Psychologist Kiley Hamlin expected the 8-month old to pick the “good” puppet.
Hundreds of studies had already confirmed that infants, 166 of the 188 tested by Hamlin, prefer the helper puppets in experimental “morality plays” (where, for example, the “good” puppet returns a ball to another puppet, while the “bad” puppet steals the ball).
What did surprise Hamlin, the director of University of British Columbia’s Centre for Infant Cognition, was what came next.
In the next scene, the “good” and “bad” puppets played with a new puppet, who either rewarded each puppet with a toy or punished by taking away a toy. Hamlin assumed that babies would prefer the puppet who awarded the good one; but she was shocked to find that babies also favored the puppet that punished the bad one.
She repeated the study with younger children — and even 4-month-olds approved of harming those who harmed others.
The experiment suggests we’re born with knowing more than right from wrong — we’re also born with a sense of justice. Or at least vengeance.
“Quite a lot of what is happening very early is pretty sophisticated and consistent with how adults view the world,” Hamlin says. “This was a surprise to say the least.”
‘BABIES ARE MORAL ANIMALS’
This study — and similar research conducted at Yale, Harvard and others — fly in the face of notions of toddlers as blank slates who come into this world devoid of character or morality. Nor are they, as philosopher Jean-Jacques Rosseau described, “perfect idiots.”
Instead, research now shows that aspects of morality are inborn — and that babies are far more sophisticated in their judgment of the world than ever suspected.
Paul Bloom, author of “Just Babies: The Origins of Good and Evil” (Crown), out next month, summarizes the advances made in the field — in large part thanks to the Yale Infant Cognition Center that he runs with his wife, Karen Wynn.
“Babies are moral animals,” Bloom writes, “equipped by evolution with empathy and compassion, the capacity to judge the actions of others, and even some rudimentary understanding of justice and fairness.”
The field of infant cognition is a young one, with many of its significant advances occurring in the last decade, mostly because babies are notoriously difficult research subjects. Babies are “even harder to study than rats and pigeons, which can at least run mazes and peck at levers,” Bloom writes.
It was only in the 1980s that researchers developed a way to study infants by tracking the movement of their eyes as “windows to their souls,” or at least a window into their likes and dislikes. Babies often fixate on things that surprise or please them.
With this knowledge, psychologists found that babies adhere to what they call “naive physics,” or a knowledge, however limited, of the laws that govern the natural world (for example, what goes up must come down).
One study out of the University of Illinois at Urbana-Champaign showed that babies are mystified by magic tricks. When blocks float midair or objects disappear and reappear somewhere else (actions that violate our natural laws) babies tend to stare longer, showing researchers that babies “think of objects largely as adults do,” Bloom writes.
Bloom’s wife, Wynn, conducted a follow-up study in the 1990s, testing rudimentary math abilities in infants. In it, two Mickey Mouse dolls are shown and then placed behind a screen. When the screen is lifted and more or less than two dolls are shown, babies tend to look longer. Babies seem to know that 1+1 does not equal 3.
“Naive physics” grew into “naive psychology,” or the ability for babies to “get into the minds” of others.
One study showed that a child of only 15 months can grasp the idea of a “false belief.” In one study, an adult is shown an object in a box. He is then blindfolded as the object is moved to a different box. The baby watches and knows the object has been moved but still will expect the adult to look for the object in the original box, showing an “appreciation of the minds of others,” Bloom writes.
HARD-WIRED ALTRUISM
In 2000, Bloom, Wynn and Hamlin, then a graduate student, returned to the lab to take these assumptions one step further with the help of puppets.
One of the first puppet morality play goes as follows: A child watches on as a red ball tries desperately to climb a hill. Meanwhile, a yellow square blocks its path or a green triangle gentle pushes the red ball over the impasse. Yellow square is bad; green triangle is good. (These colors and symbols are all interchangeable, lest you think infants just like green or triangles better than yellow and squares.)
They tested 10-month-olds, then 8-month-olds and eventually 3-month-olds (who could only show preference with eye movements). Babies almost always picked the good guys.
“These experiments suggest that babies have a general appreciation of good and bad behavior, one that spans a range of interactions, including those that the babies most likely have never seen before,” Bloom writes.
The results were published in the journal Nature, shaking up the developmental psych community.
“This was mind-boggling. Before [that study], we had no idea that babies were capable of making these judgments. It just amazed us that we could be uncovering things that could fundamentally change the way we understand the infant mind working so late in the game,” Hamlin says.
Other studies followed. Dr. Felix Warneken, professor of psychology at Harvard, released his “helping studies,” working mainly with toddlers.
Warneken establishes “real life” situations in the lab, where an adult is encumbered in some way. Perhaps they are struggling to pick up an object out of reach or they are trying to open a door when their arms are full. In almost all cases, toddlers will offer to help without any prompting or promise of reward. (When children are offered a reward, helping responses begin to taper off, perhaps because they begin to see it “as a job,” says Warneken. Intrinsic motivation, or something you do for an internal reason, seems to be at work here.)
These studies were replicated with chimpanzees and they, too, helped other people and chimpanzees without a prompt.
There is other evidence to support the idea of hard-wired altruism.
Even a few days after birth, the sound of crying is unpleasant to babies — and they actually tend to cry more at the sound of another baby crying than of video recordings of their own cries. (Babies are not alone in this; rhesus monkeys avoid pulling levers for food if doing so will give another monkey a painful electric shock. So will rats.)
Empathy, it seems, has an evolutionary benefit and is caused by “biological or natural factors that are not due to society and culture alone.”
Interestingly, when children are old enough for society to filter in, Warneken found that children are more picky about whom they help. In one sharing study, 2-year-olds shared with everyone, including those that rudely refused to share with them. By age 3 ¹/₂, children became more discerning, showing higher rates of declining to help if the person in need had been mean to them in the past.
A SENSE OF JUSTICE
So babies choose the good guys over the bad guys. They also tend to help adults in need. But are they capable of more active responses to what they perceive as a slight or a wrong? Do they have a “moral code”?
These were the questions that led Bloom and Hamlin to pursue more complicated morality plays.
First, they added one more element: After seeing the “good” and “bad” puppet help or hinder another one, babies could decide to give a reward, like a toy or treat, or take one away. Babies almost universally rewarded the positive character and reprimanded the negative one. They were, in effect, dispensing justice.
Similar results were found when justice was administered by a third party.
In a test with 8-month-olds, a puppet was added who either punished or praised the “good” and “bad” puppets. Babies showed preference for the characters who praised the good and punished the bad.
In a third simulation, good and bad puppets played a game of catch with another puppet who either returns the ball or runs away with it.
While the 5-month-olds preferred the good puppet no matter what, older babies, around 8 months, were “drawn to bad actors when those actors were punishing bad behavior.”
“So, at some point after 5 months, babies begin to prefer punishers — when the punishment is just,” writes Bloom. And babies 21 months doled out the punishment themselves by physically taking away a toy or treat.
But can a baby distinguish between a person who intentionally harms someone and someone who unintentionally harms another person — key components of morality?
Hamlin says yes, citing a study she published this year in the journal Cognition.
In her study, a puppet repeatedly shows its preference for a duck. When given a choice between a duck and a flower, the puppet always picks the duck. In the next scene, the puppet tries to get the duck but is thwarted in some way. Another puppet joins in and either helps the puppet get the duck or helps the puppet get the flower. (Landing the duck is a win, getting the flower is a loss.)
Babies will only hold a grudge against the second puppet if he or she was originally present to see the original puppet’s preference for the duck. But if the puppet was not originally there — thus had no intention — then they will not.
“As adults we not only care about who gets harmed and who gets helped, it’s also about, ‘Did you mean to harm? Did you mean to help?’ ” says Hamlin. “Now we are seeing that babies are capable of using that same criteria.”
Babies don’t just like nice puppets. They like puppets with nice intentions, even if the outcome is not positive.
“If you help but fail, [babies] will like you just as much as if you help and succeed. And if you try to help and the character gets harmed, they will like you more than if you try to hurt and the character gets helped,” says Hamlin.
The ends do not justify the means in the world of baby morality.
“Maybe this all shouldn’t surprise us,” says Hamlin. “We are so incredibly complex in our social lives it makes sense that babies are doing some of that, too.”
DISTRUSTING THE OTHER
The wrinkle here is that though babies can identify good from bad, sometimes what’s “right” is skewed.
Babies make distinctions based on what is familiar and what is foreign almost immediately. (Newborn babies, for example, already show a preference for their mother’s face over that of a stranger’s).
This “simple preference for the familiar” extends to gender, language and race. Babies raised by a primarily white family tend to prefer white faces; infants raised by women will prefer female faces. But they don’t just prefer the familiar; sometimes they are actively hostile to the unfamiliar.They are naturally fearful of strangers and generally prefer their “own kind” in all manners, whether its someone who looks the same, speaks the same language, or just likes the same cookies.
Babies can be tyrannical about their preferences — if someone prefers graham crackers over chocolate chip cookies, or is a man rather than a woman, this is often grounds enough for punishment.
“This is not moral at all. Infants like those who are being nice, but they also like punishing those who are different from them,” Hamlin says. “I get the sense that babies go by the saying, ‘the enemy of my enemy is my friend.’ ”
There seems to be a evolutionary benefit, one that traces back to when humans lived in tribes. Survival depended on knowing the difference between those who are one of us (safe) and those who are other (dangerous).
“Kin has always mattered; it makes perfect Darwinian sense to favor someone who looks like you” or acts like you, for that matter “because that individual is likely to share more of your genes,” writes Bloom. And those who share our genes must be less likely to kill us.
But we do not live in hunter-gatherer units anymore. Which makes our innate moral system severely limited on its own.
Yes, we’re born nice (at least to those who look or act like us), but extending this niceness to those who are different from us?