Navigation auf


Department of Economics

The neural foundations of human behavior

Neuroeconomic methods can be used to explore the biological basis of human behavior in unprecedented ways and allows us to better understand the motives and processes that define our behavior.

Ernst Fehr is a dedicated advocate of the importance of behavioral and neuroeconomics for furthering our understanding of economics. In this interview, he explains how neuroeconomics provides answers to fundamental questions the field faces.



You are a pioneer in behavioral economics and co-founded the Zurich Center for Neuroeconomics. Why did you turn to neuroscience for answers?
It has always been clear that human behavior has a lot to do with processes in the brain and that economists must include these in their models. However, for a long time there was no technology to tap into this field.

In the 1980s and 90s, new methods for non-invasive measurement of brain activity emerged, such as functional magnetic resonance imaging (fMRI). Test subjects are put in a scanner and solve tasks or make decisions. Thanks to imaging techniques, we can visualize the active neuronal networks while they are doing these tasks.

During the same period non-invasive methods to stimulate brain activity – for example by electromagnetic impulses – arose. By combining such transcranial magnetic stimulation (TMS) with fMRI, we can identify active networks in the brain, and stimulate or inhibit them. This allows us to identify a causal influence of specific neural activations on decision-making behavior. Such methods for non-invasive, causal manipulation of brain activity are central to understanding the biological and neural basis of human behavior.

Methods for causally influencing brain activity are central to understanding human behavior.

What happens during brain stimulation?
During TMS a coil that transmits magnetic pulses through the skull is brought up close to the areas of the brain that you want to stimulate.

In transcranial direct current stimulation (tDCS), a cathode and an anode are attached to the head, between which a very light current flows. The current flow is so weak that the subjects do not notice whether they are being stimulated or not. Nevertheless, the stimulation has an effect on the subjects' behavior: Professors Michel Maréchal and Christian Ruff have shown that brain stimulation can significantly change subjects' honesty. In addition, a team around Christian Ruff and myself has shown that brain stimulation can significantly influence compliance with fairness norms. 

I assumed that honest behavior is based on values and moral convictions...
A large part of the brain activity runs without us being aware of it, just like other biological processes. Yes, you have the same moral convictions as before, you have not become a different person. Nevertheless, you will behave differently after the stimulation. This means that brain activity is causally co-responsible for behavior. It doesn’t matter whether you are aware of this activation or not.  

How do findings from neuroscience contribute to economics?
Neuroeconomics allows us to measure processes that have an influence on decision-making but could not previously be measured by economists. Especially in basic research, we can close some knowledge gaps with the improved understanding of the neuronal and biological basis of human behavior.

Neuroeconomics allows us to measure processes that have an influence on decision-making but could not previously be
by Economists.

Can you give an example of such a knowledge gap?
A central element of traditional economics is the idea of utility. But this concept is only an auxiliary construct, because we cannot observe utility directly. This is why economists say that "people behave as if they are maximizing their utility." This "as if" expresses that we do not know if there is such a thing as utility. When we see a person choose an apple rather than a pear from a mixed bowl of fruit, we cannot conclude that apples have a higher utility than pears for this person. I cannot prove this because I cannot measure the utility of the apple without refering to the shown behavior or choice of this person chosing the apple.

How can neuroscience measure the utility of the apple?
Today, we have quite a good idea about the areas of our brain in which subjective economic value is encoded or represented. This includes the so-called ventromedial prefrontal cortex (vmPFC) in the front part of the brain. In other words, the intensity of neuronal activity in this brain area is a measurable substrate of subjective value, i.e., a neurophysiological representation of utility.

Of course, an apple has different value to different individuals. But for all of them, this value is mapped in the vmPFC. The greater the subjective utility of the apple for a person, the more activity we will see in this area.

This is an important finding. We can measure the activity in the vmPFC and, based on this measurement, predict how people will tend to behave in the future.

I choose the apple and you see a lot of activity in that brain area because I like apples and they give me great value or utility. But I also like pears, and I could just as likely have chosen the pear.
This brings us to the next aspect. Behavioral decisions are influenced by several components. Besides the general utility component, there is a component of randomness. Neuronal activity has a random component. Thus, if subjective value is encoded in the brain by specific neural activity, then decision-making behavior must also have a random component. Because we do not exactly know the structure of this component, we cannot draw clear conclusions about the utility component based solely on observed behavior. This is a fundamental problem. We cannot identify the preferences (utilities) of individuals based on their decision-making behavior alone.

We need additional information to identify individual preferences. Therefore, I look not only at behavior, but also at response times or brain activity during decision-making. This data is called non-choice data and can help to solve the problem of preference identification.

The closer together the utility
of two options i must choose between, the greater
the response time.

What kind of information does non-choice data contain?
Let's take response times as an example. If we take them into account, we can solve the fundamental problem of preference identification. The closer together the utility of two options lies, the longer it will take me to make my decision. If I like apples almost as much as pears, I will think longer before choosing. If I strongly prefer apples, my response will be much faster. The speed of my decision contains information about utility differences; it correlates negatively with the difference in value of two goods.

In a recent study, Carlos Alós-Ferrer and Nick Netzer show that based to this insight, we can solve the problem of preference identification. It is quite possible that other physiological measures – pupil dilation, blood pressure, brain activity – also potentially conceal systematic information about preferences. Insights from neuroeconomics can significantly improve economic models based on observed behavior alone. They are a great asset in the toolbox of economics.

The research of the SNS-Lab often deals with topics such as altruism, risk-taking, morality, or self-control. Are these the main drivers of human behavior?
Neuroeconomics looks at the processes in the brain that are related to the subjective valuation of choices or goods. The motives you mention play a role in the valuation of goods, because most goods have a time, risk or social dimension.

The mathematical formulation of the neuroeconomic findings is a central task. only then can they be integrated into the models of standard economics.

How do insights from neuroeconomics find their way into standard economic theory?
The neuronal and cognitive processes that underlie our preferences and evaluations are fundamental for understanding human behavior. A better understanding of these processes allows a better understanding of decision-making. Neuroeconomics can help solve fundamental problems in economics. However, it is necessary to translate these findings into the formal mathematical language of economics, as this allows us to integrate neuroeconomic insights into the standard economic models.