Infants as young as nine months old demonstrate a sophisticated, and somewhat unsettling, social evaluation system: they prefer individuals who harm those who are different from them. But this striking conclusion comes from a seminal study conducted by Kiley Hamlin, Neha Mahajan, Zoe Liberman, and Karen Wynn, published in Psychological Science in 2013. Now, their research, titled "Not Like Me = Bad: Infants Prefer Those Who Harm Dissimilar Others," fundamentally shifted the scientific understanding of the origins of social bias, intergroup conflict, and moral development. It suggests that the roots of prejudice and "us versus them" mentalities are not solely learned from culture or explicit teaching but emerge remarkably early in human development, potentially reflecting an evolved psychological adaptation for coalition detection.
The Core Question: Nature vs. Nurture in Social Bias
For decades, developmental psychologists debated whether infants enter the world as "blank slates" regarding social groups or if they possess innate mechanisms for navigating complex social hierarchies. Prior research by Hamlin and Wynn had already established that babies prefer helpers over hinderers—they like puppets that assist others in achieving goals and dislike those that obstruct them. This suggested an innate preference for prosocial behavior That's the whole idea..
Still, the 2013 study introduced a critical variable: similarity. The findings revealed that infant morality is not abstract or universal; it is deeply parochial. On top of that, the researchers asked a profound question: *Is the preference for helpers unconditional, or does it depend on who is being helped? * Do infants evaluate actions in a vacuum, or do they calculate the social relationship between the actor and the recipient? Babies do not simply prefer "good" people; they prefer people who are "good to my group" and, disturbingly, "bad to the other group That alone is useful..
Easier said than done, but still worth knowing.
The Experimental Design: Puppets, Snacks, and Social Choices
The elegance of the study lies in its simplicity, using a methodology perfectly suited for pre-verbal infants: preferential looking and reaching behavior. Since nine- and fourteen-month-olds cannot fill out surveys, researchers measure their preferences by observing which puppet they reach for or look at longer after watching a social scenario unfold.
Worth pausing on this one.
The experiment unfolded in several distinct phases:
1. Establishing Similarity and Dissimilarity (The Food Preference Task) Infants were first presented with two food items (e.g., graham crackers vs. green beans). The infant’s preference was noted. Then, two puppets were introduced: one puppet shared the infant’s food preference (the "Similar" puppet), and the other puppet disliked the infant’s preferred food and liked the alternative (the "Dissimilar" puppet). This created a minimal, arbitrary "group" distinction based solely on a trivial preference, stripping away race, gender, language, or familiarity.
2. The Moral Scenario (Helping vs. Harming) Infants then watched a series of puppet shows involving a third character—a "Target" puppet trying to open a box or retrieve a toy. Two scenarios were played out:
- The Helper Scenario: A puppet helps the Target achieve its goal (e.g., opens the box lid).
- The Harmer Scenario: A puppet hinders the Target (e.g., slams the box lid shut).
Crucially, the researchers crossed the Similarity factor with the Action factor. Sometimes the Similar puppet was the Helper and the Dissimilar puppet was the Harmer. Other times, the Similar puppet was the Harmer and the Dissimilar puppet was the Helper It's one of those things that adds up..
3. The Choice Phase Finally, the infants were offered a choice between the two puppets (the Helper and the Harmer). The dependent variable was simple: Which puppet did the baby reach for?
The Startling Results: "The Enemy of My Enemy Is My Friend"
The data revealed a clear, reliable, and developmentally stable pattern across both age groups (9 and 14 months) And that's really what it comes down to..
1. Preference for Helping Similar Others (The Baseline) When the Similar puppet helped the Target and the Dissimilar puppet harmed the Target, infants overwhelmingly chose the Similar Helper. This aligns with previous findings: babies like helpful individuals, especially those "like them."
2. The Critical Reversal: Preference for Harming Dissimilar Others This is the finding that shook the field. When the Similar puppet was the Harmer and the Dissimilar puppet was the Helper, infants preferred the Similar Harmer.
Let that sink in. Infants chose the puppet that slammed the box shut on a struggling character simply because that puppet shared their snack preference. Conversely, they rejected the puppet that helped the struggling character because that puppet liked a different snack It's one of those things that adds up..
3. No Preference for Helping Dissimilar Others When the Dissimilar puppet helped the Target, infants did not show a strong preference for that Helper over a neutral character. The "bonus" points usually awarded for helping behavior vanished when the recipient was an out-group member Nothing fancy..
4. Developmental Consistency The effect was present in 9-month-olds and 14-month-olds alike. This suggests the mechanism does not require years of socialization, language acquisition, or complex cultural immersion. It is an early-emerging, core component of human social cognition.
The Scientific Explanation: Why Do Babies Do This?
The researchers and subsequent theorists have proposed several non-mutually exclusive explanations for this counter-intuitive preference.
1. Coalition Detection and "Us vs. Them" Psychology
The leading evolutionary hypothesis is that the human mind contains specialized cognitive systems for coalition detection. Throughout evolutionary history, survival depended on identifying "who is on my team" and "who is a threat." Similarity—even arbitrary similarity like food preference—serves as a heuristic cue for coalition membership.
- Helping an in-group member strengthens the coalition (Good).
- Harming an out-group member weakens a rival coalition (Good).
- Helping an out-group member strengthens a rival (Bad).
- Harming an in-group member weakens one's own coalition (Bad).
From this perspective, the infant isn't being "mean"; they are engaging in sophisticated coalition management. They are tracking social alliances and preferring actors who advance the interests of their "team."
2. The "Common Enemy" Effect
The study supports the "common enemy" hypothesis. Infants may view the Dissimilar puppet as a potential rival or threat. When the Similar puppet harms the Dissimilar one, the Similar puppet is effectively acting as an ally against a rival. The infant prefers this ally. This mirrors adult geopolitics and playground dynamics alike: The enemy of my enemy is my friend.
3. Expectation Violation vs. Active Preference
Some critics initially wondered if infants were simply surprised by the Dissimilar Helper (expecting dissimilar others to be bad) rather than actively liking the Similar Harmer. That said, the reaching measure indicates an active approach motivation—they want to interact with the Similar Harmer. This goes beyond mere expectation violation; it reflects a positive valuation of the harmful act toward the out-group member.
4. Not Just "Familiarity"
Control conditions ruled out simple familiarity or perceptual matching. The infants didn't just prefer the puppet that looked or acted like them; they tracked the relational dynamic. The preference for the Similar Harmer only appeared when that puppet harmed the Dissimilar target. If the Similar puppet harmed a Similar target, infants disliked them (consistent with standard helper/hinderer preferences). The valence
