Newborns may have innate sense of prosocial behaviour

Newborn babies seem able to distinguish between prosocial and non-social behaviour in new study, though questions remain.

15 August 2025

By Emma Young



We are a remarkably cooperative species (at least, on the whole), capable of working together to everybody's benefit. Research into the cognitive and emotional underpinnings of this has pointed to everything from our ability to feel empathy to a fundamental preference for prosocial, rather than anti-social, people.

The fact that these abilities and attitudes are seen across cultures, and that many have been observed in even young babies, has led to the idea that human morality itself may stem (at least in part) from evolved adaptations for sustaining cooperation within large groups, note the authors of a recent paper in Nature Communications.

However, even the youngest babies in these earlier studies had at least three months' experience in the world, write Alessandra Geraci at the University of Catania and colleagues. In a bid to explore whether some of these abilities might really be innate, rather than learned, they recruited groups of healthy newborns, with an average age of just five days.

In the first of three experiments, 36 babies were held on a parent's lap while videos depicting two grey balls were presented side-by-side on a screen. One of the videos was designed to depict 'social approach' — with Ball B moving towards Ball A twice, before ending up close to it. The other video was designed to convey 'social avoidance' — with Ball B moving towards Ball A then backing up, and finishing up further away from it. These videos were positioned and designed in such a way that even newborns, with their limited vision, will have been able to see them.

Half the babies saw a 'social' version of the videos, in which both balls moved, and so appeared to have some agency. In the other, 'non-social' version, though Ball B moved, Ball A did not — making Ball A an inert object, rather than a social agent, the researchers write.

The team monitored which side of the screen the newborns looked at throughout the minute-long display (or until they looked away for at least 10 seconds, suggesting that they had lost interest).

The results showed that in the social condition, newborns spent more time looking at the approach video, rather than avoidance video, suggesting they could tell the difference between the two interactions. In the non-social condition, however, they didn't show any preference.

This experiment provides evidence that newborns can distinguish between an extremely simple form of prosocial versus antisocial interaction, and that they pay more attention to the social one, the team writes.

There is another possible interpretation, however. As Ball B's approach wasn't clearly beneficial to Ball A, it's possible that their attention was drawn because they interpreted the attention as a potential threat.

In a second experiment, the researchers sought to create a more moral context in the videos. This time, 36 newborns saw side-by-side videos that were designed to depict helping vs hindering. In both videos in the 'social' condition, Ball A appeared to try to move up a slope, but failed. In the 'helping' scenario, Ball B then moves from the bottom to 'push' Ball A up the slope. In the 'hindering' condition, Ball B moves down from the top and pushes Ball A to the bottom.
In the non-social condition, Ball A did not move independently — it was stationery until it was 'pushed' either up or down the slope by Ball B.

The team's analysis of the results revealed that newborns in the social condition, but not the non-social condition, looked for longer at the 'helping' than the 'hindering' behaviour.

The third and final experiment was a partial replication of this study. The team weren't able to recruit enough babies to run a full replication, so all 18 newborn participants saw the social version of the helping/hindering videos. Again, the team found that the newborns spent longer looking at the 'helping' rather than the 'hindering' actions.

The researchers conclude that the work marks a clear step forward for research into whether we come into the world able to distinguish between types of social interaction. "These findings go significantly beyond past work in lending support to claims that humans are in possession of unlearned mechanisms for detecting and evaluate (sic) key features of the sociomoral world," they conclude.

There is, however, still some discussion to be had about exactly what the newborns may be interpreting from these videos, and whether it's truly social at all. Another conclusion that could be drawn from the results from the partial replication, in the third experiment, could be that the babies spent longer looking at a video that violated their expectations of what would happen to a ball on a slope, given the force of gravity. (Other research suggests that we are born equipped with some 'intuitive physics'.)

In experiment two, the babies in the non-social condition didn't spend any longer, overall, looking at what might termed the 'anti-gravity' rather than the 'pro-gravity' video. But they spent less time, in total, looking at these videos than the babies in the 'social' condition, and this could have affected the results. (As there were was more going on in the 'social' videos, it's perhaps not surprising that they were more visually engaging.)

Exploring potential social — not to mention moral — judgements in newborns is clearly fraught with difficulty. But future studies will hopefully address some of the other factors that might have influenced the results in these experiments.

Read the paper in full:
Geraci, A., Surian, L., Tina, L. G., & Hamlin, J. K. (2025). Human newborns spontaneously attend to prosocial interactions. Nature Communications, 16(1). https://doi.org/10.1038/s41467-025-61517-3

SOURCE:

https://www.bps.org.uk/research-digest/newborns-may-have-innate-sense-prosocial-behaviour(accessed 21.08.25)