Are Toddlers Noble Savages?

The bluestreak cleaner wrasse is a trusting fish. When a large predator swims into its cleaning station, the tiny wrasse will often enter the gills and mouth of the “client,” picking off ectoparasites, dead skin and stray bits of mucus. The wrasse gets a meal; the client gets cleaned; everyone wins, provided nobody bites.

This is a story of direct reciprocity. Nature is full of such stories, from the grooming of Sri Lankan macaques to the sharing of blood by vampire bats. In fact, such reciprocity is an essential component of biological altruism, or the ability to show concern for the wellbeing of others. Despite our reputation for selfishness, human beings are big believers in altruism, at least when it's rooted in reciprocity. If somebody gives us something, then we tend to give something back, just like those fish and bats. 

But where does this belief in reciprocity come from? One possibility is that were hard-wired for it, and that altruism emerges automatically in early childhood. This theory has been bolstered by evidence showing that kids as young as eighteen months dont hesitate to help a stranger in need. In fact, human toddlers seem especially altruistic, at least when compared to our chimp relatives. As Michael Tomasello, the co-director of the Max Planck Institute for Evolutionary Anthropology, writes in his recent book Why We Cooperate: "From around their first birthday - when they begin to walk and talk and become truly cultural beings - human children are already cooperative and helpful in many, though obviously not all, situations. And they do not learn this from adults; it comes naturally."

It's an uplifting hypothesis, since it suggests that niceness requires no education, and that parents don't have to teach their kids how to be kind; all we have to do is not fuck them up. As Tomasello writes, “There is very little evidence in any of these cases…that the altruism children display is a result of acculturation, parental intervention or any other form of socialization.” If true, then Rousseau was mostly right: every toddler is a noble savage.

However, a new paper in PNAS by Rodolfo Cortes Barragan and Carol Dweck at Stanford University suggests that the reality of children’s altruism is a little more complicated. Their study provides powerful evidence that young kids do like to help and share, but only when they feel like they're part of a sharing culture. They want to give, but the giving is contingent on getting something back.

The experiments were straightforward. In the first study, thirty-four 1 and 2 year olds were randomly assigned to either a "reciprocal play" or "parallel play" warm-up session. In the reciprocal play setup, the scientist shared a single set of toys with the child, taking turns rolling a ball, pushing buttons on a musical toy and passing plastic rings back and forth. The parallel play condition featured the same toys, only the scientist and child each had their own set. In both conditions, the scientist sat three feet away from the toddler and flashed a smile every thirty seconds.

Then, six minutes after play began, the scientist removed the toys and began testing the willingness of the children to offer assistance. They demonstrated a need for help in reaching four different objects: a block, bottle, clothespin and pencil. The children were given thirty seconds to help, as the scientist continued to reach out for the object.

The differences were stark. When children were first exposed to the reciprocal play condition, they offered help on roughly three of the four trials. However, when they first played in parallel, the rate of assistance plummeted to an average of 1.23 out of four.

In the second study, the scientists replicated these results with a stranger. Instead of having the children help out the same person they'd been playing with, they introduced an unknown adult, who entered the room at the end of playtime. Once again, the children in reciprocal play were far more likely to help out, even though they'd never met the person before. As Barragan and Dweck note, these are "striking" shifts in behavior. While children in the parallel play condition tended to ignore the needs of a new person, those in the "reciprocal play condition responded by helping time and time again, despite the fact that this new person had previously done nothing for them and now gave them nothing in return."

The last two studies extended these results to 3 and 4 year old children. Once again, the young subjects were randomly assigned to either a reciprocal or parallel play condition. After a few minutes of play, they were given the chance to allocate stickers to themselves or the adult. Those in the reciprocal play condition shared far more stickers. The last study explored the cause of this increased altruism, showing that children were more likely to say that a reciprocal play partner would provide help or share a toy, at least when compared to a parallel play partner.

I have a selfish interest in this subject. As a parent of two young kids, a significant portion of my day is spent engaged in negotiations over scarce resources (aka toys). In my small sample size, appeals to pure altruism rarely work: nobody wants to share their Elsa doll to cheer up another toddler. However, if that Elsa doll is part of a group activity – we can dress her up together! – then an exchange of Disney characters just might be possible. As this paper demonstrates, the key is to make sharing feel like a non-zero sum game, or one in which cooperation leaves everyone better off.

And this is where parents come in. As Barragan and Dweck note, their data contradicts “the notion that socialization has little or no part to play in early occurring altruism.”  Instead, their work demonstrates how the modeling of adults – the mechanics of our playing - strongly shapes the sharing instincts of children. I’ve made the mistake of believing that my kids will share once they’ve got enough toys, that altruism depends on a sense of abundance. (Ergo: my many trips to the Disney store.) But this appears to be an expensive mistake. After all, the parallel play condition offered kids the same playthings in greater amounts, since they didn’t have to share the ball or rings with the grown-up. But that surplus didn’t make them generous. Rather, their generosity depended on being exposed to an engaged and attentive adult, willing to get down on the ground and roll a ball back and forth, back and forth. My takeaway? Buy less, play more.

The wrasse and the bat seem to be born knowing all about reciprocity; those species have quid pro quo in their bones. But human kindness is more subtle than that. Unless we are exposed to the right conditions – unless someone shares their toys with us - then we never learn how much fun sharing our toys can be.

Barragan, Rodolfo Cortes, and Carol S. Dweck. "Rethinking natural altruism: Simple reciprocal interactions trigger children’s benevolence." Proceedings of the National Academy of Sciences 111.48 (2014): 17071-17074.