it's a huge domain that is way bigger than each one of these other domains. Yeah.
00:38:24
Speaker
And I don't know a single person who studies language as a
Bridging Language with Other Cognitive Domains
00:38:30
Speaker
thing. You always end up focusing either on production or on comprehension. And even within that, you focus on one particular aspect of it because otherwise it's just too big. No one does. And the motor control of speaking and retrieving things from memory, no one does both because it's just completely different.
00:38:51
Speaker
So I think from a more, you know, maybe because how things evolved over time within the cognitive sciences, you either have people who study language or something else.
00:39:02
Speaker
And what I've always tried to do is bridge. And what I noticed when trying to bridge the gaps is different things actually. So some people would say, the people from these other domains, they look at me and say,
00:39:22
Speaker
You're doing language. Why are you even trying to talk to us? Whatever you're doing is language. Or if you find something that could be interesting to us, we'll say, but of course you're finding that because your task is an episodic memory task. Well, it's not. It's a language task. Yeah, I don't know, but there's episodic memory in there. That's why you're finding something that looks like our thing.
00:39:42
Speaker
When I would talk to language people about how I was trying to bridge it, I would also get these, why do you need that? Language is a unique thing or it's unique to humans, it's its own thing. It doesn't need all of these other domains to be involved. Language is unique and there's something, there's core language and we don't need to study these other domains.
00:40:07
Speaker
And then surprisingly, there are these other people like you, and I was talking to my colleague today, this morning actually with Sabine, and the reaction is like, isn't that trivial that you should be studying these things more? No linking them to each other. Isn't that obvious? And isn't everyone looking at the problem in that way? And now you're asking the same thing.
00:40:31
Speaker
Like, I don't know if I've been missing entirely, who is doing that? Because I've not encountered these people. Right. And over the years, I felt that I'm most of the time doing this quite on my own for many of the cases. So I don't know, I agree with you that that's how we should be looking at it. Maybe not uniquely, like just doing that, but we should also be doing that.
00:40:56
Speaker
And maybe I should go back to the literature and see if I'm missing something, because to you, it also appears to be trivial. Well, no, I guess I really have no answer to this. I always try to always give people the benefit of the doubt, because there's always a chance that you miss something always. So be open for that. But when I ask probing questions, I just never get a satisfying answer. It's just always like, oh, yeah, that's that.
00:41:26
Speaker
So, how was that an argument against what I just said? It's like, why don't we, I don't know, look at language and motor control? Oh, because the sky is blue. What? That's how it feels to me.
00:41:45
Speaker
What did you just, that makes no sense. But now, so I'll ask you something, Mr. Motor Control, the researcher. Do you know that actually- It's been some time to be fair, okay? No, but even that time, that's one of the most fundamental things that I think motor control people may not know.
Motor Control in Language Production
00:42:01
Speaker
Okay. That speaking is actually the, like in terms of the amount of muscles you're using,
00:42:09
Speaker
in terms of how fast you have to use them, how well coordinated you have to use them. Speaking is the most complex and the ultimate motor control problem. I can imagine. It's the fastest discrete thing that has to happen in a perfectly coordinated way. I can imagine. It's speaking.
00:42:35
Speaker
I believe you immediately because it just makes sense. But if you guys don't talk about that, like I don't think the motor control domain, like people, they'd realize that, you know, they are also neglecting a huge part. And then there's these people who only do motor speech, motor control, which is, you know, yet another thing that needs a specialization. But yeah, that's I know I agree with you. It's
00:42:57
Speaker
It's not something I looked into, but it makes complete sense from a motor control perspective from the amount of
00:43:06
Speaker
the amount of muscles you need to use in a coordinated way, the distribution of those muscles, the timing of the sounds you produce, and not only the timing of the sounds you produce, but accessing what we just talked about, the concept from concept to label to word form and then production of sound.
00:43:30
Speaker
There is so little time to do all of that. It's insane. It's amazing that we can actually do it, right? And that's why I find production just so... And that's just from a motor control perspective. It's insane that we can do that. But...
00:43:47
Speaker
If you just look from a motor control perspective, we don't give it that much credit, I guess, maybe. So now I'm just saying things, right? There's no signs backing this up. Maybe because everyone can do it. Do you know what I mean? I mean, if we look at sports, from a motor control perspective, of course those things are amazing.
00:44:14
Speaker
But they pale in comparison to the details we just talked about, what's required for language. But yet, if you can hit a baseball with a bat, you can earn millions.
00:44:32
Speaker
But if you can just speak normally, you're normal. It's fine. However, do you remember this, what's the name of this guy? I forgot his name. This fastest talking man on the world. Yeah, no, but I can imagine. I don't know who that person is. And this received a lot of praise because it literally is even more insane.
00:44:58
Speaker
it's uh yeah so whoever whoever is listening just just type that into youtube faster speaking man i think there's this video where he's doing michael jackson michael jackson's bad and he's just speaking it the the whole song and i think he does it in i don't know 15 seconds or something it's in and you can hear every word that's that's crazy you can hear every word
00:45:23
Speaker
but I have no idea how he can do this. So it's lots of practice from a motor control perspective, a lot of practice, but still. And probably not retrieving anything from memory really, or at least not like the words themselves probably lose meaning for him. Yeah, it must be a complete forward model that just gets played run down. But still, no, I agree with you. We are just as guilty. But I just really just never got it. I just didn't understand how you could
00:45:53
Speaker
Yeah, how you could not study that together. And actually, so I'm sorry that I'm still talking. But there was a hold on, I wrote this down, actually, you know, in the we developed language, humans around what time do you know?
Evolution of Language in Humans
00:46:15
Speaker
you mean it from birth until later or when it came in evolution? Evolution and evolution. I never know these things. So, okay. So I looked it up. Oh, okay. Because I was going to say, like, I think it's again debated, right? And we are pushing this further to a bit earlier times than we thought at first. Yes. Isn't it? Yeah, exactly. It's, it's, it's, I mean, we, we can't know for sure because nobody had a recorder, right? So you, what can you rely on some paintings, some,
00:46:44
Speaker
Nobody really knows about word production, I guess, but must be writing then, something like that. But what I found interesting was I think a few years ago, a researcher, Aldo Faisal, Faisal Faisal, he's working at Imperial College London, and he's normally studying motor control and those types of parts.
00:47:09
Speaker
And what he was looking into, he was looking into the development of stone tools in the Paleolithic period, which is 2.6 million years ago. So I was thinking, what has that to do with anything? What are you talking about? And he was showing these images of the stone tools that were very primitive. And then he showed other periods of the Paleolithic period. It's more recent.
00:47:38
Speaker
which showed quite a big development in the complexity of the stone tool. So even if you just look at the stone itself of how the pieces are being chipped, how the stone is being designed.
00:47:52
Speaker
Now what does this have to do with movement? What they did is they taught people nowadays, normal humans, how to make these stone tools. So you have a stone and then you just have another stone and then you chip it away and you have to learn how to do this.
00:48:11
Speaker
And while these people were doing these movements, he was recording those movement patterns, and he was using some fancy machine learning algorithms to dissect specific patterns of movement. He was looking at it from a perspective of a grammar.
00:48:29
Speaker
How can we, is there maybe a movement grammar, an action grammar involved in this? And when he looked at the early ones, the grammar was very primitive. And when he looked at the more complicated ones, the grammar was much more elaborate. And what I thought was fast, it was very fascinating. And it coincides with our increase in brain size as humans and
00:48:59
Speaker
it also coincides with roughly the period where people say that's when we develop language. Do you think that, I mean, just maybe I'm going too far putting these two things in the same bucket because one is motor and one is something else. Do you think just to go one abstraction level higher, it has to do with our ability to
00:49:24
Speaker
to abstract things, like in a grammar sense, that gave us the ability to actually develop language in the first place. Because what you think you're doing with such a movement grammar is
00:49:43
Speaker
being able to abstract away from a certain thing. Yeah, so it's not like they couldn't produce those movements before. The muscles were there, but I guess the abstraction level from the conceptual perspective. And then the coinciding with a
00:50:04
Speaker
increase in brain size. But it could be something that gave rise to both, right? Yeah, of course. I think once you start understanding that a certain shape leads to a certain type of cutting that has those consequences, then you're going to move more towards making the tool look that way rather than in another way. Of course.
00:50:24
Speaker
And maybe it's that the ability to maybe plan and understand the consequences of a certain shape giving leads to this that will later give lead, give rise to that. That's the thing that enabled both the tools to become more complex and language to become what it is. I would say rather that.
00:50:46
Speaker
Well, which it is boiling down to maybe the ability to create abstractions from certain things. But I don't think it's the motor skills that enable us to do that. No, no, no. I'm not talking about the, I'm not saying everything's motor. No worries. I'm not like that anymore.
00:51:10
Speaker
I remember I used to. But no, no, just from an abstraction perspective, I just thought this was, I just never thought of movement as
00:51:23
Speaker
a grammar. And already just looking at it from that perspective, that's pretty cool. Definitely. And I mean, it's not like he was not... And it started to get more complex. It wasn't just this one movement that resulted in some type of muscle combination as a grammar.
00:51:45
Speaker
but a specific type of movement combinations that then, you know, the abstraction became more and more and just as we have with language and the concepts we have, I was just, I was blown away. That was insane.
00:52:00
Speaker
But unfortunately, I have not seen that being published anywhere yet. But maybe I just don't have access. You know how that goes with a paywall. Right. Yeah. Well, but OK, sorry. I just went on this rant. This was quite a strange tangent. Coming back to memory then, how do you look at language and how it interacts with memory?
Interaction Between Memory and Language
00:52:24
Speaker
Because you are interested in that. Right. Yeah, and I think also it depends a lot on
00:52:31
Speaker
when people talk about memory, what they are talking about, because they think there's lots to be said about why. Or, you know, there used to be lots of good evidence that memory and language are really two separate things.
00:52:47
Speaker
But how? I mean, do you have to- For sure episodic memory, right? Oh, okay, okay, okay. So that's one thing. But then there's also this, I think, evidence coming from AJM, this patient who had both of his Hippocampi resected, who developed severe memory problems.
00:53:05
Speaker
I think more of the evidence at that time was that his semantic memory was completely fine, which would be things like, you know, what is the capital of France and, you know, what is X, that type of knowledge that was not affected, even though this person could not form new memories whatsoever. So that was, I think, the evidence, the ultimate evidence that semantic memory and episodic memory are very separate from each other. And then language was always
00:53:35
Speaker
I like to see lots of aspects of language as being part of semantic memory. But many people don't make it that explicit or don't really see it that way. They see language maybe or our lexicon as a special part of our memory system.
00:53:55
Speaker
was still memory. I would argue that's still memory. Yeah, so would I, but yeah. What the fuck, people? I mean, seriously. No, sorry, sorry. But I had this, as well. I was studying motor control, and I was looking into motor learnings, also memory, and be like, oh, those are implicit memories. What the fuck? Yes, memory. That's the word you should focus on here. And I go, that's different than episodic memory.
00:54:18
Speaker
Yeah. I think people just, people just like chopping things into smaller pieces. Maybe we feel that it makes it more graspable in terms of our understanding, but I'm not sure. Well, there must be a limit. Yeah. Right. Yeah. There's a, I'm not trying to.
00:54:34
Speaker
to take a dump on those people. I understand that you need to put things into categories and study them, but there is a point where you need to put them back together. Right, exactly. And we are not very good at doing that. No, you're very right. So there's this thing, right? That episodic memory and semantic memory are two distinct things because patients with hippocampal damage will have episodic memory, profound episodic memory deficits, but no semantic memory deficits.
00:55:01
Speaker
And now there's been quite a shift, I think, in the field. Even last year at a very big cognitive sciences conference, there was an entire symposium on semantic versus episodic memory as no longer being this versus thing. It's way more
00:55:21
Speaker
intertwined. So there's lots of good evidence for that. But then what really I thought was striking is that at some point
00:55:32
Speaker
What then happened is that people are saying, oh, you know, there's episodic memory that is not that different from semantic memory. But then they started, they started again, compartmentalizing semantic memory itself. So we have personal semantic memory and we have personal, whatever type of memory. And then we have whatever semantic, like, and then all of a sudden semantic memory was no longer
00:55:56
Speaker
similar to episodic memory because we had created five different categories within semantic memory and I was like, oh God, this is just not helping it, you know. But is there evidence for it? Well, they had good evidence that these are not exactly the same thing.
00:56:10
Speaker
But I find also in lots of the cases, it's more of, I find at least, it's such an artificial thing caused by the way we put people into experiments. Sure, yeah, yeah, yeah. Right? And so we create all of these different categories within domains that may not really exist in that particular way. It's just that that's how we usually test people in experiments. And that's just a very bad way of going about it, I think.
00:56:37
Speaker
I guess that's kind of funny. Studying language, or actually anything, is limited by the linguistic concepts we have in our own head. Right. It's kind of a catch-22. Exactly, yeah. So what I find really cool to try to do is see how, for example, the way that semantic memory is organized
00:57:05
Speaker
if that affects or not necessarily affects, but if that is reflected in the way we use language, for example, right? So there's again, this very nice things that I find really, really cool again, coming from patients with brain damage.
Semantic Memory and Language Errors
00:57:23
Speaker
So you find these semantic paraphasias, right? So the person is trying to say wife,
00:57:29
Speaker
But he can't find the word. So he says, like, my mother. No, no, no. My sister. No, no, no, no. And then the therapist asks him, oh, do you mean your wife? Yes, my wife. Right. So he's in the right domain within his women that are close to him in part of his family or whatever. But he can't get to that exact label. And so those are the things that I find really cool, how that to me is how
00:57:58
Speaker
the way that semantic memory is organized is reflected in how, in language, as we are using it, in errors, for example. So those are the things that I like studying. And how you can't just let, let's say, once you have a concept and you want it to get to a label, the accepted idea in the field is that lots of things will also get activated. And sometimes you do grab the wrong label.
00:58:27
Speaker
So activation can't just spread without any breaks or whatever, because it becomes a mess. So there needs to be something either controlling it or selecting whatever you want, if you want a more homuncular explanation, which is not really an explanation. What do you mean with activation in this case? Yeah, that's the way that cognitive scientists call activation that we have no clue what it means.
00:58:56
Speaker
What do you mean like brain activation? Even at the level of a concept that if you want to, you know, or maybe should I think about it like, like, uh, um, what are those things called? You know, these word, uh,
00:59:12
Speaker
Always, I think. Word bubbles, no, not word bubbles. Ah. But I think I see what you, I also can't find that particular word. Or you can also find them, I think, on Thesaurus, where you just type in the word and you can just click on synonyms and then you can get like a graphical representation of how far these things are away from each other. Or you mean like that type of activity? No, no, but do you mean word cloud? Word cloud.
00:59:41
Speaker
No, not really. See what that again shows, you know, like how, you know, the way things are organized in memory. No, more like, I don't know if you think of, you know, memories in terms of features, I think like feature based explanations are nowadays quite more or less, I guess, accepted. So a concept will have multiple features and probably
01:00:09
Speaker
If you're trying to use that concept or if you encounter that concept, activation here would maybe mean that neurons that are associated with those features are firing in synchrony, for example.
01:00:26
Speaker
But I think the way that cognitive scientists use activation is not really thinking about it in terms of neurons. A lot of the times they think it's more inspired by computational models where activation would be, I don't know, you know, the level of that particular node in the network is more active.
01:00:45
Speaker
But it's more the things spread and they can't just spread forever and in any way that the network would allow for it because they think if you do that, you get a mess and you don't get to anything in the end, right? Right. I mean, then it would just the whole thing would get activated. Doesn't matter where you start. Yeah. And then you and then it doesn't have any problems like, OK, now you have a thing that just always it's like a switch that
01:01:13
Speaker
It's always on. It doesn't matter what you do, the light's always on. So like, okay.
01:01:19
Speaker
No, I agree with you. Yeah, those models don't help. No, exactly. So it's more in those terms of thinking of activation in those terms. Okay. But I don't think we exactly know what it means in terms of what the brain is doing. I think there are attempts to, you know, like, what is a word? Yeah. Like, I don't know of any good evidence of what a word is in terms of, you know, neurons and what they do. I don't know what a word is in the brain.
01:01:47
Speaker
Right. Exactly. You didn't have that question on your list. No, I was just thinking, how would I look into this? And then I thought, OK, I would look at this from a reading perspective, but then we are in perception research. And then we move into the visual cortex. But that's not a word. You're still at the processing of the visual features. And you need to reach some level of abstraction to then get to the word.
01:02:18
Speaker
Yeah, I have no idea. So what would be a plausible thing for a word to be? Would it be a couple? Lots of neurons firing together and it's just there. The sum of the parts, that's what makes a word, but no individual neuron has the information about a word. Even though you have these concept cells, right? They're supposed to have information, one cell
01:02:44
Speaker
representing one concept regardless of the way you you input your input is but yeah I would I would I would I wouldn't think so no exactly because if I don't buy it because it's I
01:03:00
Speaker
there's too much cell death happening in the head. And language is still too prevalent. So it can't just be one neuron equating to one word. There must be some type of network activity. I was also never really a big fan of
01:03:26
Speaker
It's not one neuron that has the word, but it's seven. I was also not, never really a big fan of that either because it's like, that's what it, but it needs to be distributed. I think we can all agree on that, but how exactly it would be represented. I have no idea because the.
01:03:48
Speaker
the level of abstraction is insane if you think about it. If you really think it through, it's insane. And even if you don't get to language and just study vision, the level of abstraction there is already insane. And language is one higher, I would say.
01:04:06
Speaker
Same goes for, if we were to stick to just perception, same goes for hearing words, auditory perception, then it's the same. And if you were to think of, it's called Braille. Yeah. And then for touch, it's the same. It's always one up. But I have no idea how that would be represented in the brain.
01:04:34
Speaker
I guess not by just one area. Because as you said, it's just it's everything. But what would that be? I don't know. Yeah. Yeah. Yeah. So those are the things that I find really cool to think about, but not that I really haven't. And it's just so hard. You know, I think a lot of we we can advance a lot by looking at single cell recordings in animals. Yeah.
01:04:59
Speaker
But then for language, that doesn't really work, right?
Animal Communication vs. Human Language
01:05:02
Speaker
I never thought about that, you're right. There's a huge gap to fill that we won't be easily able to fill for language because it is restricted to humans. There's no other way you can study it.
01:05:15
Speaker
But how... Wait, wait. So, okay, I heard very often, actually, during my entire studies, language is specific for humans. It's one of the things that makes us so special. It's the thing that makes us human. Okay, let's say it's the thing that makes us so human.
01:05:34
Speaker
But there are also some cases where, for example, animals learn sign language or parts of sign language. Is that just faking language? Yeah, I guess if you would put it that way. I think that one of the main things with that is that we are able to generate new sentences, new everything by just putting
01:06:01
Speaker
Words we already know, using rules we already know, but we are always able to combine them and produce new stuff. And I think a lot of which these animals cannot do is, for example, that kind of thing. So they probably learn the meaning of certain combinations and they reuse that. But there's no generation of new patterns by applying those rules, for example.
01:06:25
Speaker
So again, it's not exactly a literature I'm very, very familiar with, but I think that that's what makes the entire difference is knowing particular symbols and what they mean versus being able to generate.
01:06:41
Speaker
new structures by combining things in a particular way. Yeah. There was this one case that I was just thinking of now. Have you heard of this gorilla who has been taught sign language, who has been trained with a cat? No. No? OK. So OK, just bear with me, all right? So there was this gorilla, and it was in captivity. And they tried to teach the gorilla sign language.
01:07:11
Speaker
And the vocabulary was, you know, was all right, but not...
01:07:16
Speaker
For a gorilla, of course, outstanding, but not what the learning didn't go as fast as, I think, the researchers were hoping for. And then at some point, the researchers introduced a cat, a kitten, into the setting. And they saw that the gorilla was super excited about the kitten. I mean, it played well with the kitten. It understood this is a fragile.
01:07:44
Speaker
So when the researchers saw this, they started to use the kitten as a learning tool, as a reward for if the gorilla was doing a good job, then it could play with the kitten. And then the vocabulary of this gorilla went through the roof.
01:08:03
Speaker
And then one day, unfortunately, the kitten died. It got out and got hit by a car, and it died. And the gorilla was super excited. So there's this video you can watch on YouTube. It's, yeah. So you can watch it. I'm not making this up. And then the gorilla wanted to see the kitten, and the researchers had to tell the gorilla that the kitten died.
01:08:32
Speaker
And when they did, you could see immediately that the gorilla understood what happened and went into mourning. And I was just like, okay, this is insane. But for me, it's just very difficult to differentiate. This is now...
01:08:53
Speaker
Is this now language? Is this, what is this? It's, I have no idea. Yeah. Then I guess it depends on your definition of language, right? Because. What is the definition of, because I don't know, I always, I always hear also, you know, I always hear these things where.
01:09:12
Speaker
where communication is equated with language. Is it? Is it the same? I really have no idea. No, I guess not right, because you have things that do convey a message by just being a symbol that we know either because we learned what it means or some things are just so intuitive from the way they look that we know that they mean something. So they convey information, they communicate something even though it's not via language. Right.
01:09:38
Speaker
Um, so I guess, I don't know, like even though I study language, I don't, I never really bothered of thinking so much about what a good definition of language is or what other people think about it. I find it very difficult. Yeah. It's like the words thing, right? How do you define a word? There's a problem with almost every definition you can give. So these are things we, we know so well where they are and yet we don't.
01:10:04
Speaker
I mean everybody, I guess it's like, as you said, at the beginning when learning all these languages, at the core you understand the principles and I think we all understand these language principles.
01:10:18
Speaker
But we don't really know. It feels one removed, kind of. It's a strange feeling. And it could be just me, right? So there's likely that there are very good definitions of language out there that would allow you to say why the gorilla case is not language. But I guess the gorilla at least understood the message being conveyed.
01:10:47
Speaker
So maybe the gorilla does understand all of the concepts, especially if you chunk them together, there's a higher order of the message than just the, you know, each unique concept, but more once you put them together, they also mean something at the level of the combination. But then if the gorilla can't produce that himself,
01:11:11
Speaker
There may be still lacking one aspect of iterate that would make language be complete. I think that would be then interesting to look at that exactly this part again from the anthropological perspective that we talked about, about the ability to abstract.
01:11:29
Speaker
And maybe then it makes more sense. I don't know if that's, maybe that's part of language definition. I should have looked this up. Yeah, me too. What are definitions of language? But this thing they said about the reward, I do find that really interesting.
01:11:48
Speaker
If you look at most animal research, how it's done with reward, right? So in even like very, um, cruel situations that you remove all of the food and the animal only gets food once they do something. And that's not how we test human participants in the lab. Right. I don't think ethics would really like that. We just yell. I'm just kidding. I'm just kidding.
01:12:16
Speaker
I would have wanted to sometimes. I never did. I would have wanted to sometimes, but sorry, continue. No, but if you do look at what reward does to animals, right? Then it's, it's just huge and it makes the whole difference sometimes. So I also wonder a lot from the, from the animal literature.
01:12:42
Speaker
Yeah. How much the fact that it's really on the basis of punishment and reward, how much that shapes the behavior in a way that we are kind of not paying enough attention to. And that's why we don't find some parallels with humans simply because we don't put them in this punishment and reward type of situation. Right. But a lot of
01:13:08
Speaker
Like we are confounding the research by just introducing these maybe overly focused tools of reward and punishment to make the experiment setting. Yeah, but more like if it makes...
Animal Research and Human Behavior
01:13:27
Speaker
You know, how much can you generalize from the animal behavior if it's so shaped by reward and punishment? How much is it then really true about in general? So for example, in terms of the brain, right?
01:13:42
Speaker
If we find certain things, that's how it works. And by that, by using animal research, we try to explain how the brain works. But all of it is always in the context of reward after you starve an animal almost. Then I do think it has quite an impact on what you will find. And so how much does it really tell us about how it may be using humans?
01:14:04
Speaker
in general, that's the thing I'm not entirely sure about, but it's more, it's always been my issue with neglecting the fact that they are put in those situations in the first place that we don't deal with humans. How much it really tells us something about how much can we really translate to human behavior or the brain for that matter. But that's an aside. But no, the reason why I wanted to say this is because
01:14:34
Speaker
I think then also in terms of doing experimental research on language, most of the time we're using language to communicate something that has a purpose. There's a reason why we're trying to communicate that. And I think maybe we don't realize enough that the fact that we do get that communicated is also rewarding for us. And then you never have that in the lab situation where there's absolutely no reason why people are speaking.
01:15:03
Speaker
There's no message really to be conveyed. Oh, that's interesting. Because I know how frustrating it can be when somebody doesn't understand what you're trying to say. Right. It doesn't feel rewarding at all. No, exactly. Right. So that's something that I've always thought a bit about, like in terms of, you know,
01:15:19
Speaker
realizing that there's way more reward going on in real communication by simply, you know, by just being in the communication and accomplishing that the person you were talking to does whatever you want him or her to do or, you know, understands whatever. There's a lot of reward involved there.
01:15:36
Speaker
I mean, it is goal directed, right? And if you frame it that way and then achieving that goal is the reward in itself. Just remember from a motor control framing things that way from a reinforcement learning perspective, it just, it made sense. Why would it be different for language? You're right. Yeah.
01:15:59
Speaker
Yeah. I mean, exactly, right? Why would the brain come up with 50 million different concepts if one overarching one works? Right. Yeah. But okay, coming back to language being making us so special, would we also develop language if we wouldn't grow up with one?
01:16:22
Speaker
Are there any, is there any research in that? I mean, I don't know how you would study this unless you're just insanely cruel.
01:16:30
Speaker
you mean where there are people with spontaneous. So I think there are a couple of cases of deaf communities that invented their own like sign language that emerged spontaneously simply because of the need humans have to communicate and socialize. So there have been a couple of those cases. An entire deaf community just
01:16:54
Speaker
You know, it arose from their interactions, a sign language arose from that. So yeah, I do think there's something, like it's intrinsic to us to communicate in a certain way that then gives rise to language. So I do think it would always happen. But now that you put the community part into it, do you think then it's related to us being quite a social animal?
01:17:23
Speaker
Or do you think this would also happen with someone who is completely isolated for, I don't know. So like you would take some person who's raised by a pack of wolves. Yeah, let's say that. That's actually completely not hypothetical. That's actually a better example than what I had. It's a more humane example than what I had in mind.
01:17:48
Speaker
So the person has actually never. Never interacted with other humans. And never even known that there's language. Never known there's language. I mean, whatever that human might want to communicate to the wolf, they wouldn't understand what that
01:18:05
Speaker
Yeah, I don't know. I don't think real cases of that exist. No, I have no idea. Yeah, exactly. That's the thing with this question. You would hope this will not happen. But isn't there a story of a child raised by wolves? Or is that just a myth? I have no idea. In all honesty, I don't know. But just, I mean, the social aspect of this is so strong. Yeah, but then it could be that the communication would be there, right? The person just wouldn't...
01:18:34
Speaker
I don't know. It's actually a very good question. I think maybe it, because we have a disability to generate and combine symbols in a particular way that we do do even like with the sign language communities or the deaf communities, right? That's, they came up with that because of the need to communicate. Then I guess maybe the, the child raised by the wolves would,
01:19:01
Speaker
use their means they would howl or whatever but probably in a I don't know I'm just making this up as well but I would at least expect that it would do it in a more systematic or maybe call it complex or at least like being able to combine multiple things and assign different types of
01:19:26
Speaker
sounds to different types of needs and then combine them in a way that the wolves wouldn't. That would spontaneously emerge from this person, I would guess.
01:19:37
Speaker
But I don't dare putting that to the test. Well, whoever knows, this would be quite interesting to know. I was just thinking if you were to study this from a purely scientific and you wouldn't care about any moral and ethical things and were to just keep a human in complete isolation for their entire life and see what's going to happen.
01:19:57
Speaker
just because we are so social, social animals, that human would already be so, I think that human would just, just die for, because it's missing one. Sure, it may have water and food and things, but we know what happens to humans when they are kept in isolation and, and they just, yeah.
01:20:19
Speaker
Yeah. Yeah. So I think I fully agree with you. So the rage, for example, or the rage, but it's well known that people who either, you know, due to aging, start losing hearing or the ones who do develop a language disorder or aphasia due to brain damage, they are way more likely to become isolated and more likely to develop a depression because of that isolation than people who do keep their language intact.
01:20:49
Speaker
So again, right? It's the thing like just not being able to communicate as well and there's a huge impact on quality of life. Does there, does there, um, does it have to be communicating with someone else or would it be also fine if you were to just talk to yourself? I guess not, right? Because like, when do we, when do we, you know, I think we do talk to ourselves to be able to organize our thoughts and, but most of the time I don't think that that's what,
01:21:18
Speaker
We get the most outro in terms of satisfaction in our lives, right? It's about sharing things with other people, ideas and thoughts. So I don't think just a monologue would satisfy anyone. No, but that's true. Unless you are, you know, one of these people who really seek isolation on purpose. But I think these are more the special cases.
01:21:40
Speaker
Yeah, that's true. And it only, it only really works in doses. I mean, right. It's actually like I'm saying, right. It's, I don't actually know if that's true. Uh, I was just thinking for myself, if sometimes I do, sometimes I do enjoy, uh, isolation, but then that is, it's sought out and rather than the other way around, it's, um, yeah.
01:22:07
Speaker
No, that's crazy. Yeah, this is something I was wondering about for a long time. Whether we should test that? No, at least I can't think of anything of how to morally do this morally and ethically in an okay way. I don't see a way of how that's possible.
01:22:28
Speaker
And if that's not possible, I don't think it's a question worth answering. Well, I guess if you are a misantrope and you do believe that human beings are horrible people, then for you it's morally and ethically the best thing to do is take a human being away from the contact with other humans.
01:22:48
Speaker
But I think he would be quite unique in thinking things that way. It would not be approved by almost anyone else. And if that is the way you think, you are confirming... No, no, that's true, but you're also confirming that thought by doing that in the first place. I mean, there's... Yeah, I mean, you confirm your own theory about human beings by being... Yeah, no, I don't think that's okay at all.
01:23:18
Speaker
Yeah. What happens, so now we've mainly spoken about a single language, except for the beginning when we talked about how many languages you speak.
Impact of Multilingual Upbringing
01:23:35
Speaker
What happens when we learn multiple languages? Is there
01:23:40
Speaker
Because some, I've heard so much stuff about this. And I really, it would just take such a load off my mind if you could call bullshit on so many of those things, which I don't know if it's bullshit or not. That's the thing. You just hear so much about it. Everyone has a perspective on language, of course. I just don't know what of that stuff is bullshit and which isn't. For example, to give you an example. Yeah, exactly. I need that, because otherwise... Yeah, to give you an example.
01:24:10
Speaker
I grew up bilingual, I guess you as well. Not really. Not really, okay. So I know other people that also grew up bilingual and then I know people that of course did not grow up bilingual and I have been told
01:24:27
Speaker
That you're superior. No, no, no. I have not been told that I'm superior. I have been told I'm inferior. Oh? Yes. And with more languages that you grow up, the more inferior you grow. That's what I have heard. That's what, well, not what I have heard. This is what I have been told. Right. Specifically by friends who now have children and they are multilingual.
01:24:54
Speaker
very proficient in multiple languages, and they don't know how they should, you know, interact with their child. Should they speak to their child in German, in English, in Dutch, in whatever? Or should they just stick to the country they are currently residing in? And they have been advised by
01:25:18
Speaker
by teachers from kindergartens and things that they should just stick to one language. And if you do more, it's gonna slow the child down.
01:25:30
Speaker
Wow, okay. So that's not exactly my field, but I think there's one crucial rule. You can speak even, I don't know, 16 languages at home, as long as the input the child is getting is from a proficient speaker. If you're, say, you're both not from that country, but you live in that particular country, and then you both start giving the language of that country as the input to the child because it will be more beneficial,
01:25:59
Speaker
but you're both not proficient in that language, that's a bad idea. The input needs to be, you know, following all of the rules because that's what will enable the child to learn something and also, you know, the learning machinery itself can only develop well if it's learning through consistent rules. So as far as I know, the bad cases have been when the input is not
01:26:23
Speaker
The right input is just, you know, the person who doesn't speak the language well and is using that to communicate with the child. That's when it goes wrong.
01:26:31
Speaker
And as far as I know, there's absolutely no evidence that growing up with more than one language at home is going to affect you. In any case, what could do is that you have, they were slightly slower in the beginning, or if you measure in terms of vocabulary size, say a child of five should have a vocabulary size of X. If you're going with two languages, you will have a vocabulary size smaller than X in each one of those languages.
01:26:58
Speaker
at that particular age. But hey, if you think about it, you actually have two vocabularies, right? So actually you have way more. It's just that for that particular language, you have less fewer words than someone who's only growing up with one language. So as far as I know, there's absolutely no evidence that it's detrimental as long as the input is good. It has to be good language, the input. So if it's not your mother tongue, don't use it, I would say.
01:27:25
Speaker
But if it's her mother tongue, go for it. Even if it's two or three at home, I would go for it. To me, there's absolutely no evidence. And I can confirm this with my colleagues. No, I believe you. I couldn't find any on this either, but nobody believes me because I'm not a language scientist. So that's good.
01:27:43
Speaker
So I'm not slowed down or anything. No, exactly. I always thought you were a bitch. I have issues. That's fair. But then there's the other side of this, right? That the claim has been for a long time that bilinguals are actually way superior because they grow up learning or having to switch between languages all the time. So that would give them superior cognitive control abilities.
01:28:12
Speaker
which has also been largely... Debunked. Yeah, I wasn't going to use that term. It's called bullshit, I think. There's quite some evidence showing that it's not really like Dutch. Okay, I just know, I just remember when I was learning Dutch, so when I moved to the Netherlands and I was learning Dutch, and I had to switch.
01:28:37
Speaker
within a span of 10 minutes between four languages, I was having massive headaches by the end of the day. Now it's fine, but at the beginning it felt unbearable. Even though you grew up switching between two languages and you were fine. Yes. Then it's just getting old. I was 25.
01:28:59
Speaker
But I guess that's all for language. I don't know. Well, for learning a new one, it is. So then it's going to be more effortful range for sure. That's true. That's true. But it was really, it was.
01:29:11
Speaker
I didn't feel like I was having trouble learning Dutch, but just this switching back and forth. Now it's fine. Now it goes with ease. I can even do it within a sentence, no problem. And people are just completely confused what's happening. But I remember at the beginning, this was really something I had to practice. It felt like an additional skill that had nothing to do with the language itself, but literally,
Limits and Challenges in Language Learning
01:29:39
Speaker
being able to, well, I guess if we stick to what we said at the beginning, to pick the right word form, I guess? Or at which point, when we actually learn a new language, what is it that we learn? We don't learn new concepts. No, I don't think we do. But the abstract labels themselves are already language specific. OK. Yeah. Whoa. Yeah.
01:30:08
Speaker
Because it's not a label for the concept, right? Because you can have concepts that don't have a label. So it's really the word, if I say cup, it is cup, but it's not yet cup with the sounds of cup. It's kind of, you know, it's a difficult thing to explain, but it's not a concept, but it's not yet cup with the sounds it has.
01:30:34
Speaker
You know, it's this abstract step between the concept and the way it should sound. Is there a limit to how many languages we can learn? I think there was this question... No, the question I think was, are there languages that are not, that are unlearnable? I would say no, but...
01:31:02
Speaker
I would say that too, but I don't know. I know some, there was this, um, he just finished his PhD not so long ago and he was one of these polyglots that, I don't know, like 12, 13, 15 languages, something like that. So, yeah. So I don't think there's, I think it's more, there's a limit. There's a time limit to how much time you have in your life to spend on learning languages. But I guess, I guess there's no limit.
01:31:30
Speaker
Practically, there is, but theoretically, I don't think there is. Yeah, I mean, yeah, yeah. You have to, of course, one, you need to have the time to do it, and two. The opportunities to practice it as well, right? Because if you don't choose it. You have to speak it. And I already have problems with mine. I don't know how you do it. So like when you were talking about the switching, I remember at some point it was my birthday, and I had people, I think, from multiple countries
01:32:00
Speaker
at home, but there were six of the languages I spoke were at that table. And I was able to just during the entire meal, switch all the time between those six languages. I had no issues with it whatsoever.
01:32:17
Speaker
And now I can't do that anymore because I don't practice anymore all of those languages. I speak them just every now and then and it got much worse. And even my mother tongue is just not as good anymore because I don't use it. So it is kind of use it, not use it or lose it, but you lose some of the fluency, some of the abilities. You don't lose much of the language itself.
01:32:40
Speaker
But I think more this fast use, that it's more witching. Yeah, definitely. No, I noticed that as well. My Farsi is terrible by now. I mean, it's not... I can still speak it.
01:32:51
Speaker
But if you put me on a marketplace now and it needs to go back and forth quick, I would be, I would look very bad. I wouldn't look like a proficient speaker at all. Okay, that bad. Yeah, but it's, I mean, it just happens. You just got to keep at it. What do you think about
01:33:16
Speaker
all of these language learning tools, these apps that are popping up left and right. The most popular one, of course, Duolingo. At least that's to me. I don't know how it works, actually. I remember following, I think it was Rosetta Stone at some point. It works quite similar. Yeah. Yeah, it's just for free. That's the only difference. I remember what I hated about it is that they didn't tell you what the rules were, and you were having to guess it yourself, but then
01:33:46
Speaker
For many people, it may work. For me, it does not because, you know, you give me the rule and I can go on and apply it because I understand what these rules should mean. I see. So for me, never worked. And I think a lot of these these tools are very much focused on. I don't know, like single words or things that come together, right? So you get whole sentences. I guess if you don't have any other option,
01:34:16
Speaker
at least you're getting something done. But that's true. Yeah. Um, I've been quite happy with, uh, sorry, sorry for interrupting you. I've been, uh, I've been quite happy with Duolingo. I have to say I'm not getting sponsored by these people, by the way, just if somebody might be, no, I'm not. I've been, um, using it to learn Spanish because it's.
01:34:38
Speaker
It reminded me of Rosetta Stone. And you have this, sure, you have this gamification aspect to it, which I sometimes find a little bit too much. But what I do like is, surprisingly, the thing that you dislike about it is I actually do like that they don't tell you. Don't tell you the rules, which works fine, because these are all Indo-European languages, right? So you already speak German. And in Spanish, it's one step away. It's not the same, the exact same family of languages.
01:35:08
Speaker
But they do follow of there's a very systematic saying between German and Spanish. Yeah. Now you go and apply this to Japanese or Farsi, right? That is not or Farsi.
01:35:22
Speaker
I shouldn't. I don't think so. At least that's not how it feels to me when I form sentences. It feels structure-wise very different. Right. So then it's probably not in the European, but I don't know. I may be making him... I don't know. Just don't go look it up yourself. Yes. Japanese I know is a concrete, because I tried to learn Japanese using Rosetta Stone.
01:35:44
Speaker
And then it doesn't work, right? Because you've got, you know, for sure that the rules are completely different. We've got no clue, absolutely no clue how you're supposed to put words together in a sentence. So I'm not sure if people have tried to use
01:36:00
Speaker
these types of tools for learning something that is completely different typologically if it works so well. I do think you need some explicit information about how sentences are to be formed because it's completely different. I do think that's the case.
01:36:16
Speaker
I mean, there was specifically with the Spanish. Of course, the first thing I did is I was looking for a Spanish buddy to try out my newfound Spanish. And I have been left at. Left and right. Which is fine. I mean, I totally get it. When you say things like, I own a suitcase, they think, why do you learn how to say that? But you have to start somewhere.
01:36:42
Speaker
But then I was making sentences where I thought I understood the structure and then they corrected me. I was like, wait, wait, but it should be like this because in that scenario it's this. And then they explained the underlying rule and then all of a sudden like, oh, okay, that would have been handy to know. Right. But yeah, so it's maybe it's.
01:37:05
Speaker
to remove from that. Once in a while it would actually be useful to know it. Exactly. So I don't think it should all be, you know, you learn the rule and then you go and just repeat the rule, just that, just drilling. It might be of a German school language. Great, like that. Just drill the rule, so maybe not that, but also sometimes it is good to know what the underlying rule is because then you get the aha and then you can go on and apply it and if you just have to
01:37:30
Speaker
implicitly derive it yourself. In some cases, it could be that you're not as efficient in doing it. So that would be my issue with these tools that sometimes some more rules would help.
01:37:43
Speaker
Yeah. Hmm. Okay. But what you did is very good. I think like put into the practice, put into practice what you learn as soon as you can. That's to me really the way to go. Right. I did this with Dutch as well. I just got the, I wanted to do the Dutch training course here, but I had to pay for it. It was 1,500 euros. I didn't have that money. So I thought, well, the Germans had come here and learned that in four weeks. How hard can it be?
01:38:11
Speaker
Well, little did I know. So I just bought this little book. What was it called? Pawakus nederlandisch für Deutschen. So that's what it was called. Not for Demis, but for Deutschen. And I pushed through it in four weeks. And then, of course, you learned these basic rules, how to make plurals, what's the articles and things like this. And from that point on, I just tried to drill.
01:38:37
Speaker
vocabulary into my head. And to all my Dutch friends, I said from now on, Dutch, no more English, which was very difficult for them because they are very accommodating. At least they tried to be. And in this case, it just didn't help. But my language, my Dutch language then improved. Yeah.
01:39:00
Speaker
I would say exponentially, just because I had to use it, I had to put things into some type of a form. Yeah, exactly. It was terrible at the beginning, but after four weeks of suffering, I could definitely tell there was improvement. So I thought, hey, okay, I have to do the same with Spanish. Yeah, and well, unfortunately, I paused.
01:39:21
Speaker
I have to admit it's the time that's the problem. Yeah, right. Yeah. And then it's easier if you already have those people around you, right? Exactly. But if you have to find a budget that then will have to make time, it's going to be one hour a week and it's already not really enough. That's exactly the one hour a week. So I was doing the Duolingo every day. So that was good. And then I was trying to just.
01:39:41
Speaker
call them once a week, but even that faded away. And it's, yeah, that was difficult, but it's, it's, I guess it's the way to go. And lots of people like learning new languages. That's at least what I learned through Duolingo. So Duolingo, if you're listening,
01:39:57
Speaker
Maybe put some grammar rules at some point. Maybe I just didn't get to that level yet. It's possible. I don't know. But if you didn't just do that. Yeah. Every now and then dropping like, this is the rule we are trying to teach you. That would help. Yes. Yes.
01:40:17
Speaker
I just have a few more questions for you. Would you mind if we go hop back into the science part? No, not sure.
The Misconception of the 'Language Gene'
01:40:23
Speaker
So there has been this, I mean, you know, I've been out of this for some time, but there has been this gene. Oh, yeah. They called it, back then at least, the language gene, Fox P2, what happened to the thing. You know that the person who discovered it is also here on campus.
01:40:43
Speaker
I know, I know that. But I thought it's unfair to ask Simon, or Dr. Fisher, I shall say, I have never met him in person, directly about this, because I'm sure he would be excited to talk about it. But the reason why I bring this up with you is because in this conversation, this presentation that you had back then, that you were specifically talking about FOXP2,
01:41:12
Speaker
Oh, really? I don't remember that because it's actually really cool. If you listen to Simon himself, he says also like we never said this is a language gene that makes absolutely no sense. It's just I think how the media picked it up and I don't know, thought it was fancy enough to put it out there in that way. Can you
01:41:33
Speaker
Because I know what it is, you know what it is. Can you just tell us a bit about the background of what this gene is? How has it been found? Yeah, I think it's a particular family, the Ke family in, I think in England. And many people in the family showed
01:41:53
Speaker
I think it was something right again, maybe not exactly how it is, but most of it will be right. At that time was called specific language impairment. Nowadays it's called developmental language disorder. So a lot of the family members had that in combination with dysarthria.
01:42:13
Speaker
What is that? Dysarthria is when it's really more at the muscle level is when there are problems with the coordination of muscles. Okay. And so you also get problems with speech, but really from the motor side, whereas the, the specific language impairment is really, it's more language and not, you know, it's more than the real core language thing. So these people,
01:42:40
Speaker
We'll have, if you look at their production, there will be things, especially when things become more complex. Again, the way they form sentences is not always correct. If there are complexities in the, not even complexities, but if you use, you know, a relative clause, like the man who did this walked out right now, that kind of construction is already more complicated for them to produce.
01:43:08
Speaker
So this family had many members that had this type of deficit. And I think they studied this family quite extensively because there was a whole, I think, lineage how it's called, right? And they found that FOXP2 was affected in this family. So that's how it started. And I think by now there's quite some
01:43:34
Speaker
There's much more that came, I think, out of those studies, but also other genes that have been linked to. So mutations in these genes linked to quite consistent findings in terms of development. And sometimes it affects language a bit more than just development in general. And then I think what they showed in mice, for example, is if you do something to FoxP2,
01:44:06
Speaker
I need to remember this because I even attended his course at that time, but it's something really at the level of learning. I don't remember if it's exactly sequencing that you can do that at the motor level as well with animals, but it's something very fundamental.
01:44:23
Speaker
in learning itself that seems to be affected that then has these far-reaching consequences for language as well. But it's not a language gene at all. Okay. So, yeah. But that's not how it should be interpreted.
01:44:40
Speaker
All right. Because I think FOXP2 genetics is really not my thing, but it controls lots of other mechanisms in terms of development in the brain as well, right? So it's a much deeper thing than being a language gene.
01:44:56
Speaker
Yeah, because then, I mean, if it was a language gene and other animals that don't have language, that would be now my reasoning.
Media Misrepresentation of Science
01:45:07
Speaker
That would be now my reasoning, but it might be too simplistic. Benefit of the doubt to the...
01:45:13
Speaker
to the media. Because I don't think scientists really thought there was a language gene. I don't think they thought that. I think it was really the way the media was talking about this. I really doubt any geneticist thought this was a language gene, because it's a very well-known gene that has, like, it's a very important gene that does all kinds of things, regulates all kinds of things. So I don't think any of the geneticists thought this was a language gene. Okay. Well, this is comforting.
01:45:43
Speaker
I've been bashing the media quite a lot lately anyways, but unfortunately they just, they just confirm it. They keep on confirming it, but this is, this is definitely, this is good. Yeah. Science is not that lost, right? No, no, no, no. No, but that's, that's good. You know, you always hear these things of scientists need to communicate their work better and things like this. And then when you get these types of,
01:46:13
Speaker
massive findings and then you hear the scientists talk about it, you just say, what happened? At which part of playing the telephone game did things go wrong? Definitely. I had that with one of my studies that, you know, I was at Berkeley and they thought it was cool.
01:46:34
Speaker
their media office wrote a piece about it, and then it started like the media caught it, right? And then it started appearing in all kinds of places with claims that I was like, oh my God. And then people started tweeting about like, oh, look at this. This is all wrong relating to my paper. But it was not that my findings were wrong. They were very upset about
01:46:58
Speaker
what the media had made of it. But if you don't go back to the source, people were saying that what I did was all wrong, but the media had wrong what I had done, right? So then I did feel, this is so unfair. I never said that. But I was being attacked for having said that, which I didn't do. So yeah, that's definitely how it goes, unfortunately.
01:47:19
Speaker
So I've been criticized for doing these podcasts that they are so long. But, you know, generally speaking, I think there are these complex topics. People are interested in them and they know
01:47:36
Speaker
They understand that you cannot understand all of this in five minutes. They understand that. And just to me, I sometimes think the only people that don't seem to get it is the middleman, AKA the media. The receiver gets it. The person investigating gets it. Just the person in the middle doesn't seem to get it. And that's, I find this very odd. Very odd. Yeah.
01:48:03
Speaker
But talking a bit more about the research that you do, what are some of the latest findings that you had in your line of research which really excites you?
Shared Cognitive Processes in Language Tasks
01:48:15
Speaker
or from my own. Yeah, from my own. We can split it in your own and, generally speaking, but I'm more interested in your own. Right, yeah. So I'm really excited about, so I work a lot at the brain level nowadays, right? And with electrophysiology in general, so we record the electrical activity of the brain, neurons use that to communicate, you could say,
01:48:46
Speaker
So the things that I'm excited about are, for example, when I started finding these links in terms of if you look at the signal and what we know the brain is doing in episodic memory tasks, we find the exact same signals when we are doing a language task.
01:49:03
Speaker
And these things were thought in the hippocampus, for example. These things were thought to be exclusive for episodic memory. And now we are finding that it may not be exclusive to episodic memory. When we are using language, the brain is also using these same building blocks, let's say. So that's something that got me really excited. And also, once there's damage to the brain, for some reason, it could be a tumor, a stroke. Those are the ones I'm more familiar with.
01:49:31
Speaker
There's lots of changes happening more acutely but also over time the brain adapts.
01:49:38
Speaker
And the exciting questions, important questions are also like, why do some people recover so well whereas others don't? Are there certain patterns of adaptation that are more beneficial or not? And these are also big questions in the field right now. So we do find some shifts to the right hemisphere once there's damage to the left that could be very meaningful for understanding what this adaptation is.
01:50:08
Speaker
So that's also part of what we are going to do more in the future. Okay, you say adaptation, that's what I used to study, so that sounds super exciting. But that's pretty cool. Yeah, definitely. So you can, just from what I hear now, you can, studying the patient population, it actually tells you quite a lot about how language is processed in the first place, right?
01:50:36
Speaker
Yeah, depending on how you put the question or how you ask it also for sure, because if you know, for example, we have cases, I know there's a certain process I think is going on in language and now there's someone with damage in that particular brain area and that process is just disrupted entirely, then it does help us understand the role of that brain area for that process. Yeah.
01:51:02
Speaker
Or even if, you know, how do you go from a concept to a label? And if you see that patients have a particular type of deficit, it also helps us understand how it actually is that you go from the concept to the label, for example. So, yeah.
01:51:17
Speaker
But I don't, because that's the approach of many people, just using patients to test a model. And that's not exactly what I find. I like that, but I also think that we can help answer questions that are more important for these people in turn as well. So I do try to work a bit from both sides and trying to understand the patterns of adaptation that could also be beneficial for the same populations, right? They're providing the data.
01:51:46
Speaker
You mean like, for example, patients that suffer from a stroke already knowing the patterns of development that they have to go through to recover and just trying to, you know, let's say make that as optimal as possible for them.
01:52:02
Speaker
Right, for example, that would be like the ultimate goal. I don't think, I'm not there for sure, and the field is not there, but that would kind of be the ultimate goal. Like if we know what the good pathways are, can we boost those or try to help them with some kind of therapy that they would go through that path? Yeah, that would be pretty cool. Yeah. That would be super cool. Yeah, definitely. And then you get the same, when it comes to anything made for patients, you get the doping questions for the healthy population.
01:52:33
Speaker
It goes immediately there, isn't it? Yeah. Cause I would want to know. Right. Exactly. What will make me be even more fluent? Why would you? Yeah. But no, I think that's pretty cool that you do that. And you're right. Lots of, at least the, the, the patient work that I've been reading about, it's about the confirming the models rather than, yeah, looking at
Reading Disruptions and Marketing
01:52:58
Speaker
it from the perspective of what can we learn and how can we help those people?
01:53:02
Speaker
Yeah. But these are very different traditions again, right? So if you look at the cognitive scientists doing fundamental research, most of the time they will use the patients for testing their models, which is fine. Yeah. But I'm not happy just doing that. It just doesn't satisfy me as a researcher to do just that. But that's more me and not a criticism to the field. Yeah. All right. Let me back paddle as well a little bit because I don't need to dump on those people either.
01:53:29
Speaker
I think this is also just much cooler to do it this way. But of course, that other work is also important to do, except for confirming the models. That's not a good approach. I have never been a fan of that. Do you have a few more minutes? Yeah, I do. Because there were some questions from people from industry, and I don't think they ever talked to a language scientist before.
01:53:55
Speaker
All right, so I have come across some people who have been just very curious about the effects of capitalizing letters on reading. Do you know something about this direction?
01:54:17
Speaker
Not really, but would it then be all letters? Yeah, so for example, you have these different schools of thought that say schools of thought, air quotes, you have different camps. And one camp says, you know, it's really good to capitalize all letters because then it means like, you know, they make these associations like then you're
01:54:37
Speaker
then you're grown or if you use all lowercase letters in, let's say, your company name or whatever, you are small, you are young, those things.
01:54:53
Speaker
Is there anything to it? But I think it's more of a marketing thing rather than a really language thing, right? Because it's really what will attract the attention. I just know I have a lot of problems if everything is in capital letter. I have a lot of problems reading. Yeah, exactly. I do think it's way more disruptive. I don't know if there's research on that, but that's my experience and also what I've heard many people saying also when we are.
01:55:21
Speaker
creating whatever catchy names for our projects. So they say like, oh, just use some lowercase as well. Don't use uppercase for everything because it looks like you're shouting and it's heavy for the reader. So it's more of a marketing thing, but I think from a
01:55:39
Speaker
processing perspective all uppercase is not very beneficial either yeah and then they have been thinking of okay what if i make every word in so let's say you have to write a sentence what if i make every word the first letter capital
01:55:58
Speaker
Or if I just have the regular way I normally write and only the first, speaking of the English language, only have the first letter of the sentence capital unless I use names and things like that. But that's also how it's naturally used, right? So I would say unless you're trying to convey something very
01:56:18
Speaker
If there's a very good reason why you're deviating from that, then don't. These are marked cases, as we call them. You're deviating from how it's usually done for a very good reason.
01:56:33
Speaker
But then if you're writing an entire sentence like that, you may be disrupting more than doing good. I would guess. I would guess as well. And I guess then it's also language specific, because many more words in German, for example, are written more capital. And I guess if you don't write them in capital. Then it's also a statement, right? So that would be deviating from the default also for a reason. So I would say just stick to the default in the language.
01:57:02
Speaker
So it has nothing to do with some type of language processing model.
01:57:09
Speaker
Oh, no, no, I don't think so. You're drawing more attention to certain things by deviating from the default, right? Yeah. But then I would say you should use that in the cases when you want to draw particular attention. But if you do, if you apply that across the board, you may be doing more harm because now everything is receiving attention and you end up not emphasizing anything that you would want to in the first place. So.
01:57:33
Speaker
Yeah, I guess you're right. I think you can see it more like using a scalpel or a chainsaw. You can cut with both. But it's just like you would decorate anything, right? And if you just put that same decoration on every single centimeter on your wall, it stops. It's no longer a decoration. I think it's the same way of thinking.
01:57:59
Speaker
Yeah. Nice. Interesting. Thank you. And to all the UX people that hear this, there you go. My final question for you, which pressing questions in your field from your perspective still need to be investigated, where you think this is where the focus should go?
Pressing Research Questions in Language Processing
01:58:23
Speaker
from what I think and not what the field thinks. Not what you think. Yeah, what I think. Well, can I have two? Sure. So the one is more in case of brain damage, what are the patterns, the adaptations or the new patterns once it has adapted?
01:58:48
Speaker
What is, is it even a fair question to say that some are good and some some patterns are good and some are not? And can we help people by improving the good or helping them use the good patterns? And from the more fundamental part,
01:59:10
Speaker
I think it can make it very complex, but at a more fundamental level, maybe understanding a bit more once we are using language. It's like these things that we talked about before. It's not naming words or just using words in isolation in a monologue.
01:59:29
Speaker
We're always trying to convey an idea or some thought and it's an interaction. It's an entire chunk that we are producing. It's never, not often isolated words. So how do we actually do that? How am I planning my own speech while you're talking to me, right? Like it's all happening at the same time. It's all very fast, very interactive. We know very little about that.
01:59:55
Speaker
Well, and you would assume that we know actually quite a lot, but we don't. Yeah, it maybe depends on, yeah, exactly. So we do know a lot about interactions themselves. And when you give me the right to speak, right? I pick up on these cues and now it's my turn. Yeah.
02:00:10
Speaker
But on these more the psychology side of language itself, how am I planning my own speech, my words, putting together my sentences while I'm listening to you, while there's something else going on here that we don't really know so well.
02:00:26
Speaker
That's fascinating and exciting. For the ones that think it's frustrating that we don't know yet, you are working on it. We've been working on it for decades. We still don't know it. But I think it's great. It's really, really good.
02:00:47
Speaker
Are there any books or other resources that you can recommend to listeners that want to learn more about what we've talked about? Or maybe there are many books. I honestly have no idea. No, and I should thought about this because I do know you ask this question. Would you recommend a book by Chomsky?
02:01:06
Speaker
Oh, no, I don't think so. Not really. No, but there's Kenneth's link. Kenneth's link. Yeah. And that's online. Yeah. So it's not a book, but they do write about different scientific topics. And there's someone in particular who likes language a lot. There's a couple of people and they do write about language many times. Then I will, I will put that in the show notes for sure.
02:01:28
Speaker
Yeah, and I will think of a book and then I'll give you... Can also be multiple. Yeah. No problem. Yeah. And those, if you know any YouTube videos that we talked about, or maybe you mentioned, I mentioned a lot of it. But I will put those in the show notes as well. Well, this... It was fun. This was fun. Thank you so much. My pleasure. For people that want to get in touch with you, how can they reach you?
02:01:56
Speaker
Yeah, if you Google my name, we have also Twitter. If you Google Vittoria PI, I am the only Vittoria PI really present on the internet. So he would immediately find me. So either my email or via my website, you can find contacts there as well. All right. So I will put your Twitter handle in the show notes, your website, and also your
02:02:21
Speaker
your email address. And you have a website for your research group as well, right? Yeah, but it's not as updated as my own. OK. OK. Then that will go in there. Well, thanks again for your time. And well, I had a lot of fun. Thanks for having me. Yeah, me too. And actually, there's so much more to talk about this. I bet it's not the last time. Oh, cool. Well then, thanks again. And to everyone listening, have a great day. Thanks.
02:02:52
Speaker
Hey everyone, just one more thing before you go. I hope you enjoyed the show and to stay up to date with future episodes and extra content, you can sign up to the blog and you'll get an email every Friday that provides some fun before you head off for the weekend. Don't worry, it'll be a short email where I share cool things that I have found or what I've been up to. If you want to receive that, just go to ajmal.com. A-D-J-M-A-L dot com. And you can sign up right there. I hope you enjoy.