Back in March, Anderson Cooper famously compared Donald Trump to a 5-year-old. The presidential nominee had tweeted a flattering photo of his wife Melania juxtaposed with a less-flattering photo of the wife of his then-opponent Ted Cruz, and, when Cooper pressed him, he defended the tweet with a kindergartener’s default excuse: “I didn’t start it.”

Trump turned 70 this summer. But there may, in the broadest of senses, be science to support Cooper’s analogy—at least when considering what his election would look like to preschoolers. A lot has been written about kids’ less-than-favorable perceptions of the Republican presidential nominee: Children are turned off by his bullying tactics; his threats of deportation make them anxious; and they look up to role models, like parents or teachers, who despise him. But this reporting typically relies on feedback from children who are already reaching the upper levels of elementary school. To younger children, Trump may actually be quite appealing.

According to a new study published in the journal Developmental Psychology, those who are domineering and greedy might have an advantage in 3- and 4-year-olds’ social worlds. This is a world in which the bossy, aggressive, selfish kids thrive at the expense of their less-bossy, less-aggressive, less-selfish peers, and one in which unfair social hierarchies are something to be retained rather than eradicated. These preferences reverse with age, fading away around age 5, and by age 8, according to the study, they’re pretty much the opposite: Kids want to counteract inequality, to be good samaritans and help the little guy.

While a large body of research has explored how children perceive dominance and equality, the researchers—most of whom work at the French National Center for Scientific Research (CNRS)—say this goes one step further by investigating how children evaluate and respond to unfair situations by reallocating resources. Its findings offer powerful, if limited (and potentially controversial), insight into how a young child’s understanding of social justice and her political outlook evolves over time. And while the study wasn’t designed to inform educators, scholars suggest the study may prove itself especially useful now that schools are increasingly expected to focus on developing kids’ social and emotional skills on top of academics. Even the Every Student Succeeds Act, the newest version of the United States’s federal education law, calls on schools to pay more attention to social-emotional learning.

In one of the study’s experiments, the researchers staged a puppet show for 3-, 4-, 5-, and 8-year-old children, who were divided up by age. In the presentation, the puppets depicted identical boys, Léo and Nico, who on three occasions disagreed over what game to play. In one instance, for example, Léo wanted to play marbles yet Nico wanted to play ball; in another, Léo wanted to run but Nico wanted to jump. In all three instances, they ended up doing what Nico wanted to do: They played ball, they jumped, and so on.

The kids in the experiment were then presented with a box containing two chocolates, one piece bigger than the other. The more chocolate the puppets have, the researchers explained to the children, the happier they are: To whom would you give the big piece and to whom would you give the smaller one? Depending on their age, the kids acted in different ways. While a sizable majority of both the 3- and 4-year-olds said they’d give the larger piece of chocolate to the bossy, pushy Nico, almost all of the 8-year-olds—93 percent—opted to give it to the unassertive, subservient Léo. (The 5-year-olds were roughly split down the middle.) Notably, for the children who favored Léo, the likelihood that their reasoning focused on his misfortune increased with age.

A second experiment was equally revealing. Rather than divvy up a coveted resource, the children had to proactively redistribute such a resource—to take a coin from a child and give it to another of their choice. In what the researchers dubbed the “Robin Hood” paradigm, the children were presented with a drawing showing three boys, two of whom were “subordinate” and one of whom was “dominant,” having proclaimed to be “the boss” and convinced his two friends to play the game he wanted to play. The difference between the two subordinate boys was that one was poor (he only had one coin), while the other, like their dominant peer, had three coins. The researchers asked the children in the experiment to take a coin from either one of the rich boys and give it to the poor one. The results were pretty similar to those for the first experiment: A solid majority of the 3- and 4-year-olds took a coin from the rich subordinate boy while almost all of the 8-year-olds took a coin from the rich dominant one.

These experiments suggest a lot about kids’ social impulses when they’re not influenced by surrounding factors, which is both a strength and a weakness of the study. “Every individual is walking through this world with these competing motivations”—whether to be dominant or cooperative, for example—“and which takes precedence over the other is going to depend on the context,” said Meredith Martin, an assistant professor in educational psychology at the University of Nebraska-Lincoln who studies how children and adolescents deal with interpersonal relationships and bullying.

This may help inform efforts to combat phenomena such as bullying by showing how developmental factors and a child’s history influence those tendencies, minus all the contextual stuff happening in the given moment. At the same time, she continued, “It’s really hard to take the adolescent out of that context and understand how this would play out in bullying. Bullying is by its very nature a social phenomenon.”

Marjorie Martin, who’s taught elementary school for three decades, most of them in second grade but some of them with 3-year-olds, was even more skeptical. “I know that, generally, people think that [children] don’t really start to feel that inequality until they’re a little bit older, but in my experience the 3-year-olds and 4-year-olds want everyone to be happy,” she said, noting that she’s always taught in a private-school setting.

Marjorie (who isn’t related to Meredith) also emphasized that the experiments involved made-up individuals; if Léo and Nico were real, maybe the children would’ve reacted differently. “Even if you want to argue that their sense of reality is not fully formed at 3 years old, I still think they understand that they’re puppets,” she said. “They understand they’re not really hurting anybody’s feelings.”

Still, the findings contribute significantly the field of existing research, in which some studies suggest that children of all ages, even infants, favor equality, while others find that children actually tend to perpetuate inequality. What’s probably most noteworthy about this and similar studies is what they can illuminate about kids’ trajectory through school.

For one, they consistently point to age 8 as a developmental milestone. That’s when feelings of empathy—what the researchers in the latest study describe as a “vigilance for antisocial behavior”—really start to solidify. In the two experiments described in the new study, for example, the 8-year-olds likely evaluated what the dominant individual was doing and associated negative emotions with him, like guilt and sadness. The preschoolers, on the other hand, may have assessed the situations based on the outcome and associated the dominant figure with positive emotions like success.

Another study, titled “‘I had so much it didn’t seem fair’: Eight-year-olds reject two forms of inequity,” found that 8-year-old children frequently rejected unequal offers of candy even when that meant they themselves were the one who got the bigger share. The 4-to-7-year-olds almost never did. It’s also around age 8, according to the University of Nebraska’s Meredith Martin, when kids transition from using aggressive behavior to gain resource control—whether that resource be a toy or a child’s place in line—to more sophisticated, “prosocial” behavior. “What I see in kids that age is that they come up with more clever ways of gaining control, like, ‘Here, let me show you how to use that toy better,’” she said. It’s no coincidence that educators often describe third grade, which in the U.S. typically coincides with age 8, as the most important year in an individual’s academic career: It’s when children stop learning to read and start reading to learn.

Even more enlightening is that the latest study shows age 5 to be a particularly ambivalent turning point. In both experiments, something statistically similar happened around 5 years old: Roughly half of the 5-year-old participants leaned toward the preschoolers’ behavior, while the other half tended toward that of their 8-year-old peers. “What’s intriguing for me is the fact that at 5 years old, we really have two groups of children … Some children become more egalitarian where others still [favor] authority,” said Rawan Charafeddine, a cognitive scientist at the CNRS and the study’s lead author.

Educators may find this research useful as they look to better develop kids’ social-emotional skills, stamp out bullying, and ameliorate achievement disparities. After all, about one in four kids reports being bullied in school, making it the most common form of child victimization—and one that adults may be able to better address by encouraging young children how to respond when they’re a bystander and have the option to either condone or counter bullying. Meanwhile, a recent analysis on 15 years’ worth of research finds that improvements to school climate—which are marked by qualities including safety from bullying and student camaraderie—helps close the achievement gap.

The findings may also help teachers develop more advanced social skills in kids at an earlier age. With the Robin Hood experiment, Meredith Martin said, “there appears to be this developmental emergence of equity understanding and the desire to act equitably. And when I see something like that I think: What an awesome opportunity to capitalize on that emerging ability and to help shape it in a way that’s going to promote cooperation and compassion and understanding.” But as scientific as this new research might be, age 5 clearly isn’t a tipping point for everyone. Trump, for instance, doesn’t seem to have crossed the threshold from authority to equality.

Even Charafeddine, who conducted these experiments through a scientific lens without an eye on education, suggested that the findings can prove useful in classrooms. “Maybe what the study shows is the existence of a sense of justice that unties from dominance tendencies” around age 5. “So maybe [educators] can appeal more to these tendencies—maybe that would help the children to develop earlier rather than at 5 years old.”

“The core thing”—that sense of justice—“is there,” Charafeddine continued, “but the children may develop one or another tendency based on what they are encouraged to do.”