 It's time to get started. So take a seat. It's the Science Cafe brought to you by OE. I am a neuroscience and behavior student. So I was really interested in this topic. So a little bit for selfish reasons. I proposed this panel and I would like to have all my questions answered from this panel of amazing scientists. So I want to ask them to introduce themselves and give a little bit of information about themselves. So if you don't mind saying who you are, how you got here, and just a little bit about your work. We're going to have a little bit more chance to talk about your work in the end. And maybe if you want to talk about yourself. This work? Okay. All right. My name is Maggie Gimini and I am a 50-year neuroscience and behavior student at UMass. I'm almost done. I just scheduled my defense today. You're all invited. It's March 29th. No, please don't come, but you're invited to the chair afterwards. Just kidding. So I study language processing in children. I use EEG to study that. And from my dissertation I'm really interested in what happens if we make an experiment for a relevant and real world? Do we get different results? And how can we understand language processing using brain measures versus behavioral measures? Does that make different answers, better answers, stuff like that? Okay. I'm Adrian Stout. I'm a professor in the Department of Psychological and Brain Sciences. I am what is called a psycholinguist. So I study language processing. Most of my work, louder. Okay. I'm a psycholinguist. I study language processing. Most of my work involves understanding how we recognize words in reading and how we assign grammatical structure to sentences as we read. I have one foot in the UMass linguistics department, which is a very famous and important linguistics department, actually, usually ranked second or third in the world. We don't have any linguists here because they're all at a department event tonight. And if they were here I would put one of them here in my class. But so I got a mama's panel because I got to know Maggie. She's helping me implement an experiment which we're tracking reader's eye movements at the same time as we record their EEG to compare what the brain is doing and what the eyes are doing as people are recognizing words in reading. And she's writing, well, she nominated me for this panel. So here we are. I've never been to one of these before. After this, I'm going to the World War II Club in Northampton for Thursday night's period. So I've had a couple of beers here and I'll have a couple of beers there. And my faculties have continued to decline, maybe linearly over the course of that. Hi, everyone. My name is Dan Bahaba. I'm a lecturer and postdoctoral researcher over at Smith College just down the road. And I recently finished up my PhD in the opposite direction on Route 9 over at UMass, where I was studying the role of hormones in regulating how birds hear and learn bird song during development. My fun fact while researching for this panel and learning about Cocoa the Gorilla, which we'll talk about later, turns out me and Cocoa both have the same type of orange minks cap, a bobtel cap. So I have something in common with a pretty advanced gorilla, it turns out. Thank you. Would you like to share a fun fact with us? Yeah, I didn't give you a fun fact. Okay. Yeah, okay. I can say my fun fact is graduating soon, but I will give you another fun fact. I have a French bulldog. Her name is Charlotte. She recently completed a genetic study to see to something that was done by a postdoctoral researcher who used to be at UMass for their PhD. And they're interested in genetic correlates of behavioral dogs. And I found out that my dog is such a purebred dog that her parents are only 22% different in genetic material. So that's a really pure dog. She's one. Yeah. All right. Thank you, everyone. I think we're a pretty good company to answer some of these some of your questions and our questions about language. We have someone that studies language and children. Someone who studies reading and writing. But I think it's a very important component of language, if I'm not mistaken. And someone who studies communication, maybe language. I don't know. I guess we can talk about it a little bit. So our little poll is working really well here. We got some really cool answers. So I think I think we're going to start by reading a couple of these answers and answering probably the easiest question tonight, which is what is language? By those faces. Was I wrong? Is that not an easy question? That's the hardest one. So I'm going to read some of these some of these answers first, okay. So what is language to you in a few words? Words. Conveying emotion. Sending a message. Means of communicating wide range of thoughts, feelings, communication. So communication, a lot of it. Mental and behavioral way to apprehend experience. Wow, that's deep. All right. So what do you guys think of these answers? And how do we think about that? Are we thinking about this question correctly? What is language? Is it related to communicating feelings and emotions? Are we on the right path? So I think it's an interesting approach to defining language that moves away from the procedural definitions that people try to make in science to say that it is conveying a message that is specifically emotional or otherwise more meaningful than just maybe a signal that something is immediately wrong. So maybe more abstract concepts. I think that's a really interesting approach. I will say that I'm not a huge fan of super specific definitions of things like language because I think that we get into arguments about it that are unnecessary. Yeah we do. But I think that all of your suggestions are really great. Would anyone like to add something to this technology? What is language? As someone who doesn't study human language, I would argue, and we'll talk more about it hopefully tonight, that language is really any learned vocal communication in an animal, human or otherwise. And that's as broad as I would like to define it because like Maggie, I think you get really deep into the specifics instead of really stringing goalposts that make it easier and easier to distinguish humans from other animals. However, I think the animals do possess the language. Okay so the grounds of debate are set because my language is not necessarily vocal and the animal communication is not language so we'll talk about this. Okay, back out a little bit later. But it was a good start. All right, I think in that aspect, I'd like to give Dan a little bit of a chance to answer to that. So does language really makes us special, are human special snowflakes because we have language and... No. Humans are special for other reasons but there exists, I would say, a lot of spectrum of vocal communicators or learned vocal communicators. So certainly I'm able to sit here tonight and talk to you all about language and sort of meta-analysis, whereas maybe other species that learn vocal communication wouldn't be able to engage with sort of meta-cognition. However, it doesn't discount their abilities as well. So maybe I should show some of those slides for this. So one thing, I tend to think about language and parallel humans with animal vocal communication or non-human vocal communication and you all probably have some vocal communicating animal at home. You probably have a cat or a dog that keeps you up at night or begs for food. However, well there's a lot of animals that communicate using vocal cues like barking, gnawing, so on and so forth through the next slide. There's actually a really small subset of animals that actually require learning early in life in order to produce these vocalizations. So even though you're cat meows, it's sort of pre-programmed with that ability. If it never heard another cat meow, it would still continue to meow. However, if you look at across the annual community, vocal learning is exceedingly rare. That's why we really champion humans is being really special, right? We learn our language. However, we're not that special because there are other animals out there that do this as well. One of the most prominent examples is animal you'll start hearing soon as the spring approaches which is songbirds. They learn their song in their seasonings, so every year in the spring or early development. Hummingbirds are another example and parrots as well is probably another obvious one, their vocal mimics as well. But there's also other mammals that learn vocal communication as well that you might not think of as well such as whales and dolphins as well as seals and walruses and even bats have learned vocal communication as well. So we're part of an exclusive club but we're not the only ones in it. So it sounds like the answer is it's complicated and depends on who you ask and depends on how you define languages. Am I right to assume that? Is that fair? Okay, so I guess it's complicated. I think we'll have more opportunity to talk about animal versus human communication. But I actually have a question for you. So it looks like humans, of course, we're really famous for knowing language and a lot of ethnicities and all over the world we develop language somehow and I have an impression that writing and reading also seem to be universal to language. Is that a fair assumption or writing which is essential? I'd say no. So in fact for the humans the, for most of the human history of the approximately 100,000 years we think during which humans have communicated by means of spoken language, we did not have writing systems. So writing systems are a very cultural, our cultural version that is very new. Language is also sort of new in evolutionary terms but is evolutionarily much older than writing. Writing is so new as to be really considered, I mean really is just a recent social invention and in fact still, you know, much of the world, much of the many, many people who have perfectly well developed abilities to speak and comprehend language do not read or write. So it's not, it makes use of our capacity for spoken language. Writing makes, right reading and writing make use of our capacities for spoken language and for comprehending language but are not in any way entailed by that capacity. Cool. That's so I was wrong. Awesome. So I guess we can move on from this debate a little bit. I don't want to get physical here. So let's make it a little lighter. How do we actually learn it? Okay so Dan, as we said, emphatically the animals do learn how communicate and but it's not a dispute about that we go through this process and we don't have so let's talk about the mechanisms a little bit. No mechanism but maybe in practice how do we learn language? Okay yeah so your first language is a really interesting thing because you can acquire your first language without explicit instruction. So you just need exposure. You do need specific kinds of exposure but it's different from learning a second language like Spanish or you know you go to school, someone teaches you very explicitly how to speak that language. Your first language is very different and what's interesting about this is there are sort of kind of implicit ways that parents will help their children acquire their first language. So one of these ideas is motherese so it's the kind of sing-songy voice that you use of the baby and you probably do this naturally to babies but I do it to pets too. You just find yourself doing this to babies or young children and you don't really think too much about it but this is actually helpful for them. You're exaggerating certain things in a speech signal that really helps the child acquire language. Children are also more likely to pay attention to you if you do this type of speaking and as long as you're not overly simplifying what you're saying this kind of thing is great. It does very culturally but especially for native English speakers this is something that happens and it's something that helps babies learn their first language in a way that requires no explicit specific instruction. So it's great that you mentioned first versus second languages because I'm very interested in this because English is not my first language but the first thing that people notice when I start with my second and third or fourth beer is that my accent might come out. No, I'm doing no drill. My accent starts coming out so but a lot of people have accents that are stronger than mine so I've heard even when I drink. So how does that work? So I started learning English really young so does that play a role? Yes, definitely. So there is debate about whether you guys have probably heard about the idea that if you're really young that's a great time to learn a second language and as you get older you get worse at learning a second language. There's debate about why this is and what is in the brain, is it social, is it even real but there is some evidence that learning a second language without an accent is way easier if you do it younger and this comes down to something called phonological categories. So a phonological category is kind of the category of sounds that you think is one letter. So for example in Japanese the letters R and L are within the same phonological category which means they're not two different letters. So if someone is a native Japanese speaker and then they come and learn English it's very difficult for them to distinguish these two letters because their entire life that's all been one thing. It's like asking you to discriminate between slightly darker blue and slightly lighter blue as two different things but it's even harder because speech signals happen so fast they go away it's so rapid. So changing the boundaries of your phonological categories is something that happens in infancy and early childhood and it can be difficult to push those boundaries around and that's why accents can persist even if you have exposure to your second language for years and years and years. Some people have this ability to completely get rid of that and they don't have an accent anymore but not everybody. So Dan can we learn anything about animals from humans? Do the mechanisms or the ways that animals learn their communication? Do they relate to the way humans learn or all of that? Is it similar? Yes. That's it. So one thing that Maggie was sort of approaching is this idea of learning language earlier in life being easier and this is this concept called critical period plasticity. Maybe people have heard about this about why learning language early in life is a lot easier and we know this is true in age-limited vocal learners. The best examples that we know about this and the most well-studied vocal learners and ones that I studied are songbirds in particular. So zebra finches which is the type of bird that I studied in my grad research they first have to listen before they ever produce their own speech. So much like human infants don't really make many sounds maybe they cry but that's about it they don't really begin to babble right away birds are the same exact way they listen before they speak and we call that the the social phase or sensory phase where they're just sort of taking all that information in and craving what's called an auditory memory or you know a memory of dad's song or mom's song or that kind of bird which is thought to be pretty similar to humans as I understand it. Similar to humans as well after this auditory memory is formed they go through what's called sensory motor learning and what that entails is now that they have this auditory memory they start to produce their song and it sounds like crap it sounds like for those who are old enough to remember what winding the cassette tape sounds like that's essentially what birds sound like when they first began producing their song it's this warble noise it sounds just as bad forward as in reverse but what they do is they start to compare that noise to that auditory memory they compared and then they're like oh wow what I produced was an absolute junk I should change it a little bit and so then they refine that song little by little and that's what's called air correction learning so they realize how different what the noise they're producing is from the memory that they've created of their parent song and then eventually they create a pretty good copy or a iteration of their parent song and that is you know a loose analogy to how we think humans learn language as well so let me start by returning to my earlier remark which I disputed your characterization language so from so human language researchers often point out that each of the sentences that one hears or most most of the sentences that one hears are sentences that one has never heard before so in your response this now you uttered a large number of sentences which presumably many of them most most of the people in this audience had never heard before and you had probably never uttered anything so you have very limited you have certain linguistic elements at your disposal right you have speech sounds and then you have words and you can combine them in completely novel ways according to a certain set of rules and the rules are the grammatical rules of your language and then the combination of those words according to the grammatical rules of your language gives rise to certain very specific meanings which you have in mind as a speaker and which your audience is then able to recover as a listener so each instance of you using your language is essentially you are communicating a novel message right you're doing something you're you're combining a set a repertoire of linguistic elements in an entirely novel way so the difference that human language researchers see between human language and animal communication is in this creative that is what we are doing as humans is on the fly using a set of linguistic elements in this entirely novel way on each instance to produce messages that have never been produced before communication can consist of something as simple as you know smoke signals or semaphore or for that matter right so these are these are stereotyped behaviors which correspond to very specific reasons and so human language researchers see that this enormous gap between what any language communicate what any animal communication system is actually doing and what human language is doing so that so that's why i just want to explain why did the human language researchers see language as something really different in kind not just in degree but really different in kind from animal communication um as for learning uh human language researchers so people who study language acquisition actually don't see human language development as in any way an instance of copy okay so human language researchers think that in fact the human infant is abstracting of very complex set of rules and representations from the input that is let me say that in a less in a less from a lot let the bless lingo the human infant is actually learning a lot of very abstract things about their language which they are then using to generate novel utterances which again nobody has ever heard right so they're not actually copying individual utterances that their parents have produced their parents have probably never produced most of the utterances that they're that they're producing they're learning a set of rules which govern how they're going to turn a message into into new utterances which is a really different thing from copying what your parents have done so it looks like we need to go back to our definition of language for the night one more time to update it is it fair then to say that language is not one thing it's made up of several different processes like vocal learning like auditory learning like innovation creativity to make new phrases and perhaps since we like to we like evolution a lot here perhaps animals are halfway there they mastered vocal learning they mastered auditory learning but they haven't developed the sophistication of our language is that fair so that's a really i mean so the question of how to relate in evolutionary terms human language to animal languages are really difficult questions that i think nobody has a satisfactory answer to um i'm certainly i don't i i don't have it i mean so so how many people here have heard of gnome chomsky okay so so chomsky is the father of modern linguistics and he he has has a theory of i mean one shouldn't even call the theory he regards human languages so as so completely distinct from any animal communication system that he thinks that the only way that human language could possibly have evolved is that there was essentially a sudden mutation which sent human language off in this completely different direction right um now i cannot evaluate that i have no idea whether that hypothesis is correct or not but that's one of the live hypotheses on the table because it's really hard to see how a more continuous trajectory would work right how we get from anything that looks like animal communication systems to what we have as as human language i imagine it wasn't easy for him to put this out there without any debates huh oh over there yeah very seriously yeah awesome all right i think we are ready for some questions for the audience from the audience so we have the first hand here she will only speak Spanish and i'm wondering if you have any words of wisdom that i can share with them i'm sure they're not the only ones who've ever done it but um you know i think it's a little different than what you said of you know she learns english very young you learn english very young but i think it's different from being hit immediately with two languages so before we answer that did everyone hear the the question is that good okay yeah so that is an awesome idea um there is no drawbacks to that there's some there was some concern that maybe a child that is exposed to two languages at once might speak later might have a hard time might be confused sometimes that happens but it's not a big deal the the main thing that you're getting is the benefit of having two languages so there there is work that shows that children who are bilingual are better at something called executive function so you can kind of define that as like their ability to switch um their attention to something new stay on task kind of higher level cognitive abilities um there's a lot of there's some brain imaging research and there's also correlational research that suggests that being a bilingual child is great for you in that regard so i think that's an awesome idea and i have no advice i then do it all right here in front yeah so the question is if it if the which of the parent speaks which language if it matters because you imagine that maybe the mom has more interaction maybe with the child is that your question that's very interesting i think that also interacts with what are the mother's language is the language of everybody else around the child so for example if the mother is speaking spanish and the child is here in an area that speaks english um how that would be different actually i don't have any like experimental evidence for you for how that would work but i imagine that the amount of input the child is receiving really matters so if the child has a mother who speaks english and they're with that mother a lot their community speaks english they are going to be dominant in english as opposed to maybe the father speaks spanish and they only encounter the father speaking spanish um yeah so that would definitely be a factor but there's still going to be the benefits of being about language oh interesting right so you maybe your primary caregiver you're more bonded to them and then you are more likely to speak their language i think that that's also entirely possible yeah great yeah so one of the things i'm not a linguist but one of the things that i hear linguists say is that there really aren't such things as separate human languages that they integrate um at the boundaries and that really you have one phenomenon of human language and you can't identify people that are old and non-lingual but that there really aren't such things as separate languages that there are integrations among them all such that it's all one continuous human phenomenon yeah do you have any comment on that that seems like interesting okay yeah so so um one of the things that that we learn from from contemporary linguistics is that the commonalities among human languages are rather striking and in fact one might argue that human languages are much more similar than they are different and in fact that this is related to the learning issue right so so it has been argued that one of the ways that kids one of the things that makes it possible for a kid a kid to learn language to learn their the language that they're hearing around them is that they have a they have knowledge even innate knowledge about what the possibilities are and the possibilities are actually really tightly constrained right like so there are only so many things really there are only so many possible languages that they could be hearing because human language can only vary in certain tightly constrained ways and then they have to figure out which one of these sort of varying sort of close variations they're actually hearing right um now of course human languages one way that human languages are often distinguished is just the possibility of communication between people right so i mean if i'm a monolingual speaker of english and you're a monolingual speaker of mandarin right we won't be able to communicate whereas if i'm a monolingual speaker of you know standard american english and you're a monolingual speaker of um you know a dialect from norleans right we will be able to communicate so that's how we distinguish sort of the notion of dialect from the notion of language is simply the possibility of communication between between two individuals but there's no sort of deep difference there right it's just do we manage to communicate right so yeah i mean i think i think you're right that that there are enormous similarities the generalizations that holds at all levels at the level of sound at the level of how words are formed at the level of how sentences are formed there are generalizations that hold across all human languages that are really profound yeah cool uh we are just going to get one more question from you losses and then we can move on to the next session and then we're going to have a q and a a longer one i've certainly grown up i heard a lot of repetition um specifically for my mother as in pick up your room pick up your room uh with a lot of variation to then please pick up your room please for the love of god pick up your room and i know i'm just singing then a lot of our conversations with a lot of the same so the third hi how are you yes thank you oh it's raining out but there is also that capability of putting together completely new sentences that we haven't said a lot about a lot of our stuff is casual congratulations and okay so let's say have a long response yes you're absolutely right so so there are constructions there are idioms that occur a lot absolutely um but memorizing those is not a good way to learn language right um and memor and producing those is not a good way to interact right and that is so someone who who simply memorizes set expressions and is not able to use language productively by product and what productively means for a linguist is basically create new new forms that that person actually will not really be a speaker right so if you if you encounter someone who seems at first glance to be a speaker of english but all they can do is produce second memorized expressions you realize aha that person actually doesn't really speak english what's that yeah like that could grab us right right right on the other hand someone who really is a speaker of english can do crazy things involving novel forms right so like you can read jabber walking by lewis carol and you can recognize that even though the individual words are not even words there is a sense in which these are well-formed sentences right and you can debate anybody else and yeah yeah so so you're you know something about your language that goes well beyond knowledge of the specific input that you've heard before so um we're just going to call this q and a for a little bit but we we're going to have another one soon and it's going to be longer i promise um so i have i have a burning question in my head right now because we we heard from the perspective of a linguist no chunsky how language might have evolved but i'd like to hear from the perspective of maybe evolutionists and do we have other ideas out there about how how language might have evolved because we have these cool things that animals can also do might you know might be related might not be related but there could be i imagine competing ideas right dan yeah so i'm afraid to use the word language now sitting next to each other i'm just but i would say that from what i know so i know less about you know human language evolution and i know more about how we think vocal learning evolution has come to exist in songbirds for example probably the most well studied set of animals that are vocal learners and one idea is that it's originally this part of your brain called motor cortex that is involved for sort of just that doing motor behaviors some examples are these really complex gestural displays or essentially dancing in certain birds so there's these type of birds in the neotropics called mannequins that can moonwalk just as good as michael jackson i would argue and so some of the theories is that birds that have sort of branched out from these really highly motor expressing animals the certain areas of the brain sort of sequestered this motor region for motor production of vocal lisations instead so instead of having this elaborate dance now you have this elaborate courtship vocalization instead so that's sort of one idea that's persistent in the field right now this motor theory of learning as well one idea too is that auditory learning has sort of been much more evolutionary conserved so if you look at how the brain receives auditory information how we all hear it's pretty well conserved across humans across mammals birds so on and so forth the same basic pathway but something special happened in birds and humans and bats to allow them to correspond this auditory learning to a learned vocal production as well. Cool. Does anyone have anything to add? I'm just this brings up for me something that I think is interesting is the difference in variability of the productions so I feel like with animal communication the sounds that they're producing are very well reproduced I guess so they're stereotyped exactly yeah so really okay because so what I was going to say is that I know that with the zebra finches they have what they have really really specific song structures and like I will listen to things that Matt will play for me and he'll be like oh that's different I'm like no it's not like that's the same but one thing that I think that's interesting about humans is that we have a really broad like low-level acoustic properties that can count under one category so for example you can produce a given letter sound like a P very very differently but speakers will be like oh okay that's a P and I have a large continuum of P sounds that I will accept. Same thing goes for sentences it gets even crazier with sentences there's just massive amounts of acoustic variability and there's this sort of abstract concept of that it's not just the low level signal but it's this higher level concept of that sound meaning something so I'm interested to hear about other animals that have more variability because I'm not familiar with that. I should prep this all of this by saying that we're talking or I guess I'm talking a lot about vocal communication in animals and I just like to put an asterisk next to all that that vocal communication is not the end all be all in communication in the animal kingdom and in fact we focus very heavily on learned vocal communication but there's animals that communicate with odors like dogs and rodents they have a lot of odor cues in which they communicate with each other there's types of fish that will actually communicate with electricity so they can measure sort of current being produced there's you know animals that you know communicate with dance as well so we're talking about a pretty limited sphere right here because it's a best analogy that we have to spoken language in humans so I just like to prep this with all of that but I'm getting to your point about sort of more dynamic maybe song so we study in the lab very commonly this cute little bird called a zebra finch and maybe if any of you have been to Petco or have one at home you've seen these tiny little cute songbirds that have the zebra like stripes on their chest and we study them because they are simple in the lab so I think what Maggie is going to get that they have this really awful song in fact I have a clip of it but we shouldn't expose anyone to it sounds like a squeaky dog toy well are you plugged into sound yeah so no one would argue that's melodic right no one would enjoy listening to that but what it is and Adrian was also giving together is that it's highly stereotypical so that's what makes it a convenient model to study in the lab is that you know they have about four or five notes and they're syllable it's the only song they'll typically sing but it turns out even zebra finches are more complex than just the four or five syllables that they produce they have individual contact calls that they learn so so for individual recognition but other popular animals to study their songbirds include animals that you see all around here which are European starlings so people are maybe backyard birders you probably see them they're an introduced species so purists really hate them because they're kind of an introduced pest more or less but they have repertoires of you know 300 plus songs in their vocabulary so to argue that all songbirds have sort of this fixed one song is overly simplistic there's this huge dynamic range and spectrum of complexity at least in songbirds and I would argue that's why I think vocal communication or learn vocal communication is on a spectrum you have these really advanced animals such as songbirds that have this really dynamic way of communicating and then you probably have you know the top dogs which are humans you know we're the best at learn vocal communication but we're not exclusive in this club so something to add or okay I'm good okay all right all right that was really interesting um so we've learned what language isn't what language is we're kind of a little iffy depends on who you ask it's complicated definitions language is not only one thing all that so it seems like scientists haven't really decided yet on that aspect on that aspect um but when you look at the animal world and and Dan gave a lot of examples of animals that learn things in their own uh biology right so songbirds learn how to sing and then we have bats and stuff but they're we've been trying as humans to teach animals our language right um I don't know if you guys have ever seen Coco the gorilla learns a bunch of science I we have a video about Coco here um oh we're used to I don't know where we went maybe we don't no we don't I'm sorry oh here it is why do we go there so here's Coco unfortunately Coco died last year but this is uh called Coco's last message to humanity she says man Coco love earth Coco love but man stupid stupid Coco sorry Coco cry time hurry fix earth help earth hurry protect our nature see you thank you so people have taught through their entire scientists spent light their entire lives trying to teach Coco signs and things like that um there's also another example of of uh chimpanzees after I placed the canvas in the machine Lana questioned me in the following manner question you put chow in machine to which I responded chow in machine this exchange took place four additional times after which Lana came back with question chow in machine to which I responded yes Lana was not about to buy this however but she came back with no chow in machine which was indeed true I asked question what in machine she did not respond so I repeated it again question what in machine to which Lana responded cabbage in machine so we can start with Adrian um does that count if we teach animals our language does it count okay um so I don't want to I feel like I there's no way to respond to this without being a Serious Party Goober um I I'm very skeptical about what these isolated demonstrations actually show so um so okay let's take this at face value the how long answer I'll have to give you 30 seconds okay okay I'll be quick so so what this what what let's just take the second second case so question chow in machine cabbage in machine it's on right so these are um impressive feats on the part of a chimpanzee to string together signs in this way no question that that's true and it is impressive that the it is the case that eight chimpanzees gorillas have been taught to associate signs with objects and it is also the case that they have been taught to associate um and they've been taught to put together these signs in order in a way that is interpretable to their human experimenters that is all true um are they doing anything that is akin to human language so how would a human ask the question whether uh something is in the machine so first of all they would do this funny thing um in which they they invert the order of an auxiliary verb right so they would say is the chow in the machine so that's what we call a movement operation um the now I'm not saying that a movement operation is required for human language but it is characteristic of human language so every language has these complicated movement operations in which in which the message is um that the nature of the message is determined by the the the fine details of the order in which the elements are occurring instead this chimpanzee has learned that when he wants to communicate a message a question start out with this thing this this token that says question right I mean that's very impressive that's great that the chimpanzee knows that chimpanzees are very smart we share 99 of our DNA with chimpanzees they're very smart that is very impressive and it is telling us something about the symbolic abilities of the chimpanzee it is not telling us that they have anything like human language I think cool no no you would like to add something yes so I definitely agree with Adrienne about that um I think it is it's an interesting sort of work around that we have this so what this chimp is doing her name is Lana what she's doing is she's pressing buttons on this computer that are lexagrams so that is like a little picture it doesn't specifically represent what it is so for example the lexagram for apple is not going to be a picture of an apple it's like a random symbol she's pressing that to represent a word and you've given her a button that represents the concept of a question so she just has to press this question button and she knows that she can get things when she presses this question button and then follows it with a string of words or button presses um and like what Adrienne is saying that's very different from using syntactic or grammatical structure to get your point across it's a much simpler operation there was a lot of excitement about this in the beginning um later experiments showed that they do have a difficult time even without a question like more complicated grammar so for example if you ask them to do two things in one sentence they will do only the first thing they can't remember to do the second thing um if you mess with the order of words um so instead of something like the boy chases the dog you could say the dog was chased by the boy they would never understand the difference between those two sentences how we use syntax to make those different um I think Lana is a better example because we have this this button pressing behavior that gives us a very concrete record of what she's doing I think Coco was a little bit more difficult to interpret I don't know if you guys noticed that was kind of a super cut of a lot of signs and we can't really tell what order they were put in which is unfortunate I don't want to say they messed with the order but we don't know um and it's also interesting where is she getting these ideas right like these are very profound ideas from this gorilla like telling you save the earth that she's watching you time is running out like it's it's very dramatic and it's very it would be moving if you could take it at face value but that's something that's kind of difficult to do um so I definitely agree with Adrian that these are really fascinating test cases and interesting to push the limits of animal cognition and see if they're given a language system what can they do with it but it's not human so so Maggie was pointing out that you know this may be a super cut and there's a very famous example from I'm not sure if it's from Coco or from another earlier ape who was taught some human sign some signs but one very exciting development was the use of the phrase water bird to refer to a duck and this is great right I mean so this is a novel combination of concepts to refer to a particular entity and it makes sense but it turns out that in fact what the I can't even remember I can't remember who's Coco so I don't know if it's Garela or Nim okay so wash I was okay thank you so so what washo I'm gonna take your word for it what washo said in fact was water water water bird bird bird okay and the middle two expressions from that were extracted and in fact that I mean makes sense right water bird duck but in fact that the full string was water water water bird bird bird so I mean there's always been a lot of excitement about the possibility of teaching apes some sort of linguistic communication and it has led to over exuberance of the part of researchers and over interpretation of certain very ambiguous utterances or uses of sounds Dan would you like to add something to this or great right this is fascinating do you do we have any questions all right talked about Coco and everybody else the um have you done any studies or looked at any of the research that Irene Pepperburger has done with Alex the parrot so I'm not as familiar with Alex the parrot I mean I know of all of the training maybe someone else actually knows more about the parrot I love Alex what I found was that when parrots are learning to speak they make the same vocal sounds as young children okay as they speak they come up there exactly the same it's almost like babbling in your experience yeah they babble oh that's cool okay you know the parrot when they're young and babies when they're young they battle no matter what language they're learning whether around the world they make this babbling sound parrots do the same thing so Dan do you know my bird babbling to your bird's battle yeah they totally babble but when they're learning their own language not when they're learning to mimic humans right I don't know if that's what you were referring to but Alex didn't mimic Alex yeah and and that was one of the findings I mean her thing was on autistic children if I remember remembering correctly she was looking for communication how to communicate with autistic children that's why she picked Alex I don't know but Alex was able to verbalize and communicate in English right yeah I know that Alex he learned labels for things right so she could ask him like get four green blocks and you can go do it and if he wanted a walnut as opposed to an almond right he would be very specific and and tell her yeah if he wanted one right exactly so that sounds like a learned vocabulary yeah which is similar to what these animals have right they have a very specific vocabulary and there is some interesting debate about Alex so in general people say an animal has never asked a question but Alex as it's been recorded that Alex asked an existential question so people say Alex asked what color am I which is cool and like that would be a big deal and I really wish we could verify it because that would be very interesting I think everything about Alex is really fascinating because you're right it's not just mimicking he learned the vocabulary and he used it and his abilities were remarkable in their own video like you can't question that he did really really amazing thing but she worked with other birds as well and I don't know where her research has gone since then so I really don't need better yeah all right yeah right I'm gonna languish and I can't remember who did the study but they took a pool they partitioned it it was one dolphin on one side and the dolphin on the other side and on one side the trainers would be like okay hit the button on the left and the dolphin would have to say it was body on the other side hit the thing on the left and that they both got fish afterwards and they could do that consistently which seems novel because in the ocean they don't know each other if I'm on the left right right I don't know I mean that is an interesting yeah so so those are the those are like the edge case right that's the tough those the tough cases yeah I agree I would just add and it's sort of related to that a lot of animals can do auditory learning test dolphins totally learned their own vocal communication and I think we're just only recently starting to appreciate that it goes beyond just you know these noises that they make for individual recognition but just as easily as you can teach a dog you know that the to say sit you know relates into the certain behavioral response perhaps you can also translate uh maybe I have upsetting interest I'm looking for the microphone that's uh you grabbed the microphone I know I was just saying that uh you can teach behavioral associations with learned auditory information that gets back to this idea of this ancestral auditory circuit that we all have and be able to pair that with learning can we have one more uh let's see I have to pick okay so you had it for the first time yeah sorry that last part I really that's a big question what the relationship between music and bird song and the only thing I know is that back in I think the 1800s before there's a radio recorded music before phonographs came out they used to train different types of birds such as nightingales with flute music and then you would buy that nightingale and it would repeat back that music and so before there was radios and pre-recorded music you'd buy a songbird to just do that very task to listen to song because you can have music in the house recorders there you go so correction recorder not flu thank you for that yeah clearly I know more about birds and instruments but I would get back to a pretty loaded question uh John was asking are we underestimating or overestimating animals intelligence by example of squirrels getting run over I would say that's you know a huge exaggeration because just because modern inventions that are only as recent as you know last 50 80 years when was the first part by 1920s something like that team up 1980 1880s okay so I'm way 140 years the the scale of evolutionary time is a lot slower than the scale of human invention time so one good example is a counter to that um is the nine banded armadillo that isn't quite common around these parts of the woods but if you go down to Florida it's essentially the most common roadkill that you'll find down there is essentially the squirrels of the south and the reason that they're very common roadkill is because of their adaptation to react to you an oncoming oncoming predator is to jump up rather than jump out of the way so if they're in the middle of the street and a car is coming so to run quickly they run up and then you just end up getting hit by a car so I'd say by judging the animals intelligence by the basis of modern technology is a bit of an unfair comparison in that way so if anything we're underestimating their ability to adapt to you know uh their own uh environment compared to our more rapidly changing landscape I would ask that that even human language has only a very tenuous relationship to intelligence the the notion of so so every neurologically intact human learns a language essentially without any difficulty whatsoever right so I mean even um and and the um there doesn't seem to be any kind of so there's differences in vocabulary depending on anyone's education level but one's mastery of the basic rules of the sound the sound system of one's language and uh and uh the grammatical rules of one's language is essentially perfect regardless of one's level of intelligence and one's performance in other domains um on that ground and it's been argued that language is some it's sort of uh should be thought of as something like um a a species ability that is is really not even learned in any kind of literal sense but rather develops in much the same way as this as an ability like walking develops um rather unconnected to general intellectual capacity two reasons I was asking that question one is we're talking about teaching the animal human language which is a very very different thing they didn't evolve to learn human language no more do we evolve to learn to you know figure out where the local word varies but the other part is on the music for the other day that uh playing music or teaching the people how to use it helps their Alzheimer's and so they've got this some kind of connection to you in there oh i'm not familiar with the connection that you mentioned between using is it said learning an instrument or learning playing music and reduce rates of Alzheimer's yeah yeah I can't speak to that I don't know if anyone else can speak to that oh that seems like we think they're deep into the woods talking about that stuff it's really interesting which is very long thank you um just one last question you have one yeah thinking about it all these questions music I was thinking of minor chords major chords the way our sentences go up when we ask a question is that universal how animals will whine and people sort of have that same line is that you know inherent is it learned but about the zebra bench auditory memory I don't get that because in the natural world we learn children and birds I was again was learned by around them they're modeled so did you um isolate the zebra bench once it was exposed to the language it was supposed to copy then put it in a cage by itself to see if it relied on auditory memory is that how you did it yeah so that's a great question so zebra finches in the wild they're actually this very gregarious species so they roll about 150 deep in the wild and there's a whole bunch of other different birds as well they're native to the outback of Australia and it turns out that you tend to learn best from the peak that feeds you so kind of like you learn from you were asking about you know maternal care and you know that stronger bond relationship and maybe you learn more from at least we know from zebra finches that even though they're hearing their uncles and other males in the trees and in the lab in like the different cages that they'll only learn from the males that actually feed them and interestingly as well if you take away the dad but you have a brother in there they'll learn it's called horizontal transmission they'll actually learn from sort of the oldest brother in the nest as well and then you were asking as well like how do you figure out yeah so the way how do you know that that's a great question so one thing that we've been talking about is the different phases of learning language and I would argue for songbirds it's similar to human language learning in that you listen before you speak and so one thing one trick or one way that people have taken advantage of that aspect is you train a bird you just play that song to it either with a live animal or with a tape recorder depending on the species and you do that only when it's not producing its own song so they don't actually start producing song until at least for zebra finches are about 40 days old so if you train it before it's 40 you know that the only auditory information it's giving is from whatever you present it with and then if you raise it with a female at least for zebra finches females don't sing you know that it's not receiving any other male song to impact its song development and so at the end of the day when it gets to adulthood you can compare that song of that bird those are just a female to that of its father or its tutor that you presented it with and you can actually measure how similar it is and you're pretty confident that there's no influence because it was raised with another female that is not singing so I don't know if that clarifies a bit there there is some more complications to it so there's a recent study that showed that female birds just by flipping their wing can actually direct song behavior so there's this whole social aspect of song learning which is very similar to humans as well that it's a social learned aspect and that's one parallel that I would argue zebra finches are really great at some birds like white crown sparrows you can put them by themselves as a baby in a soundproof box you put in a tape recorder you just press play and they'll learn to repeat that song you do that with a zebra finch it'll be as if the bird has never heard a song in its life it won't sound anything like a zebra finch song so it won't learn unless it has a somewhat social model and so that's why we think there's these really strong parallels with language learning which is very social for life awesome thank you so much guys for these questions they're really interesting it looks like in a lot of aspects in a lot of the topics that we're talking about we stumbled on this unknown about what the brain of these animals can actually do right so we were talking about the alex lavera and then in these these apes and it looks like we maybe underestimated their brain power to to do some of these things so it looks like we need to keep studying brains in order to get some of these questions right so i'd like to move on a little bit from this esoteric topic of what language is and all the this heated debate to let's talk about the brain a little bit I want to know what's going on in our brains when we when we only produce language but just before we move on if you're thinking about getting a drink and you haven't yet please do because if the tap room likes that we do this here they will they will welcome us more time so it's my idea so so I want to show a video okay I hope I hope it's it's uh it's we can see it well but I really like um this video because it shows us what our brains do to our language and when I like Maggie to comment on it a little bit um no you can do it after but hopefully um people so I'll I'll play this video and I want you guys to pay attention if you can to this person's mouth okay so it's going to be a face and he's going to be producing sounds and pay attention to their mouths I'm going to ask you what sound you heard okay what was the sound well be yeah okay great all right so next one all right I don't know uh if you could see very well their or their projection there but what was the sound this time the V a V okay I heard D there okay okay cool let's uh listen to the third time um what about this time there was a V identical sound ba ba two sounds what happened there Maggie so I think that this is usually a cool illusion but it's even cooler here with this slightly messed up projection because some of you can see the face and some of you can't I feel like you're having it either you're having a difficult time seeing the face you've seen this before or you have really good auditory perception I'm closing my eyes you're closing your eyes okay so the things that you see about the mouth the articulatory you know structure of language matter so the entire time that die was saying ba it is the video that is tricking you into you're supposed to hear first time is ba second time is da with a D and the third time is va with a V and if you produce these sounds ba da va you can feel that it feels very different what you're doing with your mouth and it also looks very different um you have so over learned what it looks like for someone to produce a D a B or a V sound that when you see that your brain says nope that was not a ba I know that was a da and I don't care what we heard I saw that now that was a da so that that's basically what if those of you that it worked for that's what you experienced and this importance of this is called audiovisual speech so you can see it and you can hear it at the same time and audiovisual speech is important in other areas as well as like cutesy illusions so if you are able to see someone's mouth while they're speaking you can understand them much better I'm sure you've experienced this but this actually matters for your brain as well so there is an ERP component it's just like a brain wave that we can measure on the scalp and every time you hear an auditory onset it's like a loud noise your brain produces this component it's called the end one because it's negative and it happens at 100 milliseconds it's a very creative field of study so the end one is a response to auditory assets these happen all the time in speech not always at the beginning of the word just a little bit of louder burst of sound will give you an end one so if you are seeing someone's mouth while they're speaking you will actually have a larger end one to their auditory assets their loud noises then if you cannot see their mouth so this not only matters perceptually in this example but it's literally changing what your brain is doing in a way that we can easily measure just on the scalp so it looks like a brain is doing fancy stuff so so can we assume that our brain has like specialized areas for language yeah so this is a complicated debate so in the beginning of sort of language and brain research we had this idea that there were two areas of the brain that were really really important for language we have broke this area and we have Wernicke's area I don't know if any of you have heard these they're left lateralized broke his area was thought to be really really important for producing language people who have lesions or damage to broke his area have a hard time producing fluent speech and then Wernicke's area was really important for understanding the meaning of language and also producing language that makes sense so if you have damage to your Wernicke's area you can produce language though but it doesn't make any sense you keep talking and talking and talking and no one actually knows what you're saying so these two brain injuries sort of spawned this idea that broke his area and Wernicke's area are very special for language they're like the language areas of the brain and they are the most important thing now we know that it seems like this always happens with science it's way more complicated than that and language is really a whole brain experience in most people language is generally left lateralized which means you use specialized structures in your left hemisphere for language but the right hemisphere is also very involved it's especially involved for low level acoustics so just like the sounds in general your two hemispheres are processing sound at different rates so sounds that are faster on the right sound that are slower are on the left and this difference can be it shows that we're not it's not as simple as left lateralized in two areas we really have a whole brain language processing system and adrian is it the same for reading and writing and spoken language in terms of brain so there's a specific brain region that has come to be called the visual word form area that seems to be involved in the recognition of printed words it seems to be involved in the recognition of words in alphabetic languages like english but also in the recognition of characters in languages like chinese so it's actually as you might expect it lies actually at the right at the border between the visual areas of the brain at the back of the head and the language specific areas the brain in the left temporal world that Maggie was just referring to and so it's sort of it's anatomically where you would expect it to be right at the border between those regions of the sort of bottom of the left occipital temporal junction and it's exactly what the visual word form area is doing is a topic of great debate whether it's really somehow been recruited or is recognizing only letters and words or whether it's an area that can be that actually is involved in a certain kind of visual perception of complex forms it just happens to be recruited by reading because that's what we do as modern humans this is an area of debate but there definitely does exist this very this area that if you go in an MRI and you your you do a reading task will light up differentially during that task oh cool and uh do we see something similar in animals then like that learn their vocalization yes cool yeah so um getting back to Maggie's she was mentioning sort of the two sort of dominant brain regions involved in vocal communication or language processing and production we know at least in songbirds that there's similarities as well same with hummingbirds and parrots I don't know too much about in other vocal learning species but in birds at least again I mentioned there's these highly conserved auditory pathways and birds they seem to have a similar Bernanke's area which is thought to be really important for speech processing it's called N.C.M. the caudal medial night opalium and it's very similar to auditory cortex just like in you and I and seems to light up similar to what Adrian was saying and humans when maybe they're reading but in this case N.C.M. lights up when birds are hearing another song it doesn't activate when you play them tones or noises or music but specifically the birds own species song is one of this area lights up similar concept but for production there's this area called HBC which the acronym is actually the proper name for it it used to be called the higher vocal center but that's sort of fallen out of fashion and we know that this area is similar to Baroka's area and that if you lesion this area if you sort of shut it down the birds can no longer produce any song behavior as well they make some like warbling noises and one researcher back in the 70s noticed that they maybe sneezed a little bit more when they don't have this area but that's about it they're not able to sing and then interestingly as well getting to this idea of lateralization even though these are bilateral structures there seems to be a bigger impact for auditory memory in N.C.M. in the left hemisphere that seems to be the hub or some excuse me some people think of it as and similar for HBC there seems to be a lateralized function but both are required for vocal production really cool so we're running a little low on time we're almost at the end of a 730 block of 130 one hour and 30 minutes block so we're going to I'm gonna ask if you guys would be able to stick a little bit around to answer people's questions if you can if that's okay so we're not going to have a Q&A question a Q&A block right away because I want to ask a couple more questions to you guys before we adjourn okay awesome so we learned a little bit about brain well languages seems to like the answer to everything seems to be it's complicated but I want to talk a little bit about the future and where we're heading okay with language my the most important question in this panel to be is are emojis language I read somewhere that apparently 92 percent of everyone that has internet uses emoji to communicate can we consider it language now yes no do you use emojis okay emojis are sort of the modern day version of body language I would say people thinking of that as language and there's this like I guess classic example where people you know a really obvious example of a very descriptive body language is maybe I know you do some posturing like this it's somewhat universally understood but I heard an example that asked but try conveying that it will snow on Tuesday using just your eyebrows and the same could be said of emojis as well you can surely get across really basic concepts through just direct images but it no longer is what languages which I would argue is this sort of arbitrary symbolic form you know dog doesn't actually represent what a dog is but if you have an image of it you you have more information but you can't get into more complex syntax than that yeah I mean you could you can actually argue that emojis are are a return to the origins of language right so so we had what is that no no no I mean so so some of the earliest evidence of human linguistic behavior or human yeah I guess you'd say linguistic behavior are these hieroglyphics in which the in which there are representations that are direct visual representations of the things to which they refer and that's where we are now right so we've come back great don't know how good I'm back really cool so unfortunately we're going to have to adjourn right now but the last thing I want from you guys okay in a couple of words is we want to know a little bit I want to know I hope so I want to know a little bit more about your work just in a couple of words what have you found and why is it important yeah that's always a terrifying question but so I am really excited about finishing up my work at UMass right now because I've I've been studying child language development I do experiments in the lab with children trying to understand language processing in the real world but I use an EEG system that is attached to a wall the kids have to wear a funny hat I have to do the same experiment with the children every time in order to be able to compare a big group of kids so I'm trying really hard to play inside those boundaries while still giving the children something that is like like real life as close as I can get it and I think as lay people you will hear what I'm doing and you'll be like wow why haven't we've been doing this all along I'm just having them listen to stories they answer questions about the stories this is something that's very different from what a lot of people do with children in the lab and I'm finding that when you do something that's really natural and normal for children the results are different and I think that it's really important that we are doing experiments in the lab that are as relevant to real world people as possible yes I guess that's what I'm the most excited about right now awesome cool okay so what does this sentence mean visiting relatives can be dangerous anyone what does it mean good good question right so that's the question so that sentence is ambiguous it could mean one of two different things it could mean it's dangerous for to have relatives visit you or it could mean it's dangerous to go visit so language is full of ambiguity and one of the things I study and that that's an example of what's called a syntactic ambiguity so there are two different grammatical analyses of that sentence none of the individual words is ambiguous as to its meaning but the grammatical analysis of the sentence is ambiguous and one of the things I study is how we resolve those ambiguities how we deal with those ambiguities in real time as we're listening or reading these ambiguities are actually everywhere we don't realize it but they're all around us and we have to resolve them all at time so if I say put the present in the box on the table in the kitchen that sentence is actually about five ways ambiguous right the box can be on the table in the kitchen the present can be you're starting okay so just play that back in your head and you'll see it's about five ways ambiguous you have to come up with an interpretation and you have to apply the rules of grammar of your language to as you're hearing the sentence for reading it and most of my work involved reading figure out which analysis you're going to go with and so one of the main things that I study is how we how our brains automatically decide which analysis to go with and how we recover when we've initially adopted the wrong one okay so my research right now is not related to languages all I'm just studying how much voles will miss their partner when they're spending some time away from them but my dissertation is much more related to the topic of tonight which was studying again how hormones are involved in hearing and producing and learning birdsong and one of the big takeaways is that estrogens are really important in developing male songbirds for both hearing song as well as for learning song and it depends on what hemisphere the estrogens are acting within as well as returning to this idea of lateralization so with that I would like to call it a night please hang around talk to our speakers if you have more questions I'd like to thank the top room again for letting us come here and thank you all for coming this has been a lot of fun at least for me that I hope you enjoyed it we do this every month we love feedback so please tell us what you think sign up to our mailing list give us feedback and hope you enjoy it have a good night thanks for coming