Experiment 2: social contact-independent targeted helping task

In all of studies in Experiment 2, targeted helping behavior was evaluated (n = 29pairs/sex) using a custom (Med Associates; Fairfax, VT, United States) operant box developed with three chambers (Cox and Reichel, 2020, 2021; Cox et al., 2022), as seen in Figure 1A. In this apparatus, Targets were placed in 100 mm of water in the wet compartment and Observers on a dry platform with access to a chain that opened an automated door. The Target was released into a dry compartment separate from the Observer (Exp. 2a, n = 21 pairs/sex). Throughout Experiment 2, latency to chain pull was taken as an index of helping behavior (Ben-Ami Bartal et al., 2011; Cox and Reichel, 2020; Cox et al., 2022). Trials (20 total across 10 days, labeled “Acquisition”) lasted a total of 300 s (5 min) regardless of the chain pull latency. For all behavioral assessments, two trials were conducted daily during the rats’ dark cycle (Cox and Reichel, 2020; Cox et al., 2022). If the Observer did not pull the chain within the allotted time, the experimenter ended the trial and released the Target. We focused on two timepoints for analysis (see Figure 1): early acquisition (EA, average of days 1–2), as an indication of an initial helping response; and late acquisition (LA, average of days 9–10) to examine the effect of habituation to aiding a familiar conspecific. In order to determine the importance of the Observer visualizing the Target to learn to release the distressed conspecific, a separate cohort of male and female rats (Exp. 2b, n = 8 pairs/sex) was tested in the same operant box, but the Plexiglas divider present between Observer and Target was painted black to prevent either animal from seeing through it.