From Dropout to Developmental Psychologist: An Interview with Professor Jesse Snedeker
Sucheta Srikanth
Professor Jesse Snedeker is a professor of psychology and affiliate of the department of linguistics. Her laboratory explores language comprehension, development, and production across a variety of demographics, typically focusing on infants and children. She utilizes techniques such as eyetracking, electroencephalogram, and naturalistic observation to study the process of learning language and native speakers’ intuition towards it on a semantic and lexical basis. They are one of few labs that are using developmental EEG within the language context and straddle the line between language development and language comprehension, focusing on the particular niche of looking at how comprehension is developing.
Q: How did you decide that you wanted to work in the field of psychology, and more specifically, development? How did this area lend itself to language?
A: My first lab experience was with an EEG lab that studied moment-to-moment language comprehension, our understanding of words as we hear them. As I was talking to the professor in that lab, he said, “It really seems to me that a lot of the questions you want to ask are being studied over in linguistics. It might be more about the development of language.” So based on that, I went over and talked to the folks in linguistics. And I guess for me, the decision was whether to go into linguistics or psychology. I'd settled on studying how languages are learned by children.
Q: What about the general development process made you curious about language acquisition?
A: The focus that I take within developmental psychology is I am interested in how all of the rich cognitive structure and abilities that we have come to be in a human mind. Language is a really interesting case because it is clearly both very much a product of our evolution, but at the same time, it's got to be richly cultural and learned in the following sense. No species other than human beings learn languages. Other animals don't do that. I want to understand that. And then secondly, human beings speak wildly diverse languages. That involves not only the differences in individual words, but also in how words come together. How tenses are marked and the kind of clauses that you can create. What makes us different seems to be two things. One is our social capacity—our social reasoning—and the other is language.
Q: What was your process in deciding a focus? Did you pursue any other subfields along the way?
A: The alternate pathways I was thinking of were not psychology-related. When I went to the University of Chicago, it was two-thirds thirds male and it was a hard ass place. I was very attracted to that at that age, I wanted to learn everything and know everything. But, somewhere along the way, when I was there, I realized that I was more interested in growing and developing socially. I dropped out of college and began teaching preschool at that point. Eventually, I realized that I was not going to be able to make my living as a preschool teacher without a B.A., and I needed a better life plan to be able to pay my rent. That’s where the idea of going back to school and getting a bachelor’s degree and then a master’s degree in social work kicked in. I joined the evening degree program at the University of Washington, and at the time, I did not want to go to graduate school, but I only considered the idea when the professor I worked on psycholinguistics with suggested it. I don't think it's the right route for everybody. But I do think that it's a life that really suits some people. When I made the choice to become a psychologist, the biggest decision for me was whether I wanted to do something that was immediately practical, which would immediately help children and influence their lives, or whether I wanted to do something that was driven more from intellectual passion. Now, I do both.
Q: How did the work you did prior to creating the Snedeker lab influence the type of projects you now create?
A: As an undergraduate, I was in an EEG lab working with adults, capping, which is basically setting up the EEG electrodes for recording, and working on stimuli. I realized I was more interested in the child and where language developed. At the point where I applied to graduate school, I applied to programs where people were working on language acquisition, but also where they were working on language comprehension. They are two very different worlds. And I would go to these places and I would say, I would like to study how children understand sentences moment to moment. And they said, you can't really do that.
I studied word learning, how children learn words. But then during the time that I was in graduate school, people started doing something kind of interesting. They started doing these studies where they would have people listen to language while they had a display in front of them. Usually it was a set of objects in the physical world. They would track their eye movements to these objects as they talked about them. So you might say something like, “Look at the cup,” when you have a cup and a phone in front of you. What would happen is about 200 milliseconds after you hear that sound “c” in “cup,” people would start to look over to the cup, suggesting that they're moment by moment. While I was in graduate school, I started working with John Trueswell at University of Pennsylvania doing this. I spent the next roughly 20 years of my life doing these kinds of studies where we would twist and turn the sentences in various ways to study how children were building up certain structures.
Around the time that I got tenure, I started thinking, you know, there's a lot of things we can't study this way. We can only study language in a context where there's just a couple of objects around where it's really constrained and really limited. And I start going back and thinking about the EEG. Right. And what we might be able to learn if we had come up with EEG studies that kids actually wanted to participate in, ones thatwere fun and ones that were engaging. That's how we got to the point of focusing on this paradigm where we have children listen to stories instead of disconnected sentences. Natural stories that we can then manipulate or explore in different ways.
Q: How would you, as an individual, define developmental psychology?
A: I’m not so interested in defining the entire field. Like I said, there are many different communities coming from really different perspectives here. I don’t want it to seem like I’m telling people they can’t use the label. They’re welcome to it, right? The part that I am interested in is this part of psychology that focuses on the origins of knowledge: Where does our knowledge come from? What are the constraints on it? How do children build it up over time? And so the developmental group at Harvard and our psychology department, it's tiny and it's really focused. All of us are studying this question of the origin of knowledge.
Q: What are your thoughts on the direction that technology, more specifically AI, is going in the field of psychology?
A: I'd say there's three exciting things about AI as it intersects with my work. One of those kinds of obvious, and I think it's true for everybody, which is that it gives us tools to do the kinds of things that we've always done faster and better. One really clear example of this is that in my lab, we are often using something we call cloze probability. How predictable a word is as a measure of what's going on neurally. This is expensive. It takes a lot of people's time. At this point, the correlations between those kinds of cloze values and what is being produced by LLMs are so high that I am comfortable in norming stimuli in this way. There are many, many other cases.
That's one kind of thing. But the second thing is that I'm somebody who studies language. The structure of language and how language is acquired. So I've got a specific interest in AI for two reasons. One is that the sudden success of large language models has created a real interest in the wider community in how language is learned and where language comes from. I don't think that the large language models give us an answer to either of those questions, but that curiosity is invaluable.
Q: What is the hardest obstacle you’ve faced in your work, and how did you overcome it?
A: The obvious ways to study the origins of language are impossible. Two ways you might want to go about it: You might want to go back and look at human beings as they evolve and discover what happens to their language. Can't do that. And there are six million years separating us and the most recent, you know, common ancestor of chimpanzees. So looking at them is not going to be particularly helpful. The second thing you might think to do is take a child and raise them in isolation or completely control the input of speech that this child gets. We are not allowed to do that for very good reasons. One of the difficulties for everybody who wants to study the origins of language is how do you answer this question when the obvious levers are taken away from you? I can say a few things about that. What me and people in my profession have done largely is to look at the natural experiments that occur with human beings and the situations they put themselves in.
One of the most valuable methods for me has been looking at people who do not have any access to an external language and seeing what kind of communication systems they create. This is a study of what is called "home sign." These are deaf people with hearing parents who are never exposed to sign language and cannot hear the spoken language around them. About twelve to fifteen years ago, I got involved with a group in Nicaragua who were working with home signers and users of Nicaraguan Sign Language. They have to create themselves. A second kind of natural experiment that has given me some leverage is focusing on children who are internationally adopted.
Q: What is your favorite study you’ve ever conducted?
A: One of my favorite studies is this notion that you can put a sentence inside another sentence, this property of recursion. It's often argued to be one of the fundamental properties of human language that makes it different from anything else. Animal communication says they're not recursive. In fact, Noam Chomsky argued that that is the only thing that makes a language a language is it has the property to create recursive structure. So we decided to go after individual people who are creating their own languages de novo. So the case of Nicaraguan Sign Language, whether they had the capacity for recursion by eliciting relative clauses from them. Short answer, they do. They can produce a little sentence inside of another sentence. And I love just the simplicity of this, the fact that it was a case where the baby and the child methods that we have developed gave us a new answer about adults and linguistic diversity and where it is that this basic capacity of human thought comes from.
Q: How has the Snedeker Lab adapted to the recent funding cuts that research institutions have faced around the nation? How strongly do you think this impacts your work?
A: In my field, grants tend to last for a very short period of time, about two to three years. All of our grants were coming to an end during this period. We had our new applications out. For some of them, we’re still waiting for an answer. I have ones that have been in limbo for about a year now, and I'm not completely sure why that is.
So what that means is that we are not operating with the same kind of resources that we have had for the ten years or so previous to this. We're doing a few different things. I’m very grateful for the fact that EEG is a relatively inexpensive paradigm. We can keep going on very, very small amounts of money. And I'm grateful for that. A second thing to point to is that this has gotten us thinking a lot about international collaborations. So this summer, for the first time in a very long time, we're not going to have our summer internship program and stay here on campus collecting data because we don't have the funding for that. But, we are all heading out into the world and to different places to discuss work with collaborators that will hopefully keep this research going, even if we can't keep it going at Harvard. For example, I'm very excited about a collaboration with Waseda University in Japan where we're going to look at moment-to-moment language comprehension in children who are hearing Japanese. Japanese is fascinating because, like a lot of the world's languages, it is actually kind of the opposite of English. In English, we put our heads before our compliments. A great example of this, we put our verbs before our objects. This gives a chance to look at how prediction happens in kids where the information order is completely reversed.
Q: How has your background in psychology influenced your criteria while looking for people who work in your lab? What personality traits and backgrounds are important while looking for researchers?
A: You should expect people to be coming with different skill sets, right? For a graduate student or postdoc, this process is a lot about fit. Is this someone for whom my lab might well be the best place for them to be, right, where we're uniquely suited to one another? What this typically means is somebody who wants to explore a question within the domain where I might be able to add something uniquely to that question. In some cases, it's really kind of obvious.
We're one of the few labs that are doing developmental EEG focus on language at higher levels. If someone is interested in that, this is a great fit. For another group of students, this is about the fact that we are a lab that uses the methods of psychology, right, to explore linguistically relevant questions. That's one set of criteria.
Another set of criteria is, is this someone who's ready to make this choice? This is particularly relevant for graduate students. Not everyone should be going to graduate school, right? Do they know about the kind of work that's going to be involved? Is this a career path that they're fairly committed to?
I’m looking for whether people seem to be able to think deeply about data, about the relationship of research to theory and going back and forth between those things. I spend a lot of time with great care interviewing of graduate, potential graduate students. And that process usually involves several interviews and one in which we discuss research design fairly tightly. For undergraduate researchers, I'm looking for somebody who is curious, who is interested in staying with the lab for a couple semesters.
About the Author
Sucheta Srikanth ('29) is a freshman at Harvard College concentrating in Psychology (MBB) and Statistics.
Q: How did you decide that you wanted to work in the field of psychology, and more specifically, development? How did this area lend itself to language?
A: My first lab experience was with an EEG lab that studied moment-to-moment language comprehension, our understanding of words as we hear them. As I was talking to the professor in that lab, he said, “It really seems to me that a lot of the questions you want to ask are being studied over in linguistics. It might be more about the development of language.” So based on that, I went over and talked to the folks in linguistics. And I guess for me, the decision was whether to go into linguistics or psychology. I'd settled on studying how languages are learned by children.
Q: What about the general development process made you curious about language acquisition?
A: The focus that I take within developmental psychology is I am interested in how all of the rich cognitive structure and abilities that we have come to be in a human mind. Language is a really interesting case because it is clearly both very much a product of our evolution, but at the same time, it's got to be richly cultural and learned in the following sense. No species other than human beings learn languages. Other animals don't do that. I want to understand that. And then secondly, human beings speak wildly diverse languages. That involves not only the differences in individual words, but also in how words come together. How tenses are marked and the kind of clauses that you can create. What makes us different seems to be two things. One is our social capacity—our social reasoning—and the other is language.
Q: What was your process in deciding a focus? Did you pursue any other subfields along the way?
A: The alternate pathways I was thinking of were not psychology-related. When I went to the University of Chicago, it was two-thirds thirds male and it was a hard ass place. I was very attracted to that at that age, I wanted to learn everything and know everything. But, somewhere along the way, when I was there, I realized that I was more interested in growing and developing socially. I dropped out of college and began teaching preschool at that point. Eventually, I realized that I was not going to be able to make my living as a preschool teacher without a B.A., and I needed a better life plan to be able to pay my rent. That’s where the idea of going back to school and getting a bachelor’s degree and then a master’s degree in social work kicked in. I joined the evening degree program at the University of Washington, and at the time, I did not want to go to graduate school, but I only considered the idea when the professor I worked on psycholinguistics with suggested it. I don't think it's the right route for everybody. But I do think that it's a life that really suits some people. When I made the choice to become a psychologist, the biggest decision for me was whether I wanted to do something that was immediately practical, which would immediately help children and influence their lives, or whether I wanted to do something that was driven more from intellectual passion. Now, I do both.
Q: How did the work you did prior to creating the Snedeker lab influence the type of projects you now create?
A: As an undergraduate, I was in an EEG lab working with adults, capping, which is basically setting up the EEG electrodes for recording, and working on stimuli. I realized I was more interested in the child and where language developed. At the point where I applied to graduate school, I applied to programs where people were working on language acquisition, but also where they were working on language comprehension. They are two very different worlds. And I would go to these places and I would say, I would like to study how children understand sentences moment to moment. And they said, you can't really do that.
I studied word learning, how children learn words. But then during the time that I was in graduate school, people started doing something kind of interesting. They started doing these studies where they would have people listen to language while they had a display in front of them. Usually it was a set of objects in the physical world. They would track their eye movements to these objects as they talked about them. So you might say something like, “Look at the cup,” when you have a cup and a phone in front of you. What would happen is about 200 milliseconds after you hear that sound “c” in “cup,” people would start to look over to the cup, suggesting that they're moment by moment. While I was in graduate school, I started working with John Trueswell at University of Pennsylvania doing this. I spent the next roughly 20 years of my life doing these kinds of studies where we would twist and turn the sentences in various ways to study how children were building up certain structures.
Around the time that I got tenure, I started thinking, you know, there's a lot of things we can't study this way. We can only study language in a context where there's just a couple of objects around where it's really constrained and really limited. And I start going back and thinking about the EEG. Right. And what we might be able to learn if we had come up with EEG studies that kids actually wanted to participate in, ones thatwere fun and ones that were engaging. That's how we got to the point of focusing on this paradigm where we have children listen to stories instead of disconnected sentences. Natural stories that we can then manipulate or explore in different ways.
Q: How would you, as an individual, define developmental psychology?
A: I’m not so interested in defining the entire field. Like I said, there are many different communities coming from really different perspectives here. I don’t want it to seem like I’m telling people they can’t use the label. They’re welcome to it, right? The part that I am interested in is this part of psychology that focuses on the origins of knowledge: Where does our knowledge come from? What are the constraints on it? How do children build it up over time? And so the developmental group at Harvard and our psychology department, it's tiny and it's really focused. All of us are studying this question of the origin of knowledge.
Q: What are your thoughts on the direction that technology, more specifically AI, is going in the field of psychology?
A: I'd say there's three exciting things about AI as it intersects with my work. One of those kinds of obvious, and I think it's true for everybody, which is that it gives us tools to do the kinds of things that we've always done faster and better. One really clear example of this is that in my lab, we are often using something we call cloze probability. How predictable a word is as a measure of what's going on neurally. This is expensive. It takes a lot of people's time. At this point, the correlations between those kinds of cloze values and what is being produced by LLMs are so high that I am comfortable in norming stimuli in this way. There are many, many other cases.
That's one kind of thing. But the second thing is that I'm somebody who studies language. The structure of language and how language is acquired. So I've got a specific interest in AI for two reasons. One is that the sudden success of large language models has created a real interest in the wider community in how language is learned and where language comes from. I don't think that the large language models give us an answer to either of those questions, but that curiosity is invaluable.
Q: What is the hardest obstacle you’ve faced in your work, and how did you overcome it?
A: The obvious ways to study the origins of language are impossible. Two ways you might want to go about it: You might want to go back and look at human beings as they evolve and discover what happens to their language. Can't do that. And there are six million years separating us and the most recent, you know, common ancestor of chimpanzees. So looking at them is not going to be particularly helpful. The second thing you might think to do is take a child and raise them in isolation or completely control the input of speech that this child gets. We are not allowed to do that for very good reasons. One of the difficulties for everybody who wants to study the origins of language is how do you answer this question when the obvious levers are taken away from you? I can say a few things about that. What me and people in my profession have done largely is to look at the natural experiments that occur with human beings and the situations they put themselves in.
One of the most valuable methods for me has been looking at people who do not have any access to an external language and seeing what kind of communication systems they create. This is a study of what is called "home sign." These are deaf people with hearing parents who are never exposed to sign language and cannot hear the spoken language around them. About twelve to fifteen years ago, I got involved with a group in Nicaragua who were working with home signers and users of Nicaraguan Sign Language. They have to create themselves. A second kind of natural experiment that has given me some leverage is focusing on children who are internationally adopted.
Q: What is your favorite study you’ve ever conducted?
A: One of my favorite studies is this notion that you can put a sentence inside another sentence, this property of recursion. It's often argued to be one of the fundamental properties of human language that makes it different from anything else. Animal communication says they're not recursive. In fact, Noam Chomsky argued that that is the only thing that makes a language a language is it has the property to create recursive structure. So we decided to go after individual people who are creating their own languages de novo. So the case of Nicaraguan Sign Language, whether they had the capacity for recursion by eliciting relative clauses from them. Short answer, they do. They can produce a little sentence inside of another sentence. And I love just the simplicity of this, the fact that it was a case where the baby and the child methods that we have developed gave us a new answer about adults and linguistic diversity and where it is that this basic capacity of human thought comes from.
Q: How has the Snedeker Lab adapted to the recent funding cuts that research institutions have faced around the nation? How strongly do you think this impacts your work?
A: In my field, grants tend to last for a very short period of time, about two to three years. All of our grants were coming to an end during this period. We had our new applications out. For some of them, we’re still waiting for an answer. I have ones that have been in limbo for about a year now, and I'm not completely sure why that is.
So what that means is that we are not operating with the same kind of resources that we have had for the ten years or so previous to this. We're doing a few different things. I’m very grateful for the fact that EEG is a relatively inexpensive paradigm. We can keep going on very, very small amounts of money. And I'm grateful for that. A second thing to point to is that this has gotten us thinking a lot about international collaborations. So this summer, for the first time in a very long time, we're not going to have our summer internship program and stay here on campus collecting data because we don't have the funding for that. But, we are all heading out into the world and to different places to discuss work with collaborators that will hopefully keep this research going, even if we can't keep it going at Harvard. For example, I'm very excited about a collaboration with Waseda University in Japan where we're going to look at moment-to-moment language comprehension in children who are hearing Japanese. Japanese is fascinating because, like a lot of the world's languages, it is actually kind of the opposite of English. In English, we put our heads before our compliments. A great example of this, we put our verbs before our objects. This gives a chance to look at how prediction happens in kids where the information order is completely reversed.
Q: How has your background in psychology influenced your criteria while looking for people who work in your lab? What personality traits and backgrounds are important while looking for researchers?
A: You should expect people to be coming with different skill sets, right? For a graduate student or postdoc, this process is a lot about fit. Is this someone for whom my lab might well be the best place for them to be, right, where we're uniquely suited to one another? What this typically means is somebody who wants to explore a question within the domain where I might be able to add something uniquely to that question. In some cases, it's really kind of obvious.
We're one of the few labs that are doing developmental EEG focus on language at higher levels. If someone is interested in that, this is a great fit. For another group of students, this is about the fact that we are a lab that uses the methods of psychology, right, to explore linguistically relevant questions. That's one set of criteria.
Another set of criteria is, is this someone who's ready to make this choice? This is particularly relevant for graduate students. Not everyone should be going to graduate school, right? Do they know about the kind of work that's going to be involved? Is this a career path that they're fairly committed to?
I’m looking for whether people seem to be able to think deeply about data, about the relationship of research to theory and going back and forth between those things. I spend a lot of time with great care interviewing of graduate, potential graduate students. And that process usually involves several interviews and one in which we discuss research design fairly tightly. For undergraduate researchers, I'm looking for somebody who is curious, who is interested in staying with the lab for a couple semesters.
About the Author
Sucheta Srikanth ('29) is a freshman at Harvard College concentrating in Psychology (MBB) and Statistics.