You are here: American University College of Arts & Sciences News Are Rats Empathetic?

Contact Us

Battelle-Tompkins, Room 200 on a map

CAS Dean's Office 4400 Massachusetts Avenue NW Washington, DC 20016-8012 United States

Back to top


Are Rats Empathetic? An Experimental Investigation

By  | 

Albino rat peering out from a small wooden box
Rat peering from start box of E maze.

Without doubt, humans are empathetic. But what of other animals? Do they as well come to the aid of others in distress? The possibility that this might be so has captured public imagination. Popular books and videos provide claimed demonstrations of empathy in non-humans. One such description comes from Frans de Waal, who describes a chimpanzee that climbed a tree to fetch an apple for her old aunt, who could no longer retrieve it herself. Another example comes from a popular video showing a buffalo herd rescuing one of their young from an alligator and a pride of lions. Many are convinced by these accounts as they seemingly demonstrate behavior that we call empathy. As experimentalists, however, we hold this phenomenon to a different standard. If animal empathy is a robust effect, it should be readily demonstrable in a laboratory.

The presence vs. absence of animal empathy is not merely of academic interest. In humans, empathy is so pervasive an attribute that its absence is often diagnostic of psychopathology. Because of the clinical importance of addressing cases where human empathy is absent, researchers have sought animal models of human empathy so they can try to understand the mechanisms by which empathy may work. One of the simplest organisms that could provide a model of empathetic action is the rat. Recently, our lab received a grant from the National Institute of Health to investigate whether there is potential to establish the rat as a model for human empathy.

Some researchers at Peggy Mason's laboratory at the University of Chicago have claimed to demonstrate empathy in the rat, but their demonstration may confuse the pursuit of social contact (rats like to be close to other rats) with empathy. In one example, one rat (free rat) learned to release a second rat that was trapped inside a small restraining tube. When released, the trapped rat could enter a large chamber where she was not restrained. The authors claim that since the free rat didn't derive any social contact from the situation (releasing the trapped rat increased its distance from the free rat), the free rat must have been acting empathetically. However, our replications of this study showed that rats don't do this unless the releasing response has been previously trained. Moreover, once the releasing response has been trained, it persists even when it no longer works. Such a result is inconsistent with empathetic action as a learned response. In another example, free rats freed a trapped rat from a pool of water. However, this experiment lacked an appropriate comparison test of the likelihood of freeing a rat from dry ground. Hence, it is impossible to know that freeing the wet rat occurred at higher frequencies than would obtain for a dry rat.

Because of these interpretative problems, our lab has focused on creating a novel paradigm to directly test if rats choose empathetic action, while controlling for alternative variables such as the pursuit of social contact. In one study we published, we used a maze to establish whether rats prefer to act empathetically. In the maze, the rat can choose to move down an aisle in which there is another dry rat in a chamber that will be freed once the choosing rat reaches it. Alternatively, the choosing rat can choose a second aisle where it can release a wet rat from a water pool. The experimental question is which aisle will the choosing rat select? Both aisles provide social contact with another rat, but only one choice-that of the aisle leading to the wet rat-evidences empathy.

We found that the rat would save the soaked rat over the rat in a dry box, a result that, at first blush, endorses empathetic action. However, in further experiments, we found that choosing rats preferred a pool of water over a dry chamber even in the absences of any trapped rats. In other words, rats like water over dry land. Given that in the prior condition rats were available to be released from either a pool of water or dry land, the alternatives were of equivalent appeal to the choosing rat to the extent that the choosing rat is pursuing social contact. However, one alternative, that of the pool of water, was preferred to dry ground. Thus, it seems that the choosing rat released the wet rat not out of empathy, but because it likes a water view!

In a current study, we are investigating if rats will choose to save other rats distressed by being confined to a restraining tube, or shocked, over rats that are presumably not distressed. Should we be able to convincingly show that rats act empathetically, we may be able to develop a non-human animal model of empathy. If not, we may just have to resign ourselves to the fact that humans are unique and have qualitatively different minds than other animals.