We presented subject rats with three simultaneous choice alternatives: releasing a restrained conspecific, engaging a non-restrained conspecific, or not socializing; they found no preference for helping

No preference for prosocial 'helping' behavior in rats with concurrent social interaction opportunities. Kelsey A Heslin, Michael F Brown. boRxiv, Nov 20 2020. https://doi.org/10.1101/2020.11.18.388702

Rolf Degen's take: https://twitter.com/DegenRolf/status/1330015424939962370

Abstract: Helping behavior tasks are proposed to assess prosocial or empathic behavior in rodents. This paradigm characterizes the behavior of subject animals presented with the opportunity to release a conspecific from a distressing situation. Previous studies found a preference in rats for releasing restrained or distressed conspecifics over other controls (e.g., empty restrainers or inanimate objects). An empathy account was offered to explain the observed behaviors, claiming subjects were motivated to reduce the distress of others based on a rodent homologue of empathy. An opposing account attributes all previous results to subjects seeking social-contact. To dissociate these two accounts for helping behavior, we presented subject rats with three simultaneous choice alternatives: releasing a restrained conspecific, engaging a non-restrained conspecific, or not socializing. Subjects showed an initial preference for socializing with the non-restrained conspecific, and no preference for helping. This result contradicts the empathy account, but is consistent with the social-contact account of helping behavior.