In early October, neuroeconomics scholars descended on Pasadena to attend the Society for Neuroeconomics' annual conference and learn about the latest research in the field. The location was intentional: Caltech faculty have been at the forefront of the development and growth of neuroeconomics since its formal inception in the early 2000s. Two pre-event satellite symposiums focused on consumer neuroscience and cognitive economics were co-sponsored by the T&C Chen Center for Social and Decision Neuroscience, based at the Tianqiao and Chrissy Chen Institute for Neuroscience at Caltech, which brings together experts in neuroscience, computer science, economic theory, political science, and more to study how human brains make decisions.
So, what exactly is neuroeconomics? The Society for Neuroeconomics website describes as a field that "represents the confluence of economics, psychology, and neuroscience in the study of human decision making." But for Antonio Rangel (BS '93), Caltech's Bing Professor of Neuroscience, Behavioral Biology, and Economics and past president of the Society for Neuroeconomics, the aim of the field is much more profound.
"Neuroeconomics is about understanding human nature: Who is it that we are? Why do we do what we do?" he says. "I think it's a dream that we can construct this understanding of human nature that links behavior, computation, and biology at the most fundamental level. It doesn't get more beautiful than that. It's hard, but it's as deep a question about who we are as it gets for me."
We recently spoke with Rangel and Colin Camerer—Caltech's Robert Kirby Professor of Behavioral Economics and Leadership Chair and director of the T&C Chen Center for Social and Decision Neuroscience, who helped launch the field with a fundamental Science paper on ambiguity in 2005. Camerer is also a past president of the Society—to learn more about the history of neuroeconomics at Caltech and where the field is headed.
The idea of combining neuroscience and economics research really started to gain some traction and coalesce into a field when the Society for Neuroeconomics was formally established in 2005. How did the organization bring these disciplines together?
Camerer: Actually, there was an initial conference back in 1997. We knew several economists who were a little curious about neuroscience, and one of the organizers of that conference knew several psychologists and neuroscientists who were willing to come. For the most part, the economists were just silent. Almost none of the language that describes the way neuroscientists think about mechanisms in the brain—cognition, interoception, the role of memory, reinforcement learning— was in the economics vocabulary at that time.
As a side note, three people in the audience in economics went on to win Nobel Prizes. Now correlation is not causation, but I think what that means is that these kinds of accomplished scientists are often unusually interested in a lot of different things.
In 2001 or so, there was a small meeting in Martha's Vineyard—and a big storm that nearly forced us to cancel it. The next year, we organized a meeting in South Carolina, and there was another threatening storm. So, the group decided we needed to have a society, so that if we ever booked a hundred hotel rooms and nobody came, and we were on the hook for the rooms, we would have some protection. From that, the Society for Neuroeconomics was created. Paul Glimcher, a neuroscientist and psychologist from NYU [New York University], was the driving force in the organization. In the spirit of the enterprise, we decided that we would choose presidents every year who alternated between economic or social science flavored and neuroscience flavored. He was the first president, and I was the second.
Rangel: Some of the magic that happened at those meetings—and that has continued—was due to the close interactions of three communities: behavioral economists, psychologists, and neuroscientists. Everybody understood very early on that we needed to learn from each other and get the three parts right. One of the things that I'm happy about in the field 20 years later is that if I open a publication from any good lab, the economics is right, the psychology is right, and the neuroscience is right. And we train our students in all of these three disciplines.
Sometimes the culture converges properly, sometimes it doesn't. This just worked. I think there are some challenges for the field in the present, but that part of it is still right.
What were first topics you worked on in the nascent field?
Camerer: The first serious topic was ambiguity, which is a term for when you can't assign a probability to something; maybe there's a range of probabilities. Typically, people don't like to act in the face of ambiguity, so we thought, well, maybe we can see if there are different brain activities with respect to ambiguity and risk, which is when you can give a solid probability. The Caltech Brain Imaging Center had just opened, and they had functional magnetic resonance imaging (fMRI) machines we could use to observe brain activity.
So, we did a study about ambiguity risk using this imaging and we found that there was distinct ambiguity activity, independent of risk, in parts of the brain called the orbitofrontal cortex and the amygdala. The paper was published in Science, and it was a big step in the field because it provided a neural basis of ambiguity for the first time. Our goal was always to prove that even though many of us didn't come from a neuroscience background, we could, with various helpful co-authors and our gumption, produce high-quality general neuroscience that would be published in the top science journals. We didn't want to ever be accused of being a bunch of economists doing crappy neuroscience.
Rangel: In the early 2000s, I was a young assistant professor of economics at Stanford interested in understanding public policy for addiction. The typical way you think about that at a very high level is to have a model of human behavior that tells you what happens if you change features of the environment, like policy details such as cigarette taxes or the prices of products. In the process of working on that, I started reading a lot of behavioral economics and discovering Colin's work. From there, I started reading a lot of neuroscience research because it became very clear to me that the standard models of economic behavior could not produce answers that made any sense in the context of addiction. They were making fundamentally wrong assumptions about what drives those behaviors.
I spent a year reading and researching and eventually wrote an economics paper on the topic, but I also decided I wanted to switch careers. I was lucky enough to start talking a lot with Colin and spending time at Caltech. At some point, the opportunity to come here was open, and I just decided to take the plunge and completely retrain myself in neuroscience.
Camerer: What Antonio did was unheard of. He was very well trained in a particular area of economics called public finance and doing high-impact work and then said, "To move forward, I need to do something completely different." The advice we give students is get tenure for what you are hired to do, just plug away. Once you get tenure, do anything you want. But some people just break the mold like Antonio did.
Rangel: I tried very hard to do the switch at Stanford and my department was extremely supportive. But even though the psychology department and the neuroscience and the economics groups are outstanding, they don't talk much to each other. I think this transition for me was really made possible by coming to Caltech and being in a place that has no boundaries where wacky things happen once in a while.
What role has Caltech has played in expanding and supporting this field?
Camerer: One major thing is that we had a few large grants over quite a long span. Steve Quartz [professor of philosophy] was a main driver in a Packard Foundation grant, and that provided a lot of money for postdocs and initial brain scanning and so on. And after that, the Moore Foundation made several big, multi-investigator grants. Gordon Moore [PhD' 54, chair emeritus of the Caltech Board of Trustees and co-founder of Intel Corporation] was very generous in many ways, including through long-horizon grants. Moore completely understood that we want to bet on moonshots and complicated things.
Steve Quartz and Antonio also received $2 million from the National Science Foundation to fund PhD training in 2007, and major funding from the National Institutes of Health supported the Conte Center for Neuroscience Research at Caltech, which Ralph Adolphs [Bren Professor of Psychology, Neuroscience, and Biology] led for a decade and where researchers used electrophysiology and functional magnetic resonance imaging to investigate how humans make social decisions. All these long-term grants helped move the field forward. Today, the T&C Chen Center for Social and Decision Neuroscience endowment supports continued adventurous research.
Rangel: My knowledge of behavioral economics was very limited, and my knowledge of how to do fMRI and neuroimaging properly was nonexistent. But I started collaborating with Colin and with John O'Doherty [Fletcher Jones Professor of Decision Neuroscience] and our labs were kind of open to each other and the transfer of expertise was amazing. It works at Caltech with the small size and the division structure and the openness to unusual arrangements. My career benefited enormously, and I'm very grateful for it.
One thing that has also been very satisfactory is that we trained a lot of the faculty in the field. A lot of the later presidents of the Society for Neuroeconomics trained here as grad students and postdocs.
Camerer: If you look at the list of who's attending these conferences, there's a lot of second- and third-generation alums. Caltech is tiny, but we have this tremendous, disproportionate impact. In fact, four of nine current board members of the Society for Neuroeconomics and nearly half their presidents since 2004 are Caltech affiliated. The last four Early Career Awards have all been Caltech affiliated—this year one was my academic "grandchild," a student of my student who himself won in 2015—and Antonio's former student Brendan Eum won the 2026 Best Talk Award for a talk that was truly superb on neuroforecasting, an emerging subfield that uses neural data from a small population to predict the behaviors of larger groups. The Institute's influence on the field has been there from the start and continues today.
How has the field changed over the last two decades?
Rangel: Cognitive neuroscience, like many other domains in science, is very driven by technology. You pursue questions that are feasible with current technology and when a new technological innovation comes, it opens the possibility of new questions. Neuroeconomics would not have been possible 30 years ago without neuroimaging—it was essential. For the first 15 to 17 years of the field, computational fMRI was exploited to the maximum. We have been able to isolate algorithms and computations at the level of resolution of the fMRI technology: hundreds of thousands of neurons. It's amazing, but it's very low resolution. I think the field is waiting for the next evolution of technology that will allow us to resolve those circuits and algorithms in healthy humans in higher detail.
Camerer: I'm teaching an AI social science course this term for the first time. There are a few papers on how you can try to recover the cognitive architecture of an LLM and see if it seems broadly like the cognitive architecture of the human brain at a high level. It's very early days, and I think there will be a healthy debate. It's hard to keep up with the papers, between people who think there's no way the LLMs are exactly like human brains and other people who say that the LLMs were trained on human language, produced by human brains, so maybe there is some natural affiliation
Another thing Caltech has done well for many decades is experimental economics, which means you put people in a market or a strategic game, like the prisoner's dilemma. Nowadays, you can ask if LLMs do things like humans. AI is going to generate artificial subjects that are good enough that we can run pilot tests, and then you treat the humans as the rare resource. I think there's going to be some big wins and then some struggles just like with human science.
What are your hopes for the future of neuroeconomics at Caltech?
Camerer: I would love to be able to raise money for high-level thematic work, like research on habits. That topic cuts across so many interesting areas and would make a fantastic center. The psychologists who study habit in applied psychology don't really have a firm grip on the neuroscientific basis of habit, and it's important for so many things, like eating, social media doomscrolling, and forming good educational study habits. Is there a recipe?
I used to go to Santa Anita racetrack every weekend, and I'm interested in pathological gambling. If you're the devil, and you want more people to be addicted to gambling, 2026 is a great year for you. The combination of legitimization of sports betting and phones could be really harmful, especially according to some early studies, for young men. It's amazing how weak the research is on gambling pathology. It shouldn't be that hard to do, even with portable eye tracking, to see if a gambler's pupil is getting large—indicating arousal—when they win or lose. You could go do this in naturalistic settings. Those kinds of experiments just take a little bit of adventurous and patient grant money.
A third topic is the neural basis of extraordinary mental abilities, like in math or writing. Almost all the federal funds from NIH are to study normal health or abnormal disorders. We should also be trying to understand what makes some people so gifted and skilled. How do their brains develop and work differently? We could scan all the Caltech students!
Rangel: I'm incredibly grateful for the support that we got in the first two decades of neuroeconomics—it would not have happened at another place—but I think neuroeconomics 1.0 has run its course, and it's time for neuroeconomics 2.0. It's hard science that requires deep resources. Let's see what happens.
Colin Camerer (left) and Antonio Rangel (right)
