In recent decades, scientists have demonstrated that prosocial behaviors are not unique to humans, or even to primates. Rats, in particular, have proved surprisingly sensitive to the distress of conspecifics, and will often come to the aid of a fellow rat in trouble. In 2011, researchers showed that when rats were provided with a clear box containing chocolate chips, they usually opened the box and consumed all the chocolate.1 But when one box contained chocolate and another contained a trapped cagemate, the rats were more likely to open both boxes and share the chocolate.
But some rats didn’t play as nicely with others. In versions of the test that did not involve chocolate, only a rat and its trapped cagemate, researchers noticed that while some rats consistently freed their compatriots, others did not. In a new Journal of Neuroscience study, neuroscientists Jocelyn Breton at Northeastern University and Inbal Ben-Ami Bartal at Tel-Aviv University explored the behaviors and neural characteristics of helpers and non-helpers.2 They found that helper rats displayed greater social interactions with their cagemates, greater activity in prosocial neural networks, and greater expression of oxytocin receptors in the nucleus accumbens (NAc), providing clues about the mechanisms that govern prosocial behaviour.
“We appear to live in an increasingly polarized society where there is a gap in empathy towards others,” said Bartal in a press release. “This work helps us understand prosocial, or helpful, acts better. We see others in distress all the time but tend to help only certain individuals. The similarity between human and rat brains helps us understand the way our brain mediates prosocial decisions.”
To undertake these experiments, the researchers first divided the rats into pairs and allowed them to acclimatize to their cagemates for a few weeks. Then they placed the pair in the testing arena, where they allowed one rat to roam free and restrained the other in a clear box that could only be opened from the outside. While they were not trained to open the box, more than half of the rats figured out how to free their trapped companions and did so during multiple days of consecutive testing.
Behavioral analysis revealed that when rat pairs spent more time interacting with each other—engaging in behaviors like sniffing each other or lying down together—before testing, the free rat was more likely become a helper. In other words, rats with stronger social relationships were more likely to engage in prosocial behaviors.
The research team also compared the brains of helper and non-helper rats, analyzing gene expression in the NAc and the anterior insula (AI), brain regions previously associated with social reward and empathy. They found that expression of oxytocin receptor mRNA in the NAc but not the AI was elevated in the helper rats compared to their less-helpful counterparts, highlighting the role of the oxytocin signaling in the NAc as a potential driver of prosocial actions. In addition to oxytocin receptors, there were hundreds of other differentially-expressed genes in both the NAc and AI; many of these genes have incompletely understood functions and thus could be targets for future study in mediating helping behavior.
The team also examined the density of neurons positive for c-Fos, a marker of recent neural activity, in the brain-wide networks thought to regulate prosocial behaviors. Because a large number of brain regions are active during social interactions—including those related to attention, sensory processing, memory, emotion, and cognition—the researchers had previously sought to characterize the regions specifically related to rat prosocial behaviors in a 2021 study.3 To accomplish this, they compared rats’ brain activity when they were paired with a trapped cagemate, who they usually helped, and a stranger, who they rarely helped. They found that certain brain regions were more active in the first condition than in the second: In situations that elicited prosocial responses, rats showed more activity in the NAc, medial orbitofrontal cortex, prelimbic cortex, and lateral septum. In the present study, the researchers found greater activation in many of the same regions in the helpers, indicating that these brain networks may mediate both the propensity to help a familiar rat over a stranger, as well as individual differences in the willingness to help familiar cagemates.
Overall, this research reveals how specific genes and patterns of neuronal activity are associated with prosocial behaviors and suggests new directions for the study of individual differences in helping behaviors.
- Ben-Ami Bartal I, et al. Empathy and pro-social behavior in rats. Science. 2011;334(6061):1427-1430.
- Hazani R, et al. Neural and behavioral correlates of individual variability in rat helping behavior: A role for social affiliation and oxytocin receptors. J Neurosci. 2025.
- Ben-Ami Bartal I, et al. Neural correlates of ingroup bias for prosociality in rats. eLife. 2021;10:e65582