Minding bias: A Yale economist examines learning and decision making

Mira Frick, a Yale economist who develops models of how people make decisions and learn, recently earned a Sloan Research Fellowship, which recognizes early-career scholars across several scientific and technical disciplines for distinguished performance and potential to become leaders in their respective fields.

In her recent work, Frick has focused on learning biases, decisions made with ambiguous knowledge, and random choice. An assistant professor of economics in the Faculty of Arts and Sciences, she came to Yale in 2017.

In an interview, she discussed her innovative work, her path to economics, and her experiences at Yale. The interview has been edited and condensed.

What is the focus of your research?

Mira Frick: I work on theoretical models of how people make decisions and learn, both individually and in interactive settings. A lot of my recent research has been in the burgeoning field of “learning with misspecified models.” This area, and my work, is motivated by growing evidence from psychology and behavioral economics that, in many economic settings, people’s learning is governed by an incorrect, biased, or simplified view of their environment. In a series of papers with my Yale colleague Ryota Iijima and our coauthor Yuhta Ishii at Penn State, we explore how incorporating such biases alters the predictions of traditional economic models of individual and social learning.

What is an example from your recent work?

Frick: To give a simple example: Suppose a consumer is interested in buying some new product, but is uncertain about its quality or long-term safety. One way she might try to learn about this is by observing other consumers’ purchasing decisions. But a potential complication in drawing inferences from other consumers’ decisions is that she may not be entirely correct about others’ tastes, risk attitudes, or similar idiosyncratic characteristics. Traditional economic models of learning tend to gloss over this complication: they assume that consumers make correct assumptions about each other’s characteristics.

However, in a recent paper with Ryota and Yuhta, we show that even vanishingly small misperceptions of other people’s characteristics can have stark consequences and lead to dramatic failures of learning. For example, if consumers even slightly underestimate other people’s risk tolerance, we find that they will eventually grow confident in the product’s highest possible safety level — no matter how safe or unsafe the product actually is. Such highly confident but incorrect beliefs can then obviously lead to very inefficient behavior, such as the widespread adoption of unsafe products.

What are the implications of these types of findings?

Frick: That people’s learning is governed by some amount of misperception seems unavoidable: The world is highly complex, and even the most sophisticated person will not be able to correctly perceive all aspects of her environment.

At the same time, as economists, we tend to think of our models as simple approximations of reality, so that as long as people’s misperceptions are not too severe, we might feel that it is safe to ignore them. One of the implications of my work with Ryota and Yuhta is that this need not be the case: If people are even slightly wrong about their environment, this can lead to failures of learning that are drastic departures from what traditional economic models would predict. This suggests that we should take people’s learning biases seriously, and it motivates more research, both empirical and theoretical, into which biases are relevant in different economic settings and how they change the predictions of traditional models

Switching gears, how has your experience been as a young professor at Yale?

Frick: The economics department at Yale is an exceptionally stimulating and supportive environment, and I feel extremely lucky to be able to start my career here. The department and the Cowles Foundation have a long history of excellence in economic theory. Our theory group is one of the largest in the country, with an unusually well-balanced composition in terms of research interests and career stages. I think we even have an almost perfect gender balance — which is very rare in economic theory, where women are still quite under-represented. Our group organizes four or five different research events every week, and two large conferences per year. This creates an incredibly vibrant research climate, and brings our faculty and Ph.D. students together in a close-knit community that I really enjoy being a part of.

A longer version of this interview can be viewed on the Department of Economics website.