Child of Good

Felix Warneken Is Overturning Assumptions about the Nature of Altruism
Illustration by Harry CampbellIllustration by Harry Campbell
By Susan Seligson

The brief video is among the millions vying for our attention on YouTube and Facebook. An 18-month-old, gleefully and apparently of his own accord, toddles across a room to a bowl dropped by a clumsy adult, scoops the object up, extends a chubby arm to return it to its owner, and toddles off. It’s adorable. It is also a rigorously designed and carefully controlled experiment that supports a new but growing body of research indicating that very young children can be altruistic.

At the forefront of these compelling findings is Felix Warneken RI ’15, a Joy Foundation Fellow at the Radcliffe Institute this year and the John L. Loeb Associate Professor of the Social Sciences at Harvard. In addition to being able to maintain a neutral expression under circumstances that would have an amateur laughing or applauding, Warneken heads Harvard’s Social Cognitive Development Group in the University’s psychology department. A graduate of the University of Leipzig, he has been captivated by the evolutionary origins of human cooperation since his days at the Max Planck Institute for Evolutionary Anthropology, where he designed experiments with children and their wordless evolutionary cousins, chimpanzees. The recipient of a string of awards, most recently the American Psychological Association’s Boyd McCandless Award and a National Science Foundation CAREER Award, Warneken is toppling the stubborn assumption that small children are self-centered in the extreme.

“I was interested in child development and children’s understanding of other people’s thoughts, feelings, and goals—how children go beyond the information given to think about the unobservable,” says Warneken, who toils at ground zero of the nature-nurture dilemma. “There’s this strongly held belief in the scientific literature that the true origin of altruistic inclinations is social norms—that Mother Nature has induced us to be selfish and teachers and parents are always hammering into our heads that we should be empathetic and share," he says.

Felix Warneken. Photo by Stephanie Mitchell, Harvard Staff PhotographerFelix Warneken. Photo by Stephanie Mitchell, Harvard Staff Photographer

“Because of this focus on social norms as the driver of altruism, people often assume that young children, and especially chimpanzees, would never engage in these kinds of helping behaviors,” says Warneken. That was the prevailing wisdom at the Max Planck Institute when, as a young doctoral candidate, he suggested that helping behaviors might be based on innate abilities in very young children. What if he tested the notion in children under two? Would a toddler understand—and act—if someone needed help? Was it worth investigating? Some of the more senior scientists were skeptical that toddlers would be helpful at all. “What did I know?” Warneken remembers. “I was just a PhD student.”

That’s when Warneken dropped the ball. It was an accident: in an unrelated experiment, a ball fell out of his hands. It’s not overstating things to say that what happened next—a toddler picked up the ball and handed it over—sent evolutionary psychology hurtling in a fresh new direction.

When Warneken discusses his work probing a broad range of childhood helping behaviors, from cooperation to altruism to reactions to perceived unfairness toward others, he tells the story of the ball and the ensuing collision of science and serendipity. For him, and for his field, it led to studies that changed the way psychology looks at toddlers as young as a year old and their range of understanding. For Warneken, a native of the university town of Tübingen, Germany, who has been on the Harvard faculty since 2009, the ball incident led to a series of increasingly refined studies, and his initially unexpected findings have since been duplicated in cross-cultural studies (children in India, Peru, and Canada all displayed similar helping behavior) and borne out in comparative studies with chimpanzees at the Max Planck Institute and at research centers across the world.

Warneken is the first to admit that in studies with children a range of variables threaten to taint the results. Whereas laboratory mice, for example, are portable and compliant, human toddlers as research subjects pose a vast and daunting set of challenges. Kids are mercurial, restless, only sporadically attentive, and often desperate to please—qualities that can easily derail a controlled behavioral study. Their actions are affected by the presence of relatives or teachers—and, of course, the demeanor of the researchers themselves. “They do cute things,” says Katherine McAuliffe PhD ’13, a research associate of Warneken’s who is now a postdoc in psychology at Yale. “But in the trials you can’t let on that you’re amused, approving, or disapproving. My strategy is to maintain a neutral, vaguely pleasant face.”

Warneken and his team are on perpetual high alert to the ways in which results can be misleading. For example, in a study in which children, as quickly as 10 seconds later, handed Warneken a clothespin he had dropped as he hung clothes on a line, was it the child’s way of initiating a game (Do it again!)? Warneken altered the experiment, hurling the clothespin on the floor on purpose. The children didn’t respond.

And what are the limits to this helping behavior? Are toddlers capable of more-sophisticated forms of help? A video shows Warneken carrying a stack of magazines to a cabinet as an 18-month-old looks on. When, on a second trip, Warneken can’t open the closed cabinet doors and groans in obvious dismay (“We’ve all learned to be good actors,” he says), in nearly half the trials the child skitters to the cabinet and pulls the doors open. Do these children have similar cabinets at home? Warneken repeats the experiment using a box with a flap on the side and a small hole on the top. After the kids are shown how the unusual box works, he drops a spoon in the hole and fumbles to retrieve it. And yes, the children show the towering, clueless adult where the flap is.

More experiments follow, each upping the experimental ante. What if the child is given toys to busy himself with? It turns out that kids will leave their play to assist an adult who has dropped his pen. This looks a whole lot like altruism—helping another with no benefit to oneself—so Warneken went on to mix things up by offering toys as rewards when the children helped. “The result,” he says, “was that both the rewarded and not-rewarded groups helped at the same rate. In fact, the rewarding could even decrease the helping behavior; when I rewarded them over and over again, they were later less likely to help than children who’d received no award.”

Toddler chimpanzee Shangwe eats a hazelnut. Artis Royal Zoo, Amsterdam, The Netherlands. Photo by Kitty TerwolbeckToddler chimpanzee Shangwe eats a hazelnut. Artis Royal Zoo, Amsterdam, The Netherlands. Photo by Kitty TerwolbeckWhen it came to doing similar studies in chimpanzees, Warneken was able to reproduce the results he’d seen in toddlers. “It’s important to add chimpanzees to the equation—chimp parents don’t show them how to be responsible and teach them social norms,” says Warneken, whose office wall is adorned with side-by-side portraits of a child and a chimp. “And in chimps we saw the same basic helping behavior”—and not just toward humans familiar to them, but to human strangers.

They may look straightforward and spare, but the experiments Warneken’s groups conducted on helping, sharing, and fairness (one fairness study involved recruiting children on Boston Common) reflect months of brainstorming, logistical strategizing, and trial and error. “This is the most fun part of all,” says Warneken. “It’s a really creative process, with students bouncing around ideas.” For the helping experiment, using the problem of a Post-it note stuck to an adult’s shoe proved too complicated and was scrapped. Warneken “is incredibly smart and enthusiastic about the work, but he’s also a very structured thinker and encourages structured thinking in others,” says McAuliffe. “I learned from him how to be a much better experimentalist.”

His Radcliffe fellowship is allowing Warneken to synthesize the evidence he’s amassed in 10 years of empirical work. “My goal,” he says, “is to write a larger theoretical paper on the origin of human cooperation.”


Susan Seligson is a freelance writer and the senior editor of Bostonia, the alumni magazine of Boston University.

Search Year: