Navigation auf uzh.ch

Suche

UZH News

Behavioral Economics

The Costs of Rationality

Paul Glimcher, co-founder of the field of neuroeconomics and professor at New York University, recently held a keynote speech at UZH. In his lecture titled "Efficiently Irrational", he showed why seemingly irrational decisions are economical and how too much choice leads to worse decisions.
Victoria Watts
The cost of rationality: thinking needs energy, about 400 kcal per day.

Neuroeconomist Paul Glimcher makes quick decisions. Asked at which market stall in Zurich's Christmas village he would like to eat, he does not hesitate. "Here," he says, pointing to one. Why this one? "I don't know, but this is the one," says the New York University professor.

These days, with every decision having to be optimized and the best offer to be chosen from an almost infinite number of possibilities, his approach seems somewhat outmoded. But it bears witness to Glimcher's many years of study on the subject of decision-making behavior. His research is guided by one question: Why do people make crazy decisions?

The co-founder of the field of neuroeconomics provided an answer to this and other questions during a lecture at UZH.

How do we make decisions?

In the middle of the last century, behavioral researchers worked on the assumption that in our heads, we have various “little people” with different preferences wrestling with each other. While one little figure wants to lead a healthy lifestyle another wants to stay inside, where it’s cozy and warm. These inner wranglings were supposed to explain inconsistent and irrational behavior. While the model had descriptive value, it lacked any predictive power.

The 1980s saw the seminal work of Daniel Kahnemann and Amos Tversky, who depicted our decision-making using individual utility functions which allowed them to better predict behavior. "Amos Tversky was one of the best psychologists of his time," Glimcher pointed out. "His utility theory has had a major impact on our understanding of human behavior. But, good decision-making is not only about utility, it's also about cost."

Complete rationality is very expensive

Since the turn of the millennium, neuroeconomists have been studying the biological and neural foundations of human behavior. "We now know that while our behavior may look a little crazy and irrational, it is not because we are crazy or irrational, but because the cost of perfect rationality would be enormous," Glimcher points out. "Our mental resources are limited, and our brain is not a supercomputer, but a small, efficient computer with a power of about 10 watts." If we wanted to bring this power up to 100 watts, which corresponds to an average laptop, we would need a head as big as a rescue ring and we would have to eat about 7000 kcal a day to keep this brain running. "That's a high price to pay for a little bit more precision or rationality in our decisions," he argued. "Such a brain would simply not be efficient - neither for animals nor for humans. We would starve before we could decide what to eat next."

Small choice-sets, better decisions

What does this mean for our everyday decisions? Paul Glimcher recommends reducing choice-sets whenever possible. Because more choice always means more information, and this is associated with increasing costs. Moreover, more information rarely leads to better decisions. According to Glimcher, the opposite is true: a series of studies in his laboratory have shown that we make worse decisions when we have more choices available. Going back to the Christmas market: If you realize that your colleague chose the yummier food stall, that's no big deal. But when it comes to public health measures, such as the choice of a covid vaccine, the consequences of too much choice can be devastating for many people as choosing the best option becomes a matter of luck. Therefore, policymakers should take into account the limited cognitive resources and not provide ten alternative vaccines, but only two or three.


Paul Glimcher gave the keynote lecture at the 5th Annual Symposium of the Marlene Porsche Graduate School of Neuroeconomics at the University of Zurich.

Weiterführende Informationen