The science of empathy, forgiveness, and prosocial behavior

An interview with psychologist Michael McCullough.

Tell us about your background. What sparked your interest in this field?

I was trained in counseling psychology, and I became interested in forgiveness during graduate school because my advisor and I thought it offered promise in the therapeutic realm: Many people’s problems involve trying to recover from trauma and negative interpersonal experiences. As my research progressed during my first academic job, I became increasingly interested in basic research questions about forgiveness and empathy. Gradually, I found myself spending more and more time running subjects and less and less time honing my clinical skills. 

Recently, my colleagues and I have been doing a lot of psychometric research to understand what forgiveness actually is, which is such an obvious research question that we were surprised by how few people had thought about how to evaluate the question scientifically. As simple as it sounds, our work suggests that we can think of forgiveness as a psychological movement along an attitudinal continuum that ranges from hostility to friendliness. To forgive someone, it appears, is to experience reductions in hostility and a return to a friendlier, more benevolent stance. Many of the research tools that psychologists use to measure forgiveness manage to capture this hostility-friendliness continuum, though they do it in slightly different ways.

… our work suggests that we can think of forgiveness as a psychological movement along an attitudinal continuum that ranges from hostility to friendliness. To forgive someone, it appears, is to experience reductions in hostility and a return to a friendlier, more benevolent stance.

So, much of my time is spent in the laboratory, with my graduate students, working to answer research questions about how empathy and forgiveness and prosocial behavior work. But I also have a “night job” as an author, which requires broader reading and sustained effort to understand what other disciplines have to offer.

In your new book, “The Kindness of Strangers: How a Selfish Ape Invented a New Moral Code,” you explore how humans have evolved over thousands of years to be more interdependent and interested in the well-being of others. Tell us more about what research says on this issue. Why do we view relationships as important? How does this factor into Darwin’s natural selection theory?

I can’t think of any primate that is as dependent on cooperation as humans are—even on the very basic level of trying to get enough calories to meet its basic energetic needs. Humans evolved to be completely dependent on mothers and others for the first three years of life. And we continue to cooperate with others through the rest of life to accomplish countless other tasks.

One important cooperative task for our ancestors was large-game hunting. Large-game hunting, as physical anthropologists have documented, is mostly an exercise in despair and disappointment. Most days of hunting are failures, with nothing to show for them. However, our ancestors were able to ensure themselves against those failures by sharing with each other—especially when they were successful in bringing down big animals that could be shared with multiple households. Ancestral humans also cooperated in processing the other foods, aside from meat, that they gathered from their environments, but communally sharing the spoils of hunting is a big deal because of meat’s nutritional density. 

It’s not so much that ancestral humans learned the importance of cooperation; it’s more accurate to say that natural selection rewarded those early humans who were inclined to cooperate in the only currency that natural selection actually cares about: inclusive fitness. As a result, our ancestors ended up with minds that prefer cooperation over self-reliance. To be successful in the social world that humans evolved to live in, cooperation is essential. We also cooperate to make war, of course, so it would be a mistake to assume that cooperation is some sort of unalloyed moral good. We evolved to get all sorts of jobs done through cooperation.

In addition to the laboratory work that psychologists do, and the fieldwork that anthropologists do, some social scientists whose strengths are in math also devote a great deal of their professional time to formulating mathematical models that can explain how our cooperative instincts evolved. It turns out that there are many evolutionary scenarios that could have favored the evolution of a cooperative temperament in humans. Because of how natural selection designed us to operate, human beings simply cannot go it alone. No person is an island.

“To be successful in the social world that humans evolved to live in, cooperation is essential.“

“Because of how natural selection designed us to operate, human beings simply cannot go it alone. No person is an island.“

What are some of the factors that have helped humans become more empathetic, communal, and forgiving?

In addition to our natural, evolved endowments for cooperation, the last few millennia have brought huge developments in science, technology, and trade, which we have brought to bear in efforts to try to help complete strangers—including strangers on the other side of the world whom we will never meet. We have also discovered in our increasingly interconnected world that international cooperation is more profitable than war, and that the best way to ensure world peace is to bring people out of poverty. There are other historical touchstones that help to illuminate the trajectory of concern for strangers, such as the development of the Golden Rule that came out of the Axial Age; the philosophical and scientific writings of people like Jean-Jacques Rousseau, Immanuel Kant, and Adam Smith, who gave us the ideas of natural rights, individual dignity, and distributive justice; and the Geneva Conventions in the 20th century, which provided basic humanitarian norms for how nations should go to war. We got on the path to interdependence as hunter-gatherers, but for the past several thousand years we have been using our intellect and inventiveness to significantly stretch our capacity to care about the welfare of strangers.

Research suggests that we have evolved in ways that help us engage productively with others. But it also shows that there are many limits on how and when we will help strangers. For instance, your lab work has shown that people are not always willing to intervene when they witness someone who is being mistreated by someone else. Tell us more about these findings.

That’s right. It’s pretty clear that we have evolved to help people who are our relatives and who we can expect help from in the future, but we probably aren’t naturally eager (as a result of evolution, anyway) to help complete strangers—especially when helping might involve punishing a stranger who has harmed another stranger. We have performed a variety of experiments over the past few years to investigate this question. And it’s an important question because some evolution-minded researchers over the past couple of decades, including the economists Ernst Fehr, Simon Gächter, Herb Gintis, and Sam Bowles, have pushed for the idea that we are naturally prosocial toward strangers, and that one way we demonstrate this evolved concern for strangers is through so-called “altruistic punishment”: by  retaliating against bad guys who harm others, even when we don’t know the victims and will have no opportunity to interact with them in the future. 

But our research does not support the idea of altruistic punishment. In one of our experiments, for example, we set up a situation where people could administer a sound blast to retaliate against a stranger who had insulted either them or another stranger. We didn’t label the sound-blast tool as a “weapon” or anything like that, and we didn’t talk about the experiment as though we were studying aggression or retaliation. Instead, we presented the sound-blast device as a benign tool that subjects could use to deliver sound to another subject in the context of an acoustics experiment. Subjects had to discover that they could use it to retaliate (if that’s what they felt like doing). But before the acoustics experiment, all three subjects (who were anonymous to each other), wrote essays on personally important topics, which they shared with each other and later critiqued. One of the people in this three-person group harshly critiqued (ridiculed, really) either the subject’s essay or the other stranger’s essay. People were happy to retaliate against the offender when it was their essay that was ridiculed, but they didn’t retaliate when the offender critiqued the other stranger’s essay. 

We also found that people’s inaction as bystanders wasn’t because they didn’t notice the third-party harm, or because they didn’t view it as unfair. They noticed the insult just fine, and they viewed it as unfair, just as they viewed the insult they themselves had suffered as unfairl. They simply didn’t seem to be motivated to go to the trouble of intervening on the victim’s behalf. So this experiment and some of our others suggest that humans aren’t altruistic punishers after all.

Regarding forgiveness, research suggests that individuals are doing a constant calculus on how to move forward with someone who they perceive has wronged them. Could you talk more about that?

We have conducted several studies to examine the sorts of social information people consider when they’re deciding whether to forgive someone who has harmed them. We have hypothesized that people are inclined to forgive people who don’t seem likely to harm them again in the future, along with people who seem like they could be valuable relationship partners going forward. The results from our studies show that perceived relationship value is very important, but we haven’t found much convincing support for the idea that people care about the risk of future exploitation. This is a puzzling finding—or a lack of a finding, really—because it’s hard to see how we could evolve a decision-making system for deciding whether to re-enter relationships with people who have mistreated us if we don’t weigh the likelihood that they’ll be in a position to harm us again in the future. I hope we will be able to examine this issue further in future work.

The U.S. has just finished a tumultuous presidential election. Americans are increasingly at odds based on their life experiences, location, religious beliefs, race, and other factors. Much of your research has touched on empathy among people who know each other. But the political divide in the U.S. is between people who in many cases live hundreds or even thousands of miles apart. Any insight on what may be driving this divide or on how to bring these people together?

Americans have experienced significant declines over the past several decades in our trust for government, for scientists, for the media, and even for each other. It is hard to know what sort of advice to offer for reversing these trends. On a person-to-person level, I see no productive way forward that doesn’t involve meaningful conversations about facts and values, and by viewing people who disagree with you as rational creatures who are sincerely presenting their beliefs. It is also important to gently ask questions to fully understand the other person’s point of view. But productive one-on-one interactions are just small steps, and probably weak medicine for what really ails us as a country. The social scientist who figures out how to solve Americans’ trust problem probably ought to get the Presidential Medal of Freedom.

Further reading

McCullough, M. E. (2020). The Kindness of Strangers: How a Selfish Ape Invented a New Moral Code. New York: Basic Books.

McCullough, M. E., Root, L. M., Tabak, B. A., and Witvliet, C. v. O. (2009). Forgiveness. In S. J. Lopez and C. R. Snyder (Eds.), Oxford library of psychology. Oxford handbook of positive psychology (p. 427–435). Oxford University Press.

Forster, D. E., Billingsley, J., Russell, V. M., McCauley, T. G., Smith, A., Burnette, J. L., Ohtsubo, Y., Schug, J., Lieberman, D., and McCullough, M. E. (2020). Forgiveness takes place on an attitudinal continuum from hostility to friendliness: Toward a closer union of forgiveness theory and measurement. Journal of Personality and Social Psychology, 119(4), 861–880. https://doi.org/10.1037/pspi0000227

Pedersen, E. J., Kurzban, R., and McCullough, M. E. (2013). Do humans really punish altruistically? A closer look. Proceedings of the Royal Society B: Biological Sciences, 280(1758), 20122723.

Pedersen, E. J., McAuliffe, W. H. B., and McCullough, M. E. (2018). The unresponsive avenger: More evidence that disinterested third parties do not punish altruistically. Journal of Experimental Psychology: General, 147(4), 514–544. https://doi.org/10.1037/xge0000410

McCullough, M. E., Luna, L. R., Berry, J. W., Tabak, B. A., and Bono, G. (2010). On the form and function of forgiving: Modeling the time-forgiveness relationship and testing the valuable relationships hypothesis. Emotion, 10(3), 358–376. https://doi.org/10.1037/a0019349

Pedersen, E. J., Forster, D. E., and McCullough, M. E. (2014). Life history, code of honor, and emotional responses to inequality in an economic game. Emotion, 14(5), 920–929. https://doi.org/10.1037/a0036752