New study across 20 countries suggests guilt, not shame, motivates generosity

A new global study published in Nature Human Behaviour provides evidence that people tend to act more generously toward others when they are aware of the negative consequences of their actions. This effect appears to be linked to guilt, rather than to concerns about being judged by others. The findings call into question the common assumption that reputational concerns drive people to do good, at least when dealing with anonymous others. Instead, the results point toward internal feelings of responsibility and the avoidance of guilt.

Social cooperation among strangers helps societies function smoothly. Previous research has shown that people from different countries vary in how much they help others in situations involving strangers, especially when there is no direct reward. But what motivates people to act generously under these conditions remains uncertain.

Two emotional mechanisms are often proposed to encourage this kind of behavior: guilt and shame. Guilt is an internal sense of having done something wrong, felt even when no one else is watching. Shame, on the other hand, is more about how one is seen by others. Different cultures emphasize these emotions to different degrees, and some researchers have suggested that societies may develop distinct moral systems that rely more on one or the other.

Most of the earlier work in this area was conducted in a limited number of countries with similar cultural backgrounds. The authors of the new study wanted to explore whether guilt and shame work differently in a wider range of cultural contexts, and whether people everywhere are more likely to behave prosocially when they either feel guilt or anticipate shame.

“We were inspired by influential theories in psychology, anthropology, and cultural evolution suggesting that prosociality varies across societies, particularly when it comes to generosity toward strangers. Yet, the empirical evidence for this variation is mixed. This led us to wonder whether such differences might just reflect that societies use distinct ways to promote prosociality—emphasizing either guilt or shame,” said study author Catherine Molho, an assistant professor at the Toulouse School of Economics and the Institute for Advanced Study in Toulouse.

Free daily newsletter

To investigate, the researchers carried out a large-scale experiment using a type of economic decision-making task known as a dictator game. In these games, participants are given a sum of money and asked to decide how much to share with an anonymous partner. The partner has no control over the outcome.

The researchers created three variations of the game. In the first version, participants were given full information about the consequences of their decision for the other person. This setup was designed to allow guilt to influence behavior, since participants knew whether their actions would harm or help the recipient. In the second version, participants could choose to remain unaware of how their decision would affect the other person. This allowed them to avoid knowing whether they were causing harm, and was meant to test for what is sometimes called “willful ignorance.” In the third version, participants knew that their decision would be made public to other participants under a pseudonym, introducing the possibility of reputational judgment and activating feelings of shame.

The main question was whether people would act more generously when they had information (guilt condition) or when they were being observed (shame condition), and whether these effects varied by culture or individual personality traits.

Across all 20 countries, people consistently acted more generously when they had full information about the consequences of their choices. On average, 60 percent of participants made the prosocial choice when they had full information, compared to 41 percent when they could avoid knowing the consequences. This finding held true in all countries, even though the overall levels of generosity varied.

The effect of being observed, however, was minimal. Whether participants’ choices were public or private made little difference in how generously they behaved. In some countries, being observed slightly increased generosity; in others, it slightly decreased it. On the whole, the difference was so small that it fell within the range the researchers had defined as negligible.

The researchers also looked at whether people who were more prone to feeling guilt or shame were more affected by the experimental conditions. They used a validated psychological scale to measure individuals’ tendencies to experience guilt or shame. People who scored higher on guilt proneness were more likely to behave generously when given full information. They were also more sensitive to the opportunity to avoid knowing the consequences of their actions. This suggests that guilt plays a meaningful role in shaping generosity, particularly when individuals are aware of how their choices impact others.

“We find that guilt plays a similar role in promoting prosocial behavior across societies. When people are given clear information about how their actions might harm others, they tend to make more generous choices. But if people can avoid this information, they are less likely to act prosocially. Importantly, those who are more sensitive to guilt are especially responsive to learning about the negative consequences of their actions.”

At the same time, there was no evidence that people who were more prone to shame were more responsive to the public nature of their decisions. Similarly, countries where shame was more emphasized as a cultural norm did not show stronger effects of public observability. In other words, being watched by strangers did not appear to motivate generosity, even among people or cultures that tend to be more shame-sensitive.

“We expected that making people’s decisions visible to others would boost prosocial choices, given how much reputational concerns usually matter. But to our surprise, observability had only a negligible effect, and this was consistent across countries and independent of how sensitive people were to shame.”

One limitation of the study is that the observability manipulation involved anonymous strangers and pseudonyms. “Our study is that we only examined how publicizing decisions among anonymous strangers influences generosity. In reality, individuals may be more concerned with observability by meaningful audiences—such as colleagues, neighbors, or other people they interact with regularly.”

Although the study found that guilt was a more consistent motivator than shame in this context, the role of shame may be more prominent in different types of decisions. For example, public behavior in communities where reputation is tightly monitored might be influenced by shame in ways that were not captured by this study. Future research could test whether public exposure in more socially meaningful contexts influences generosity differently.