Though babies may seem innocent, a new study has found that infants might in fact have a mean streak. The study, which tested both 9- and 14-month-old babies, was published online in early March in the journal Psychological Science.  Lead author and former Yale psychology researcher Kiley Hamlin, currently a developmental psychology professor at the University of British Columbia, spoke with the News on Monday about her research methods and the surprising conclusion — babies like to see individuals with tastes different from theirs treated poorly.

Q: What were you looking for when you started this research?

A: From previous work we knew that given the choice, infants would choose a puppet with a similar taste, but we were interested in the negative side of liking those who are similar to us. It’s not weird to think that I would want to be friends with people who have things in common with me, like sharing interests or opinions. On the other hand, what that can mean is that we enjoy when bad things happen to individuals with different preferences. It’s fine to like those who are similar to us, but there’s a dark side to that which leads us to dislike those who are dissimilar to us — sometimes so much that someone who harms them is perceived as good. That’s the question we were interested in asking — how would babies react when someone with a different opinion is treated badly?

Q: What procedure did you use to explore that question?

A: We brought infants into the lab and asked them to indicate their food preference given a choice between graham crackers and green beans. Then we set up a situation in which one puppet had the same preference as the baby and another had the opposite preference as the baby. Half of the infants were then shown another puppet interacting with the puppet with a similar food preference, and this third puppet would either be nice or mean to the similar puppet. The other half of the infants saw the same nice and mean interactions, but with the dissimilar puppet. And we also included neutral interactions, where the third puppet was not explicitly nice or mean to the similar or dissimilar puppet.

Q: Were you surprised by what you found?

A: We were pretty sure that the babies would prefer the individuals who were nice to the similar puppet. But we thought that with the dissimilar puppet, there would be a competing motivation. The baby might think, “I like nice behavior, but I don’t like you very much.” We expected that babies might be confused by that comparison. In fact, we found that they weren’t confused at all. The babies just as strongly liked a puppet who was mean to a dissimilar individual as they liked a puppet who was nice to a similar individual. Obviously we thought this result was possible, because we did choose to run the study. But we were surprised by how extremely robust the results were. In the older age group of 14-month-old infants, every single baby preferred the individual who was mean to the dissimilar puppet, even though the other individual was showing niceness. The older infants also clearly showed that they liked the mean puppet more than the neutral puppet, and the neutral puppet more than the nice puppet, when it came to how they treated an individual who had a different food preference.

Q: So does this mean that babies have a “mean streak,” that they are prone to enjoy watching puppets they do not like suffer?

A: Well, it’s important to note that there are two possibilities. It could be that babies dislike dissimilar puppets and actually want bad stuff to happen to those puppets. Another possibility, which I think is probably the right one and a more reasonable one, is that the process is something like the enemy of my enemy is my friend. Rather than analyzing the action, the infants are analyzing what it means for an individual to be mean to another individual — it means that the individual doesn’t like the puppet it’s being mean to. And since the baby doesn’t like that puppet either, it thinks, “He and I should be friends.” The process isn’t necessarily nefarious.

Q: Is there any way to test which of the two possibilities is a true assessment of infant psychology?

A: Yes, there are a few things we could do. Namely, we could ask whether infants expect the individual who is being mean to be mean to others, too. If you think that this guy is a jerk, you might expect him to be mean to others, too, and if you liked those other individuals you would be upset by that. But if you think that this guy must agree with me, you should not think he’s bad in general. In a way, it’s like testing the “enemy of my enemy is my friend” theory in the opposite direction. Puppets are being nice to some individuals and mean to others, and we can access how the infant perceives those interactions. If someone you don’t like is nice to someone else, should you also dislike that someone else?

Q: Were there any obstacles you encountered along the way?

A: This study had a fairly easy time of it compared to the typical process of getting research published. And that’s probably because the data was so strong. It was quite compelling, and the results were all clear, so there wasn’t much that viewers could really say.