Eleanor Jack Gibson: A Life in Science
Date Posted:
April 12, 2023
Date Recorded:
April 11, 2023
CBMM Speaker(s):
Elizabeth Spelke All Captioned Videos Brains, Minds and Machines Seminar Series
Description:
Elizabeth Spelke, Harvard University
Abstract: More than two decades after her death, Eleanor Gibson still may be the best experimental psychologist ever to work in the developmental cognitive sciences, yet her work appears to have been forgotten, or never learned, by many students and investigators today. Here, drawing on three of Gibson’s autobiographies, together with her published research and a few personal recollections, I aim to paint a portrait of her life and science. What’s it like to be a gifted and knowledgeable scientist, working in a world that systematically excludes people like oneself, both institutionally and socially? What institutional actions support such people, both for their benefit and for the benefit of science and its institutions? In this talk, I focus primarily on Gibson’s thinking and research, but her life and science suggest some answers to these questions and some optimism for the future of our fields.
Bio: Elizabeth Spelke is the Marshall L. Berkman Professor of Psychology at Harvard University and an investigator at the NSF-MIT Center for Brains, Minds and Machines. Her laboratory focuses on the sources of uniquely human cognitive capacities, including capacities for formal mathematics, for constructing and using symbols, and for developing comprehensive taxonomies of objects. She probes the sources of these capacities primarily through behavioral research on human infants and preschool children, focusing on the origins and development of their understanding of objects, actions, people, places, number, and geometry. In collaboration with computational cognitive scientists, she aims to test computational models of infants’ cognitive capacities. In collaboration with economists, she has begun to take her research from the laboratory to the field, where randomized controlled experiments can serve to evaluate interventions, guided by research in cognitive science, that seek to enhance young children’s learning.
JOSH: Hi, everybody. Welcome to this Quest CBMM seminar. We are very, very privileged to have Elizabeth Spelke here talking about a special topic. So some of you might have seen her talk last week. She gave the Bidwell lecture in which she was talking about a summation of many things in her career and the next stage, and her big vision of how to think about the mind and how it develops.
That's not what she's going to be talking about today. Rather, she's going to be talking about the really important work and big vision of somebody else who was one of her mentors. I'll let her tell you about that.
But Eleanor Gibson-- well, I should say what I know about Eleanor Gibson is mostly what I know from Liz, which I think will be at least 10 times more after this talk, because I don't know that much. And that's part of the point of your talk, I think. [LAUGHS]
But from what I gather and from what I understand, she is a towering figure in our field. And like many, not as well known as she should be, both her life and her work.
But we all know about Liz's work. And I'm not going to try to repeat the introduction that Nancy gave where she went through all of her many honors. As Nancy said last week, she's won every possible award, which is literally true in our field. [LAUGHS] And done so much to, I think, more than anyone alive really, honestly. And this is an admittedly biased perspective, but a true representation in my perspective, to give us a picture of what the human mind is like in its very earliest stages, how it starts.
But I know, and she's going to tell us about this, I hope somewhat, that none of us-- none of us start from nothing. [LAUGHS] And so for me, it's a great privilege to learn from her about one of her main mentors. And I'm sure you'll talk about some of her influence on your work, and as well as on the entire field.
And I think all of us, understanding the stories of people, where ideas and people come from, that's one of the most interesting and important things that I think we can do in our field. So I think it's an incredible honor to have such a distinguished scientist herself here to tell us about such an earlier and incredibly distinguished scientist. And so I'll just turn it over to Liz with that for the story.
[APPLAUSE]
ELIZABETH SPELKE: Thanks, Josh. OK, can you all hear me? I understand that at the talk last time my mic was going in and out. Apologies to any of you who were going crazy and too polite to tell me. But if it happens again today, let me know, OK?
So I want to start with a couple of caveats. This talk is not only nothing like the talk that I gave last week, for those of you who were here. It's nothing like any talk I have ever given in my life, or ever expected to give.
The talk I gave last week is a pretty typical example of a talk that I would give. One, it runs over time. Two, it's crammed with findings, particularly findings that came out last week that are still kind of percolating. And it ends with a conclusion, which is my best attempt to kind of pull together the findings that it was talking about.
OK. This is a completely different kind of talk. It's a historical talk. It's not about me or my work at all. It's about the life and work of Eleanor Gibson. And it's a talk that I never intended to give to an audience like this. And maybe the one thing I'll say about me is how this talk came about.
So back in 1921, the Psychology department at Smith College was founded. And in 2021, they hit their 100th birthday, and they were going to have a centennial celebration, but it got delayed a year because of the pandemic. And they contacted me to tell me about that they were having this celebration of their 100th birthday in 2022, and they wanted to do a daylong seminar on the-- featuring the life histories of some of the eminent people who had both been either Smith College alums or had taught on the Smith faculty.
Among those eminent faculty members were Kurt Koffka, who got rescued from Germany by Smith College arranging for him and his 400 followers, or some very large number of followers anyway, in the Gestalt movement to get visas to come to the US. Smith gave him a faculty position, so he was one of the people who was recognized on that day.
James Gibson was-- they wanted me to talk about Eleanor. And I put up there Fritz Heider because there were many, many other people who-- students who went to Smith, which is to say, undergraduate women were put in contact with, actively in contact with. They were there.
I think most of you know the Heider and Simmel animations. Well, Anne Simmel was a Smith undergraduate. That film was made at Smith College. Heider was hanging out there. He didn't have a regular academic position, but they got him connections that got him a place to do his research and interact with students, and so forth. So Smith was clearly a really interesting place, and I was excited to go there and learn more about it.
Another reason I was excited to go there was that I had noticed over the years that an awful lot of really talented women who have gone into the cognitive-- brain and cognitive sciences came out of Smith. These are some examples. I'll say just a minute about Susan Goldin, now Susan Goldin-Meadow, who as an undergraduate at Smith, her advisors at Smith arranged for her to spend a year in Geneva doing research with Jean Piaget, Hermine Sinclair, and Annette Karmiloff-Smith.
So there were fairly extraordinary experiences that people could have as Smith undergraduates. And they've done-- all of the people here and many others have done extraordinary things in science since then.
The other thing that I didn't know about until I got to Harvard was one of the most eminent Smith undergraduates who went on to a career in psychology, and that's Mary Calkins. Mary Calkins-- I learned about her when I got to Harvard, because there was a move at Harvard to give this woman a PhD in 2002 or something like that, OK? She was dead at that point, but she didn't have a PhD. And the reason why she didn't have a PhD was kind of interesting.
So she had been an undergraduate at Smith. She then went on to do graduate work with William James. She did a dissertation. For her dissertation, she invented a new method for psychology. It was called paired associate learning. There was a time when everybody in psychology was doing paired associate learning.
She passed her thesis defense with flying colors. The paperwork was sent in for her PhD, and it was rejected by the Harvard Corporation on the grounds that Harvard only gave PhDs to men.
So James protested. He pushed back. And Harvard did a perfectly ridiculous thing. They decided, OK. She can have a PhD, but she can't have a PhD from Harvard, because Harvard is only for men. We'll give her a PhD from Radcliffe College.
Now, Radcliffe College was a women's college, undergraduate degrees only. Like, a peer/competitor with Smith. She had never set foot in Radcliffe College. The idea that Radcliffe College would give anybody a PhD was kind of ridiculous, and she turned it down, all right? So she spent her whole career active in psychology on the strength-- with the only credential being her bachelor's degree from Smith.
Now, with that bachelor's degree from Smith, and of course her education at Harvard, she went on to do lots of experiments. She was the president of APA at one point. And I particularly wanted to bring her up here for CBMM because she spent her life after getting the PhD as a faculty member at Wellesley. And while she was at Wellesley, she founded the Department of Psychology and was its first chair.
So all of you who have had the joy of either being a Wellesley student who got into the CBMM sphere through that or of working with students in that category, you have Mary Calkins and Smith in part to thank for that. OK. That is all I will say about Mary Calkins.
Oh, no. I will say one more thing about Mary Calkins. What do you think happened to our move? We unanimously voted to give her a PhD. Harvard turned it down.
[LAUGHTER]
She still posthumously, as during her lifetime, does not have a degree. It was too far outside the Harvard traditions to give a degree to somebody posthumously, was the way they put it. I don't know. It worked its way up the food chain all the way to the Corporation, and there it got killed. So she still doesn't have a PhD. I don't think she minds anymore. Anyway.
So I decided I'd give this talk. Smith was too interesting a place to pass up. And I estimated it would take me only about a day to prepare it, because after all, Gibson had been one of my mentors. So as her student, I had read her science books. It had been a long time ago, but still, I had read them. I figured I could review them pretty quickly.
And I knew she had also written one very long essay on her life sort of mid-stream, and then another essay at the end of her life. And in addition, in this book that's published by MIT Press, it's a selection of her readings. But there's a lot of autobiographical material in that book. She kind of gave-- she at the time of publication wrote introductions to each of these lines of research and commented on how they came about, and so forth. So I knew there'd be a treasure trove of information in there.
And I thought, OK, I'll be able to pull this all together in a day. And I got totally obsessed with her life over the next two weeks. I could think of nothing else. I was plunged back into the 1920s and '30s, and really, it was all that I could think about.
And looking back on it now, I think there were three reasons that I found her life so interesting. And I'll put them right up front, because if you're not interested in these reasons, I will forgive you if you leave now. These are the things I really hope to be pulling out of this talk.
First, it was a story of successful living under adversity. And here, the reason this struck me as so interesting is that I had come to MIT as a faculty member in 1996. This was just at the time when the MIT report on treatment of women in science at MIT had come out. It was a time I thought of great optimism in that there were concrete steps being taken, or in some cases that had already been taken, to redress the inequities and the resources and positions of women at MIT.
But then in 2001, I moved to Harvard, and so did a new president. And a couple of years later, I found myself involved in all of these discussions about, oh, maybe there shouldn't-- we shouldn't have so many women in science because they're not as mathematically talented as men are, and things like that.
And I made a personal discovery at that point. I discovered that the more involved I got in the issue-- and I did get involved in the issue, because at that point at Harvard, the main work in my lab was looking at number concepts, concepts of number and geometry in infants and how they supported learning of mathematics. And NIH always made us do analyses by gender, and we never saw any differences, so I was kind of plunged into that issue.
But I discovered that the more time I spent in discussions about inequities in science involving women-- that is to say, inequities that kind of involved me personally-- the less fun I was having in my science. And I eventually decided, OK. I'm going to take a break from all that. I'm going to go back to my science. You can change the world of science or you can love the science that you're doing, but I thought you couldn't do both at once.
And then I had this spectacular counterexample from Eleanor Gibson. Now, her life in science was a zillion times more challenging than mine as you are-- or anybody else's, I think, who was at MIT at the time, as I think you'll see over the course of this talk. But she loved every minute of what she was doing. And her love and enthusiasm for the work that she was doing comes through on every page of her autobiographies, and it came through in real life to those of us who were lucky enough to know her, OK?
So this for me raised this huge question of, how do we create environments in which people from underrepresented groups not only get their inequities resolved, but get to have a good time in the process, right? Get to be engaged in their work and supported so they can throw themselves into it, forget everything else, and have a fabulous time? OK. So that's question 1.
The second reason I got obsessed by this story was that learning more about Gibson through her autobiographies meant learning more about Smith. And here, it kind of upset my ideas about how great academic institutions come to be.
I think my assumption was that to be a great institution, you want to have a long-- you want to have a long tradition and a lot of money, and you also probably want to be near a place where, like, spinoffs of science and technology and business and things like that are happening. And then once you start being a major institution, it snowballs because then it's easier for you to get good people to come there, and then other people want to come there to work with them, and so forth, OK?
Well, Smith is a spectacular counterexample to that rule, OK? So in the '20s and the '30s at least, and I think for some time after, really great psychology was done at Smith. Gibson reports that when he graduated-- James Gibson, when he graduated from Princeton, there were three great places to go to study perception. There was Harvard, there was Yale, and there was Smith, OK? So it was really on the map at that point.
But what was it? It was a small college. It was undergraduates and master's students only. They weren't giving out any PhDs then or now. It was for women only. Places like Harvard and Yale weren't even letting women in, right? Women weren't even part of-- on the radar screen for getting educated. Plus, it was in the middle of nowhere, right? In the Pioneer Valley in Massachusetts, away from all the urban centers. And for me, that just raised the question, what made Smith so great?
And then the third reason I got interested in it is I think it's a story of how to do good science that might serve as an interesting counternarrative to the reproducibility movement that we've all been living through. Now, that's not to say that the reproducibility movement has not been a good thing. I think it's been a very good thing. But I think it suggested a picture of science that doesn't quite gel for me.
The picture is that what scientists really want is to be famous, get their next grant, get tenure, or win awards, stuff like that, rather than, what are science is really motivated by is curiosity. Like, this insatiable curiosity. I want to understand how this system works.
So I was also-- some of what spun off of this movement hit very close to home because there were a lot of people looking very, very hard at studies of infants that were based on what I thought were the ab-- and still think are the absolute best measures to use if you're going to do behavioral studies of infants. Infants are born to learn, and they learn by observing the world, exercising their capacities for perception and exploration and learning. That's how they get off the ground. So those behaviors are the ideal ways of seeing both what they're seeking to learn and what knowledge is guiding that search.
But these methods had come under intense criticism. And in reading about Gibson, I was reminded that it was her generation that developed these methods and used them to spectacularly good effect. So I started doing things like asking-- early in the reproducibility movement, I started doing things like asking in classes, raise your hands if you've heard of the visual cliff. Raise your hands if you've heard of Eleanor Gibson.
And not all that many hands went up. So I thought, OK. People need to be reminded about this work because it's a way of doing really high-quality science. OK. That's the end of the prelude. Let's go into Gibson's life.
Oh, one of the ways-- one of the things I'm going to be doing today is you're going to see a lot of text on this screen. The text are direct quotations out of her books. She was a big Jane Austen fan, and she writes a little bit like her. She's a deliciously wonderful writer. So I thought you'd be happier reading-- hearing as much of this as possible in her words, not mine.
Now, she wrote almost nothing about her early youth. She grew up in Peoria, Illinois. This was the only thing I could find that she'd written about it in any of her autobiographical materials. "I will abbreviate the beginning of this tale because my early life was so traditional as to make very dull telling," which may be why she didn't write about it anywhere else.
"I was reared in an atmosphere of middle class respectability among dozens of kindly, staunch Presbyterian relatives. The only surprising thing to me is how I managed to break away so far from this background to emerge as an intellectual and academic at least, with ideas that would seem radical to my forebears. No one ever suggested to me I was a particularly bright child or even wanted me to be bright."
And then she goes on to say that, OK, she skipped a few grades in high school and started getting the idea that maybe she wasn't so stupid after all. And she got to Smith at age 16.
Now interestingly, in her first year at Smith she thought-- she initially thought she was going to go into literature, because that's what-- she doesn't really say this, but that was a common thing, I think, for women to do. But she was able to take a bunch of electives, and she took in her first year or two three psychology courses. Introduction to Psychology, Animal Psychology, and Developmental Psychology. Hold that in mind. She's 16 years old, and those are the subjects she wants to study. Animals, children, adult minds all together.
And then she declared a psych major and described Smith this way in retrospect. "Smith was a revelation and an instant joy to me, despite my provincialism. The greatest revelation was that Smith, a women's college, was a place where women were not only permitted but encouraged to be scholars, even scientists." OK? Like, that was unthinkable.
"That is said to be permissible today even at coeducational colleges like Cornell," where she was teaching at the time she wrote this. "But the atmosphere for real flourishing of an undergraduate woman who wants to be a scientist in particular still does not match Smith at that time."
At another point, she commented that one of the reasons it was great to do science there was that there were a lot of great women scientists there. But in other places, she also talks about Koffka and all the Gestalt psychologists and the other people who were there. But here maybe is her first-- her description of the first blessing that Smith gave her.
OK. So she took many other psychology courses, but the one that she writes in most detail about was given by a new 24-year-old faculty member, James Gibson, who appeared somewhere I think in her junior year. She took this class both semesters of her senior year, and here's her description.
"The course was small." It's called Advanced Experimental Psychology. "The course was small, just nine students, and very time-consuming. Each person with a partner did four experiments each term, getting her own subjects wherever she could. Amherst," which was an old men's college not very far away-- lots of socials between the two, was a favorite place to get subjects-- "and writing up the experiment complete with background.
I made some of the best friends of my life in that course. Everyone in it went on to do graduate work. One of the experiments I did with Gertrude Raffel was published. It was very exciting, and both of us felt we were budding scientists."
And then in another place she talks about what some of the other studies were about, including "experiments on reaction time, learning, memory, perception, adaptation to prisms, the gamut of the field of the time." But here's the key point. "Every experiment was a novel problem. Many papers came out of the work begun" that year because the experiments were addressing questions that hadn't yet been answered, OK?
I hope that resonates with some of you in CBMM. I feel like there's a lot of work here that's kind of breaking ground that hadn't been broken before, and that even undergraduates can become involved in part of the process of actually figuring out for the first time, not just repeating other people's classic experiments, but figuring out for the first time how something works. So this, I think, is the second blessing that she attributes to Smith.
I want to talk about just one of the experiments that she describes that she conducted that year. This is the experiment with Gertrude Raffel. So the background to this experiment was that Smith was kind of a mecca for nonmainstream ideas. But the mainstream ideas in experimental psychology at that point came from behaviorism, and particularly centered on notions of associative learning, of relationships between sensations and elementary motor reflexes, OK? That basically, learning was a process of associating sensations with-- elementary sensations with elementary movements.
And this was something that both the Gestalt psychologists and James Gibson in different ways were pushing back against. And Eleanor, Jack and Gertrude Raffel got together and decided to do an experiment to test whether, in fact, when people learn, what they're doing is associating elementary sensations with elementary actions.
So here's what she did. The study was run on adults. They were reading at the time they were doing the study to distract their attention from the events that the study was focusing on. But while they were reading, they had their hand on a box. And while they were reading with their hand on the box, they heard a bell. And shortly after the bell, the box delivered a shock, and they lifted their hand up in the air. That's what happened on the early trials. Of course, not surprisingly on the later trials, they would start lifting their hand as soon as the bell rang, OK?
So then came the critical trials. What happens if they touch the box with a finger of the opposite hand? Now, when we feel an elementary sensation in our right index finger, we don't also feel it in the left index finger. If we put the right index finger on a hot surface, we will not remove the left index finger from that surface, right? These are separate elementary actions and separate sensations.
But of course, the subjects in this study lifted the hand from the box as soon as they heard the bell, telling-- providing evidence that what people learned in more modern terminology was about the affordances of the box and the actions that would change the state of the world such that it was no longer shocking you. They weren't learning about associations between sensations, elementary sensations and elementary actions, OK? That's just one of the experiments that she did as an undergrad.
OK. So she graduated from Smith in 1931 with a love for an experimental psychology and a love for James Gibson. They got married the next year and faced a problem. How could both of their careers progress at the same time?
And here, Smith came with a third blessing. They offered Eleanor entry into a two-year master's program. She needed money to support herself at this point. This is the Great Depression that's going on. Her family couldn't afford to support her. So they gave her a teaching assistantship to defray her living expenses while she studied for a master's degree at Smith. And for four years, they lived there. She did research for the master's and she taught, and they were together.
But there was a limitation. Smith didn't have a PhD program, and wasn't about to pretend they did by making up a PhD degree. So she had to go somewhere else. And the closest and it looked like most promising place would be Yale.
The photographs that I got-- if I were a real historian, I'd probably have gone back to the archives or something, but I went on the web to look at places where they're selling old photographs, ancient photographs. So that's a photograph of Yale from-- a postcard of Yale from 1935, the library.
She went there with the goal of studying either comparative psychology or developmental psychology. And she wanted to do it through the kind of rigorous experiments that she'd learned to do at Smith. And she knew that-- she describes how Yale was organized at the time.
It was organized as a feudal system. You had the feudal lord of each area. So for comparative psychology, it was Yerkes. For developmental psychology, it was Gesell. For experimental psychology of learning and memory, it was Clark Hull. And doing a PhD meant working in the labs of one of those people, either with them directly or with one of their minions who were working under them. That was how Yale worked at the time.
So she went there. And her first desire was to do it-- do a PhD thesis in comparative psychology, but with developmental psychology as a backup. But she quickly discovered that she wasn't going to be able to do this because the reigning developmental-- or child psychologist was Arnold Gesell who did testing of kids, but didn't do experiments.
And the reigning comparative psychologist, Yerkes, did do experiments. But when she tried to get an appointment with him, it took three weeks for her to get one. And when she got one, it lasted three minutes because she came in and announced that she wanted to work with him, and he stood up and opened his door and said, there are no women in my lab. So she couldn't work in either of those areas.
She also had other problems. Yale-- they had decided that under very special circumstances they would accept women into their PhD program, unlike Harvard, OK? However, women were very limited in their abilities to operate on the Yale campus.
For example, they weren't allowed into the graduate student housing, and they weren't even allowed into the graduate library where all the psychology journals and materials were kept. So she would have to get some male student to go if she wanted to read a journal article and get it out of the library for her.
It was also not a great time to be poor because you had to pay tuition to go to Yale. If you were a man, you were eligible for fellowships, but there were no fellowships available for women.
And along came Smith again. They awarded her a fellowship that covered Yale's tuition and also a small stipend that went some distance towards paying for her room and board since she couldn't qualify for graduate student housing. But unfortunately, they could only offer this for one year, which meant that from the moment she arrived, she had exactly one year to fulfill all the residency requirements, get an advisor for her thesis, have a thesis prospectus meeting, and get it approved before the money ran out and she had to go back to Smith.
So she turned to Clark Hull. Clark Hull, whose theory was all based on associations of elementary sensations with elementary actions, which her work had been kind of undermining. But she found a path to-- a set of experiments that they could both live with, and completed all the residency requirements in one year and returned to the welcoming environment in Northampton of Smith College. Do you have a question, Jeff? No. OK. [LAUGHS]
So here comes Smith's fifth blessing, as she reports. They gave her research facilities in which to conduct all of her experiments for her dissertation, and they gave her a half-time position, this time as instructor with a higher salary, half time so that she would have half her time to devote to her thesis research and the other half time teaching at Smith.
So here's one of her thesis experiments. It used Calkins' paired associate recall method. Her participants were adults. They were trained to associate a different nonsense syllable with each of these figures. And then they were shown the larger set of figures, so she could look and see what the training generalized to.
And she finds generalization dependent far less on things like how many times a given word had been paired with a given picture, and far more on the shape similarity between the different forms. And then wrote up the dissertation in the following way. This, again, is direct quotes from her.
"The dissertation included a long theoretical paper linking the concepts of generalization and differentiation to paired associate learning, forgetting, and various transfer phenomena." So she's trying to combine Hull's concepts, which is in the second set, with her concepts in the first set.
"I felt wary of the concepts of extinction and reinforcement as Hull used them. I was not an SR psychologist at heart, I suppose, for all my attraction to animal research. I didn't believe then or now that external reinforcement in Hull's sense was applicable to perceptual learning. I thought the differentiation, once achieved, was not extinguishable. Once you see the dalmatian in the black and white moony figures, you don't extinguish it, right? The perceptual change is permanent, even though a subject was commanded to learn new responses to whatever was differentiated.
However," she ends, "it wasn't possible to say all this in so many words at that time." And I actually interpret that lesson in two ways. On the one hand, it wasn't politically possible. Her thesis was a compromise between two very different perspectives. But on the other hand, her perspective was also still developing. She was kind of at the beginning of her research career at this point, and not quite ready to say, how is it that perception can bring about permanent changes in how we experience the world, and through what kinds of processes do we perceive?
OK. So her sixth blessing came after the PhD was finally awarded at Yale, a full-time faculty position at Smith starting in 1938. She taught her own courses, which she loved. She and James bought a wonderful home. On the day of the 100th birthday party, we all took a little pilgrimage over to the home that she had lived in there, which was a short walk from the campus. Beautiful old, rambling building. Gorgeous garden in the back.
They started a family and they lived and worked happily until World War II intervened. In 1941, James Gibson and a whole bunch of other male faculty at Smith were called up by the army-- actually, the navy in Gibson's case, to join the research wing of the war effort, and they left first for Texas and then for California.
Eleanor stayed, and the other women on the faculty stayed on, taking over all the men's courses, teaching-- doubling at least their teaching load. But at the end of that year, Eleanor and James at that point had a son, and there was a daughter on the way, so she left her teaching job at Smith and went and was inactive in psychology during the war years.
All of the work that James was doing was classified, and she wasn't allowed to share in it. So it was a very fruitful period for him. He was developing all these ideas about optic flow, guiding planes, and the ground surface, guiding planes landing. And I'm sure they talked-- they must have talked about them together at night, but she wasn't actively involved in that research.
But then comes the final blessing. The war ended. And in 1946, Smith welcomed both Gibsons back now with two full faculty positions, lab space. So they were both independently there.
Now, that's the end of the narrative part that I'm going to give of Gibson's life. For the rest of it, I want to talk just about her science. But let me start-- well, her science and the personal context in which it happened. But let me start with a chronology so you know where we're going. Here's the timeline.
In 1946, as I said, the Gibsons returned to Smith as two full-time faculty members, able to continue their careers. In 1949, they moved to Cornell where James joined the faculty. But Cornell had an anti-nepotism rule that you couldn't have two married people on the faculty at the same-- at least certainly not in the same department. I think not at the same institution, OK? So there was no position for Eleanor.
Initially, all the work that Eleanor did at Cornell she did completely volunteer. She had no status for doing it. She had to be personally invited into other people's labs, usually on the condition that she run all their experiments for free, and then they would give her-- if she had any time left, she could run experiments of her own. Then later, she did get a research position there that didn't pay her any money, but allowed her to write grants in other faculty members' names so that they could pay her as a research assistant on the grant that she had written.
That continued until 1966. In 1966, she got a half-time-- by the way, she describes all this totally matter of factly. It's not at all, like, hand-wringing. Like, oh, this was a terrible situation. It's like, oh, I got this opportunity to do this work, and here are the conditions under which I did it. If there's hand-wringing going on, you're hearing it from me, not from her, OK?
Anyway, in 1966 she got a half-time professorship in psychology at Cornell with no lab. In 1972, she actually got a named chair, the Susan Linn Sage Professor of Psychology. And for the first time, a full-time professorship and space to start her own laboratory. (WHISPERING) Yay!
In 1975, Cornell activated its mandatory retirement rule. Now, here's what went down. I was around for this part of it. Here's what went down. The Supreme Court had declared, it is unconstitutional to require people to-- I forget on what grounds. Ageism maybe or something. You can't require people to retire at a particular age, so this was no longer legal.
However, the Supreme Court had only done it either this year or the year before, and they had a grandfather clause for universities to ease their financial planning. They had the option of lifting the rule now or the option of continuing the rule. So Gibson or others on her behalf petitioned for Cornell to lift the rule, given that she'd only been a faculty member for three years. And Cornell in its wisdom decided not to lift the rule, so she was forced to retire in 1975.
OK. So with her retirement, she'd lost a bunch of academic privileges. However, the psychology department did not take her lab away. She was still able to apply for grants and do the research in her lab, but she no longer was able to officially be an advisor. She became my unofficial advisor at that point, not an official advisor, or to teach classes, which she loved to do.
So in summary, I think that after the war, Eleanor Gibson was a full-time faculty member with her own lab for a grand total of six years, the three years she spent at Smith after the war and the last three years before her retirement at Cornell. For most of the remaining time, her research was in other people's labs. And she conducted it either on her own or with the faculty member who was the director of that lab, or sometimes with the students of other faculty members who were given permission in their spare time or something to be working with her on projects.
But what she was never able to do was select on her own, except for these six years, the problems that she was going to be working on, the approaches she was going to be taking, hiring the people to do it, selecting the students to come in and work on it all. She wasn't able to do any of that.
I'm bringing all of this up not to complain, but just to show-- but just to try to paint a picture of when you ask the question, how can somebody thrive in an environment, this is the environment I'm talking about in this n of 1 case, OK? She thrived in this environment. How did she do it?
OK. Let me take you through the research that she did at Cornell. The story started the year they got there, 1949. Her first and second loves were still comparative and developmental psychology, and she was thrilled to see that Cornell had an agricultural school.
And at the agriculture school, they had something called the Cornell Behavior Farm, which was a lab for research on animals that was run by someone named Liddell, I think Harold or Howard Liddell. I didn't look it up. Anyway, Cornell now has a Liddell Lab, so this guy obviously was a big deal at the university. But he ran the Cornell Behavior Farm.
The first thing that Eleanor reports learning is that Liddell never set foot in the animal behavior farm, but he had a brother-in-law who had no training in psychology who ran all of his experiments. Now, she wanted to run experiments on baby goats. And he said, well, I'll let you run them if first you run some experiments of mine, OK?
So the first experiment that Liddell wanted her to run was an experiment in which he took goats and gave them simply unavoidable shock and reported that they were acting psychopathic on getting the unavoidable shock. She looked at the animals and thought, they're not psychopaths. They're trying to get away from this shock. They're exploring, trying out different behaviors, to get away from this shock.
So she started an experiment that randomized the animals to get either unavoidable or avoidable shock, and discovered that, indeed, that's what they were doing. And when the shock was avoidable by one behavior or another, they would discover it and everything would be fine. And it was only the ones for whom there was no way to avoid it that would eventually, after trying out all these different behaviors, give up.
So she published a paper in the Journal of Comparative and Physiological Psychology in 1952 where she talked-- where she described her experiment and this phenomenon. Now, if it sounds familiar to you, it might be because the very same journal 15 years later published another paper by Marty Seligman and Obermeyer, whoever that is, called "Learned Helplessness," never citing her paper, by the way, but basically reporting the same phenomenon.
I was curious, so I went and looked and saw. They did a grand review of 50 years of research on learned helplessness that came out not that long ago. Again, not citing her. If they had cited her, it would have had to be 65 years of research, not 50. But anyway, one way or the other. So that was the first thing she did at Cornell. Not bad for a start.
Her next experiment, though, is the one that I think would really-- really promised to be gorgeous. And it covered territory that Harry Harlow was going to pursue much less effectively, I think, some years later, OK?
What she did was the following. She was interested in maternal behavior, and she wanted to know what the effects of maternal care on infant animals were above and beyond the obvious effects of the animals getting their genes and their nutrition from their mother, and so forth.
So she decided to look at this phenomenon in goats, because goats are born as twins. So she designed an experiment where you would take the two twins at the moment they were born, and one of them is randomized to be raised by the mother under normal conditions on the behavior farm, and the other one is going to get equally high-quality care and feeding and attention from a human who is taking care of-- a caregiver, OK?
So you've got one going to the mother, one not. In this way, you obviously control the genetic relationship between the two animals and the mother, their shared genes, but you also control their shared prenatal environment-- the prenatal environment of the two goats that were randomized to the different conditions, the shared prenatal environment, and also control for the shared environmental effects that might cause one mother to be different from another, because they both got the same mother. OK.
Really nicely designed study, which was never completed because the brother-in-law who was in charge of everything, when Eleanor Gibson came in one day very much like today, just after an Easter weekend, to discover that the animals in her control group had been given away for Easter dinner, OK? Now, that's not the part of the story that drove her-- I very rarely saw her get upset.
But the part of the story that got her upset was not what I've told you so far. It's what I'm going to tell you next. She would then go on to say, but you ruined my experiment. And the guy who ran all the experiments at the Behavior Farm said, no, I didn't. It was just the control group. We've got lots of other control animals here. Here, help yourself. Take them, right? She was infuriated that Cornell University had turned this glorious facility over to someone who was completely clueless about how to run a good experiment, OK?
So anyway, where are we? Yes. It ended with the killing of her control group. It would have been Harlow's experiment seven years earlier with, I think, a much better design. Certainly a more ethical one. And on that-- upon that, she decided she had to leave the Behavior Farm.
One more good thing did come out of her time there, though. While she was doing all of the work of control, rearing the guys in the control group, substituting for mom, she was also doing the work of assisting at the birth of these animals so that she could separate them into the two conditions right at birth, and she was there for the birth of all the animals.
But she had a veterinarian helping her. And it sounds like the veterinarian understood animals way better than the brother-in-law did. But anyway, the veterinarian would help with the delivery of the first goat and then give it to her and say, put that goat down, and either put him down on the ground or put him down on a little pedestal.
And she said, well, I can't put him down on a little pedestal. He's going to fall off. And the guy said, no, he won't. [LAUGHS] So that was her first observation, that a goat placed on an extended surface would move around on that surface. But indeed, the veterinarian was right. Place the goat on a small stool, and the goat would quite calmly, not afraid, but quite calmly stay still on the stool, not attempt to walk off it or fall off it. OK. So she made that observation, but she had to change fields and couldn't come back to it for a while.
The next thing she did was leave the studies of animals and of early development and go to studies of adults where the goal of these studies was to take classical questions in visual psychophysics out of the lab and into the open air. They called these open air psychophysics experiments, and I adore them. Let me just tell you about them quickly.
The participants in these studies, they were adults. They were recruits from a nearby navy base. The studies were run at Cornell on a big, wide, open, flat-- relatively flat field. This is a picture of the field.
The materials used in these studies were poles of varying lengths. You see some examples in the picture. A bicycle, a person who rode on the bicycle, and a whistle that they gave to one of the recruits who was the subject in the experiment. And now she would do experiments on size perception and distance perception.
So for example, in the real-world version of this-- I don't know if you can see it well, but there's a pole there in the distance, and you need to say which of these poles up close is the same height as the pole that you're looking at maybe, I don't know, 100 yards away or so. It's not showing up well. I'm sorry. But it's in the middle in the distance there between the two rows of poles. So size matching experiment.
But they also did distance experiments. And for the distance experiments, they used a guy on a bicycle. He would start at one pole, and then people would have to judge-- blow your whistle when he goes half the distance between that pole and where you are now. OK, so like a ratio judgment.
Or he would place poles at different locations and they'd have to judge which one was further away, different locations and directions. Or they would have to do absolute judgments. OK. You're on the bicycle. Ride the bicycle until you are 50 strides away from-- you reach a point that would be 50 strides away from where you started. And how well could people judge distance that way?
The findings from the studies provided evidence for the importance of cues to depth that you get in real scenes that you don't get from static pictures like optic flow. They did all the studies both ways with pictures and with real scenes. They also brought some of them into the lab and did studies with shadow casters so you only had the optic flow information, and people were using that. All new ways of studying space perception applied first to adults.
Here's her description. "The field experiments were very straightforward investigations" of the things I was just telling you about. "A lot of important muddles got straightened out such as the confounding of perceptual learning with response bias and the danger of generalizing from photographs to real space. It also became pretty obvious that young human adults were very skilled without training and making relative judgments of distance even over a large area," OK? And she felt in this way like she's finally starting to stake out a research enterprise in perceptual learning of the sort she was aiming for in her thesis work with Clark Hull, but with much better experimental methods.
She also around this time started moving toward a theory of perceptual learning. I see that time is running short, so I'm not going to read her quotations on this, but I will read the beginning of the second one. This is, I think, important, that the key notion is that "perceptual learning, it's a change in what's-- a permanent change in what's perceived that's best conceived of as increasing differentiation, that you become sensitive to distinctions that you weren't sensitive to before."
And she did a nice experiment with scribbles that's a little like some of the stuff you've done, Josh, where people are just shown one of these scribbles, and then that one scribble appears in a deck many different times along with the other scribbles that are up there. And the task is to them so you distinguish that one from the others. And initially, it's virtually impossible, but people get much better at it over time. And they found-- oh, and they did it not only as changes in practice with adults, but also in experiments with children where children got better at this over development as well.
They had a big debate with Jerry Bruner and Leo Postman on the nature of perceptual learning and what drives it. On their side, the idea that there's rich information that perception can be guided by, and we learn to take better advantage of it, and also to coordinate it with our actions, to perceive affordances of things for actions, against the notion that what changes over perception is you come to have opinions and beliefs about the world that you impose on your perceptual experience top-down, OK?
So in Bruner's case, the classic example of this that Gibson keeps coming back to is the new look in perception. So you're hungry and the hamburger-- or the picture of the hamburger looks bigger, OK? Or Molly Potter's study that she and Bruner did where you're slower to see that a photograph coming into focus is a fire hydrant if you adopt the hypothesis that it's a telephone booth, right?
And the Gibsons were really opposed to this idea that somehow perception is a matter of taking highly ambiguous information and projecting your own opinions or concerns onto it. She thought it was a matter of getting information about the world from the stimulus patterns that were available.
But one of the things-- the only thing-- you don't need to read this whole thing, but the only thing I'll try to highlight here is that when she-- when they debated these issues, it was in an extremely friendly and exhilarating way, right? She and Jimmy Gibson loved to debate. So the idea that someone would be challenging their views was really exciting and fun.
And this was actually another point where I got excited in reading Gibson's description of this because it suggested to me a connection that I had never made before. When Susan Carey and I first got to Harvard, people-- I would hear from back channels that people would have wondered to each other, how can we be in a program where the two people who run it hate each other? Look, they're arguing all the time, right? They have totally different ideas about how concepts grow, and they're just arguing about them all the time, right?
We love each other. It's really fun to argue with somebody you respect and to push-- who will push back on issues and make your arguments better and all of that.
But what I realized was that actually, Susan Carey was a very wayward child of Jerry Bruner, right? Bruner was her thesis advisor. I was an extremely wayward child of Eleanor Gibson. She never approved of the turn I took into asking questions beyond perception, about intuitive physics and kind of basic systems of knowledge of the world.
But somehow, we caught-- I think we may have caught something from our advisors about this love of debate and challenge. Anyway, let me go on. I really want to tell you about, I think, her most important work, and then I'll be done.
So the next thing she did after the work on open air psychophysics was to go to a new faculty member, Richard Walk, who was doing animal studies in a lab. So she didn't have to go to the Behavior Farm. He was actually a trained psychologist doing studies in the lab.
And the study that he was doing at the time she first visited his lab in the Psych Department was the study that looked at the effects of visual enrichment on an animal's perception of forms. And the way Walk was doing the study-- it was in mid-stream when Eleanor visited the lab.
The way he was doing the study is he had these objects of different shapes in this room that half of his rats got to explore. The other half of the rats got to explore an empty space with no objects in it. So one group is living in a space with lots of objects and the other group is living in an empty space. And he was then testing them on their perception of two-dimensional forms, circles like the circle in the back there or squares like that block thing in the middle, and so forth, OK?
So Gibson took one look at his displays and said, well, yeah. They probably-- they might well do better if they're in this enriched environment with the circles, but it might not be because of the visual information. They also get to crawl over these objects. These objects are enriching their environment in multiple different ways.
If you really want to see whether there's a visual effect, let's add a new condition to this experiment and dark-rear the animals, because if it's a visual effect, then it should go away with dark rearing. So they dark-reared a group of animals. The effect did not go away, showing that it wasn't just a visual effect, and Walk's grant proposal was in trouble because he had planned this study to be looking at visual experience and how it changes form perception.
But Gibson said, never worry. Don't worry. I have an idea. We've gone to all the trouble of getting these dark-reared animals. Let's see whether they can perceive depth.
So she took the dark-reared animals that were left over, hadn't yet been tested on the visual displays in the failed study, and built a visual cliff to put the animals on. The first version of the cliff used a totally flat surface, and they learned right away that if a dark-reared rat, nocturnal animal, is in a situation where the lights go on but it's still capable of using its whiskers to navigate, it will very readily go off the two sides of the visual cliff.
So she lifted the platform high enough so that the whiskers wouldn't touch the Plexiglas. By the way, I went very fast over all of this, but does everybody know what a visual cliff is? Plexiglas.
JOSH: Don't assume that.
ELIZABETH SPELKE: Don't assume that. OK. What you can't see here is there's glass. Actually, it wasn't Plexiglas yet. It was glass at this point. There's a glass sheet that covers this whole display. Immediately underneath that glass sheet is a patterned surface on one side, providing visual information that there's a surface of support there. There's also a patterned surface under the other glass sheet, but it's much farther below on the ground, OK?
So as you look out over one side, it looks shallow, like there's the surface just a step down. And as you look out over the other side, it looks like there's a surface quite a distance away, OK? The rat is on the center board, and the question is, which side is he going to descend on? And the finding of the first study was that both the dark-reared and the light-reared rats to equal extent descended on the shallow side and avoided the descent on the deep side, OK?
So with those findings, Gibson was finally able to write a grant in Walk's name to explore this in a whole variety of different animals and also in human infants. They found that all of the tested animals-- except turtles who looked over the deep side and probably thought, a pond, and went off, [LAUGHS] except for them, all of the animals descended preferentially on the shallow side, including the human infants as soon as they could locomote. Of course, pre-locomotor infants couldn't be tested.
Then she went on to ask, how was distance being perceived by varying the information that was available for distance? One source of information came from the fact that if you put the same pattern on a near side and a far side, it's going to project different images to the eye. The texture's going to be much smaller. The texture elements will be smaller and denser on the more distant surface.
So they tested for that by using two surfaces, one with denser patterns and one with less dense patterns, and the animals descended equally on the two sides. So it wasn't texture density that they were relying on.
Experiments where you covered one eye, showed that it wasn't binocular cues that they were relying on. And mostly by process of elimination at this point they concluded that it was optic flow that was guiding the rats' decisions about which side to cross on the visual cliff.
And I think this research led Gibson to three fundamental insights, so let me at least get these out before concluding. The first is that visual perception of surface distance is not learned, and that learning, when it does occur, is not a change in responding. It's a change in what we perceive in the situation she studied. I'd say in what we know in the situations all of you guys study, or most of you guys study.
My favorite experiment that they did, though, on the innateness of visual depth perception was not the studies on the rats. It was the studies on cats. Now, the cats were really interesting because like rats, they're mostly nocturnal, and they don't start navigating by vision at birth. They open their eyes sometime after birth.
And when you put them on the visual cliff, they do something different from the dark-reared rats. Even though they're on an elevated platform and they can't use their whiskers, they don't navigate by vision the first day that they're put on the visual cliff. On the first day-- I actually put up the data. They descend equally onto the deep and the shallow sides.
But what Gibson and Walk did was they continued to test these cats on the visual cliff every day for the next seven days. When the cats were on the visual cliff was the only time they ever had the opportunity to see a surface drop-off, a visual surface drop-off.
And because the Plexiglas was there, every time they walked over the deep side they got information that the deep side was safe, OK? They never fell down. There was a surface there. In fact, it felt the same as the surface on the shallow side. It was the same as the surface on the shallow side, right? But over time as their vision's improving, they start to avoid the deep side of the cliff, despite all the experience that they're getting, OK?
Let me give this conclusion to you in Gibson's words. "The cats had equal experience descending on the two sides in the beginning. They should have learned that descent to either side was perfectly safe. If one supposes that avoidance of the cliff is learned without reinforcement, how is the eventual behavioral preference acquired?
It seems most reasonable to adopt the hypothesis that retinal processes were maturing in these animals, which required the stimulus of a certain amount of light, probably patterned light. If this is learning, it does not fit any current definition."
So I think what we have here is a quite radical rethinking of both classical and contemporary theories, both of perceptual development in particular, but also in ways of thinking about learning and constraints on learning and what drives learning in some directions relative to others. So that was the first insight that came, I think, from this work. The second insight pertains to the fundamental role of the ground surface in perception, in action and beyond, and also the role of optic flow in specifying where the ground surface is and where one is oneself as one moves through the environment.
With the visual cliff I think that Gibson and Walk showed that the ground surface is already fundamental to animals the first time they have any visual experience at all, right? To dark-reared animals. And it's powerfully affecting the behavioral decisions they're making of where to move and where not to move.
And of course, this is something that's currently actively being studied at MIT and other places. The role of the ground surface in scene perception, the way when we suddenly see scenes, we'll automatically pick out ways that one could potentially move through that scene and where there are barriers to motion through the scene, where the ground surface-- there's a ground surface available to move on and where there isn't. I think this is really fundamental.
Some of the first insights about the ground surface came from James Gibson in his work with airplane pilots and landing. But I think a good chunk of it comes from this developmental and comparative work on the visual cliff.
And finally, I think I said there were, I think, three main contributions. The third is a new approach to the study of perception. Now, Gibson was extremely suspicious of the classical psychophysics of one of my heroes, and I think many people's hero is Hermann von Helmholtz. Now, Helmholtz argued, I think really persuasively-- well, let me give you Gibson's view first.
"Although the discussions in Koffka's seminar were on a very high level, I was not attracted by Gestalt psychology, and I yearned for what I thought of as hard psychology. I didn't like introspective methods. I wanted to be objective, as I thought about it then, and I wanted to work with animals and children." And here was the problem with Helmholtz.
Helmholtz argued that we can have a science of perception for human adults, and the reason that we can is that we can do systematic, psychophysical experiments on ourselves once we train ourselves to be able to do them systematically. And a psychophysical experiment, for example, in color vision will involve something like setting up a device that will deliver amounts of light at particular wavelengths in small, faint patches, indicating any trial whether you see the patch or not. And if you do this on yourself or you train your graduate student to do it and the judgments that you give agree with each other, now you have a generalizable science.
But Helmholtz also argued that some of the most interesting questions about space perception is, where does our capacity for space perception come from? What is it based on?
He speculated that it's based on a lifetime of experience, perceiving things at different distances and very slowly learning about the relationships between different patterns of optical stimulation and different consequences of moving through the environment and encountering surfaces in other modalities. But he also goes on to say, but we'll never know whether that's really what's going on over development because we can't study animals or young infants, because we can't train them to introspect and do the experiments that we do, OK?
But what Gibson showed-- now, before I get to what Gibson showed, this argument from Helmholtz got translated to Titchener at Cornell and from Titchener to Boring at Harvard, and it had serious consequences. At Cornell, developmental psychology was taught in a school called the School of Human Ecology, formally called the School of Home Economics. It wasn't taught in any discipline connected to the sciences because children are not fit subjects for psychophysical experiments, and psychophysics is the paradigm of good science experimental methods, and it relies on introspection.
For Boring, developmental psychology, at least it was in the Faculty of Arts and Sciences, but it wasn't in the Department of Psychology. When I was an undergraduate at Harvard, I had to take developmental psychology in a department called Social Relations, whatever that means, OK?
But the visual cliff studies opened a door to developmental cognitive science. Psychophysics doesn't have to be about relating isolated points of light to isolated elements of your sensory experience. How much redness is there in this flash, or things like that, right? How detectable is this flash?
It can be a matter of relating optical transformations to adaptive action. That's what the studies with the bicycle on the plane were all about. It's what the visual cliff is about. You can do good psychophysics on animals and infants just as much as you can do them on human adults.
And indeed, people came along and did that. One of my favorite experiments is held in Hein's classic study of rats on the-- sorry, cats on the kitten carousel, adding some more insights into why the kittens in the Gibson and Walk experiments might have needed seven days to start using vision to guide their locomotion on the visual cliff.
What Held and Hein did was ask to study the effects of active, self-produced locomotion versus passive motion, passive visual motion on visual cliff avoidance by taking cats at birth, rearing them in the dark entirely, except when they were in this apparatus, which was a cylindrical shaped chamber, vertical striped, black and white stripes on the walls, and a little gondola in the middle on which one cat walked actively. The other cat was transported passively.
But neither cat was ever able to see their feet or anything else hitting the ground, right? Because both cats had this collar on that blocked their view of their own body. So they're just moving around in space.
And what they found was that the guy who's moving actively, even though he can't see where he's stepping on the ground, the very first time he's put on the center board of the visual cliff, he avoids the deep side and goes over to the shallow side. The passively moving animal doesn't, so there's an interesting role for self-produced locomotion in these visual motor systems. And I think I just told you all of that.
The discoveries also inspired studies of human infants' use of optic flow, and I'll get to that in a minute, because Eleanor Gibson did those studies. But Cornell still wasn't ready to offer her a job, so she turned to a new research area that I'm not going to talk about except I think it's very much in the spirit of some of the work you've been doing, Josh, on character recognition, and also quite in the spirit of a lot of work that's going on-- Ted Gibson and other people on interactive processing of speech and also people working on language-- effects of language processing on skilled reading.
She won-- she wrote a book on perceptual learning and development that won a bunch of awards. She won a bunch of other awards, including finally the US National Medal of Science. And she finally became a full-time named Professor of Psychology in 1972, as I told you in the chronology.
Now, her own experiments focused on studies of perception. I actually discovered-- I was surprised to see that in one of her more recent autobiographies, she said, I was the first person to do experiments in her lab. But they certainly looked to me like they were designed in large part by her.
One of them was on-- I was interested in social perception, so she must have convinced me to ask, do babies even recognize that a person's the same person when they see them in different contexts, doing different things? I was also interested in intermodal perception. I probably learned that by being a teaching assistant in her course.
But let me focus on some of her studies. Once she had her own lab, what she wanted to study was perception in human infants. And the first study she did looked at infants' perception of the rigidity or deformability of surfaces from visual information, from patterns of visual motion.
So in one study, she familiarized infants with a surface that only moved rigidly, but in three different directions, in and out, tilting this way, tilting a different way, and then tested them either with a new rigid motion-- alternately with a new rigid motion and a deforming motion with the surface kind of squished together. And infants dishabituated looked longer at the deforming motion.
Another study she did was on one-month-old infants, and it asked whether babies show intermodal transfer of a shape discrimination from touch to vision. So she familiarized babies with a pacifier in their mouths, which was either-- which was smooth and round in both cases, but either rigid or deformable, and then test tested them again with the visual rigid motions versus the deformable motions.
And there, she found that babies tended to look more at the object with the novel substance. So if they'd been familiarized to a rigid motion, they were more interested-- in their mouths, they were more interested in looking at the deforming motion, and vice versa.
And my favorite study that she did from those years was a study that used-- that put babies in an infant seat with a pressure transducer behind whatever was supporting their heads. These were, I think-- did I write it down? I think they were quite young. Three-month-old infants.
So they're leaning back, and they see a surface approaching them. And in one condition, it's a large, extended surface with a square hole in the middle. And in the other condition, it's that square that was formed from the hole from the large extended surface. So you've either got-- you've got a square contour approaching the baby in both cases. In one case, it's a surface that looks like it's about to hit them. And in the other case, it's an opening through which they can see more and more of the scene.
And when the surface is being presented, there's more pressure on the head going back, as if it's either trying to avoid a collision or else keep more of the surface in view. And when it's a scene that's opening up, the head starts to move forward as if to look through at the scene that's being revealed behind the aperture. And I just told you everything that's on that slide.
So at around the same time, Davida Teller and Richard Held were using methods like Eleanor Gibson's to do other psychophysical experiments on infants. My hands-down favorite one is a study that was done by Held here at MIT, published in 1980, where he did psychophysics on individual infants using the simplest possible measure of looking time.
Babies are wearing stereo goggles. They're seeing through the goggles. They're seeing two arrays, one of them black and white stripes that look flat, the other black and white stripes that look like they're varying in depth. And he has observers not recording anything about babies' looking time. They're simply on a whole series of different trials where the binocular disparities are changing, the side on which-- the distance variation is changing, and so forth. Across a whole series of trials, all they're doing is simply guessing on each trial.
Which of these displays do you think the baby was looking at more? OK. that's the only measure. And with that, he can plot these gorgeous psychophysical functions on individual infants, showing that stereopsis, a cortical function-- for those who have been interested in recent tics papers arguing that cortical functions aren't present in young infants, a cortical function, stereopsis, is coming in between about 10 and about 20 weeks in each-- most of these different infants. Not all of them, but then not everybody has functioning stereopsis as adults. There's one subject that didn't here. OK.
Oh, and by the way, the studies also have beautiful controls to ensure that it's stereo and not double images that are determining infants' looking preferences through really simple manipulations like taking off the stereo glasses where you still have the double image-- where now for the first time you do see double images, or turning everything so the stripes are horizontal and the disparities are vertical, which we don't see as stereo since we don't have one eye above the other, but two eyes that are side by side. OK. So really beautiful studies. Sorry, Helmholtz. It's a really good psychophysical experiment, and it's on infants, OK?
So anyway, that's it for the studies. Gibson continued to run a productive lab after she retired. She started moving around more, and at Emory and later Middlebury collaborated with people like Karen Adolph on new experiments, variations on the visual cliff.
At age 78, she wrote a really beautiful paper on exploration that anybody who thinks that these exploratory behaviors make for bad science should read. At 81, she collected all her papers as I've referred to a number of times in a great book from MIT Press.
She gave a big talk in 1983 and wrote her last memoirs in the last year of her life at age 91. And let me quote from it. "My purpose in the following pages is to tell the story of a couple of scientists married to one another and working in much the same field. I aim to show that it is possible to raise a family and do one's job pretty well in our case without sacrificing one's independence.
Yes, there were what some would consider sacrifices when I gave up my safe teaching job at Smith for a completely uncertain research future. But with plenty of love, family support, and imagination, it can turn out to be far more interesting than sticking in one safe spot. My husband and I both loved intellectual adventure. It led to frequent travel, new friends, well-educated children, and most important, some insights in science."
So on doing psychology-- science in hard times, Gibson faced setbacks at every turn. She struggled during the Great Depression. She couldn't live on campus as a woman at Yale or use the library. As a faculty member at Smith, she was in a great environment, but everything got derailed by World War II. As the wife of a faculty member at Cornell, she couldn't be hired for decades. And then once she was, she was forced to retire.
But you read her autobiographical material and you see that her life was a tremendous success and a really happy one. She loved her college years. She married the love of her life and lived-- they lived together for as long as he lived, another 47 years.
She raised two children, mixing research and parenting, she claimed-- I would claim as well-- to the benefit of both. She traveled widely, meeting colleagues who became close friends, and she never got tired of the work that she was doing, which she continued to the end of her life.
So conclusions, none. The study is underpowered to be drawing any conclusions. But questions for discussion? If you're in the mood, yes, one. Can a person lead a steady, happy life in science when the future is uncertain? Like, I don't know. Maybe today, right? Pandemics, climate chaos, financial chaos. You know, there's no shortage of reasons to kind of look around and think, this is not the best time that we're living through. Potential wars-- real wars as well as potential ones.
Can you lead a good life in science when good jobs are hard to get, when the world is anything but equal? I think here's an example that says, yes, you can. Not that it's good to have all these bad things happen. Obviously, it would have been way better if they hadn't. But they did not undermine her science or her morale as far as I can see at any point along the way. Sometimes she used them as fodder for, isn't the world bizarre, what's happening? And we'd laugh about it with her, preferably over a drink.
Second question. How can an institution become a worldwide center of excellence in science? Now, the answer might be pretty duh. Of course, it can if they've got a big endowment, long history, next to a big city, and so forth. But what if it's a small, isolated women's college at a point where nobody thought women were worth-- hardly anybody thought women were worth educating?
I think Smith's department shows it's possible. And it would be really interesting to look at all those ways in which they supported Eleanor Gibson and so many other women and think about, can we make our science better by doing the same kinds of things? Very actively seeking out the way they out all these refugee Gestalt psychologists from Germany and other countries in Central Europe. Can we make our science better by following this example?
And then third, at a time shaken by worries over the reproducibility of research, what are the best ways for research and CBMM to be not only reproducible, which I think it's far too low a bar, but to be robust and generalizable and meaningful and insightful, right? What are the best ways to do it?
Gibson thought it was to focus on a single problem and go at it from multiple different angles, each experiment independent of the others, but all of them converging on similar conclusions. I see a lot of the work that goes on in CBMM that way as well. Thanks.
[APPLAUSE]