Vicarious reward unblocks associative learning about novel cues in male rats

Many species, including rats, are sensitive to social signals and their valuation is important in social learning. Here we introduce a task that investigates if mutual reward delivery in male rats can drive associative learning. We found that when actor rats have fully learned a stimulus-self-reward association, adding a cue that predicted additional reward to a partner unblocked associative learning about this cue. By contrast, additional cues that did not predict partner reward remained blocked from acquiring positive associative value. Importantly, this social unblocking effect was still present when controlling for secondary reinforcement but absent when social information exchange was impeded, when mutual reward outcomes were disadvantageously unequal to the actor or when the added cue predicted reward delivery to an empty chamber. Taken together, these results suggest that mutual rewards can drive associative learning in rats and is dependent on vicariously experienced social and food-related cues.


20
Humans and other animals have developed a capacity for mutual cooperative behaviour (Nowak, 2006  In the compound phase actor and partner rat are either both rewarded (Both Reward), actor rat is rewarded while the partner is not rewarded (Own Reward) or both actor and partner rat are not rewarded (No Reward). In the probe trials all learned cues are presented to the actor rat without reward. (C) Here, a timeline is shown with the different components that make up a single trial throughout the discrimination learning, compound phase and probe trials.

219
Discrimination learning. Actor rats (N=20) were trained on a discrimination task with counterbalanced visual or 220 auditory exemplars. All actor rats developed a conditioned response to their own aCS+, resulting in an increase in 221 time spent in the food trough on aCS+ trials in anticipation of reward, independent of cue modality or identity.

222
Concurrently, they learned to expect no reward during aCS-presentations, as witnessed by a steady decrease in 223 time spent in the food trough on aCS-trials ( Fig. 2A). A paired samples t-test examining the mean responding over 224 the last 4 days of conditioning was performed. We found a significant difference in time spent in the food trough

256
Responding on NR day 4 however (M = 25.97, SD = 7,68; ) did not differ anymore from responding to aCS-257 averaged over the last 4 days of discrimination learning; t(19)= 1.523, p = 0.114. We conclude from these results 258 that adding a second cue, which predicts either a reward (BR) or no reward (OR) to the partner, to the actor rats 259 aCS+ did not change the actor rat responding on BR and OR in comparison to its originally learned aCS+ response 260 at the end of discrimination. However, adding a second cue, which predicts no reward (NR) to the partner, on a 261 learned aCS-response did change the actor rats' responding on the NR condition. Here, a transient increase in 262 responding was found that subsided over 4 days of compound training. Most importantly, both BR and OR trials,

263
including the aCS+ as part of the compound, still elicited more responses than NR (aCS-compound) trials.

264
Probing vicarious reward value. In the probe trials, we aimed to show the effect of associative learning driven by 265 self and vicarious reward. In an extinction setting, the cues were presented in isolation, omitting reward.

296
Taken together, these results show that the actor rats exhibited more food cup directed behavior for the pCS+ 297 cue than both the aCS-and pCS-cue over 10 trials of extinction. This means that when actor rats have fully learned 298 a stimulus-reward association producing reward for themselves, adding a cue that predicted an additional reward 299 delivery to a partner rat unblocked associative learning about this cue, possibly due to vicarious reward 300 experience. We conclude from the contrast analyses explaining the interaction effect that the unblocking of the 301 novel cue lasts for approximately 6 trials and will use this analysis window going forward. In contrast, rats did not 302 spent more time in the food cup for the pCS-cue than the aCS-, suggesting that additional cues that did not 303 predict vicarious reward remained blocked from acquiring associative value.

304
Control group

305
In this control experiment, we impeded the exchange of direct social contact by introducing an opaque wall that 306 blocked visual contact, covered the interaction windows (Fig. 1) and dampened auditory cues from the partner test revealed a significant difference in time spent in food trough between the aCS+ (M = 56.40, SD =12.89) and 311 aCS-(M =17.20, SD = 10,18); t(7) =11.54, p < 0.001 (Fig. 2D). We furthermore also found that when looking at the shorter latency to entry for the aCS+ and a larger latency to entry for the aCS-over training ( Figure S2D).

317
Social learning. In the compound phase, rats were again trained side by side on the compound cues associated

364
These results show that, contrary to the findings for the experimental group, the rats in the control experiment 365 did not show more food cup directed behavior for the pCS+ cue than both the aCS-and blocked cue over 10 trials 366 of extinction. This suggests that when actor rats have fully learned a stimulus-reward association producing 367 reward for themselves, adding a cue that predicted an additional reward delivery to a partner rat does not unblock 368 associative learning when social information exchange and putative vicarious reward experience is blocked. reward arm in a T-maze that leads to play behavior more than an arm leading to a social encounter where play 571 was absent (Humphreys and Einon, 1981). Furthermore, social play can induce a social place preference 572 (Calcagnetti and Schechter, 1992) and rats are willing to lever press for social play reinforcement (Achterberg et actor rats about the cues inducing cue-directed behavior in the partner rat and presumably increase sign tracking 641 behavior for the actor rat towards the unblocked cue. As an example of such stimulus enhancement, Heyes et al.

642
( Heyes et al., 2000) showed that rats observing other rats performing a lever pressing discrimination task have a 643 higher lever-pressing rate for a previously observed ACS+ lever. In our task, responding at the food cup by actor