The Washington PostDemocracy Dies in Darkness

The disturbing thing scientists learned when they bribed babies with graham crackers

April 25, 2016 at 9:00 a.m. EDT
(ISTOCKPHOTO)

Humans appear to have a strong and innate sense of fairness. When somebody cheats other people, breaks the rules or otherwise behaves badly, we instinctively tend to try to avoid dealing with them, psychological research suggests. This tendency is probably an evolutionary adaptation that has allowed cooperative humans to thrive, and it may be a big factor in our incredible success as a species.

What’s fascinating is that this aversion to dealing with people who are unfair, bad or immoral appears present in very young children, too – even babies. Studies have shown that infants in the first months of life try to avoid dealing with social wrongdoers — for example, sharing less with them and helping them less — and they expect others to, too.

But just how strong is this moral aversion, and can it be overcome? In a study recently published in the journal Cognition, researchers set out to answer that question — basically, by bribing babies. The experiment was based on previous research suggesting that one can detect babies’ and young children’s preferences for people by looking at who they choose to accept food or toys from. The study looked at whether infants and children could be tempted into dealing with unsavory people, and just how much it would cost.

In the first experiment, the researchers asked 160 kids between 5 and 8 years old to choose whether they wanted to accept stickers from one of two fictitious characters, one of whom would give them one sticker, and the other of whom would give them two, four, eight or 16 stickers instead. Like the rational little animals they are, the children in the study reliably chose the larger offering instead of the smaller one.

But then the tests were repeated with an additional twist. The children were told that the character offering them one sticker was nice, whereas the character offering them more stickers was mean and had hit somebody on the playground. Here, in the groups offered two, four or eight stickers, 16 of the 20 children chose to accept the one sticker from the do-gooder instead. Only four of the kids were persuaded to deal with the meanie by the prospect of a larger haul.

But there was one group that responded slightly differently: the children who were offered 16 stickers from the bad character, rather than one sticker from the good guy. In this case, the discrepancy was so large that the children tended to choose the larger number.

You can see those results in the chart below. The height of the bars represents the percentage of children who chose the larger offering. The gray bar, "baseline," represents the first situation I mentioned, where the kids aren't given details about the characters. The black bar, "character-information," is the second situation where they are told whether the character is mean or nice.

The findings suggest that, when the winnings are modest, children will avoid “doing business” with a wrongdoer, the researchers say. “However, when the stakes are high, children show more willingness to ‘deal with the devil.’”

Don’t we all.

The researchers say it’s possible that the kids generally chose the good guy because of their desire to please the experimenter. They could have been more concerned with showing the adult in the situation they know the difference between right or wrong than actually making the moral choice.

In order to address this issue, the experimenters performed another, similar experiment with a younger group of subjects: babies. From developmental research, we know that children begin to truly understand and be concerned about what other people think about them between 3 and 5 years old. Below those ages, they are too young developmentally to engage in what the researchers call “reputational consideration” or “management.”

So in another experiment, the researchers asked 80 infants, all of whom were all about 1 year old, to participate in another test. (As is the case with tiny people, 16 of those infants were excluded from the final sample because of being fussy or failing to make a choice.)

The experiments used a tiny stage, complete with tiny curtain, and two rabbit puppets. One puppet offered the baby a plate with one graham cracker on it, while the other puppet offered either two or eight crackers. As for the older kids, the babies reliably choose the plate with more crackers.

But then the experimenters started a little morality play. The babies looked on as a lamb puppet on the stage tried and failed to open a clear box to get a toy. Then, one of the rabbit puppets would either help the lamb open the box and get the toy, or slam the box shut, after which the lamb puppet would dive face down next to the closed box in despair. Then, infants were offered the crackers again.

When choosing between one cracker from a good puppet or two crackers from a bad puppet, the infants “robustly” went with the do-gooder, the researchers say. But again, the results were somewhat different when the bad puppet offered a much bigger reward. When the bad puppet offered eight graham crackers, infants tended to choose the larger number, as the graph below shows.

It’s not clear why kids are more willing to interact with wrongdoers who offer more — whether their self-interest in getting more graham crackers just trumps their moral considerations, or whether they view the crackers as a kind of apology or retribution.

But the general results suggest that people are willing to pay personal costs — up to a point — to avoid dealing with immoral or deviant people, the researchers say, and that this behavior begins very young.

Psychologist Karen Wynn of Yale University is studying whether we are born with an innate sense of right and wrong. (Video: National Science Foundation)

You might also like:

What men and women wanted in a spouse in 1939 — and how different it is today

How to tell if other people think you’re hot, according to science

Being an expert literally changes how you see things