Social groups whose members have had sustained prior experience with each other frequently exhibit improved coordination and outperform groups whose members are unfamiliar with one another. The mechanisms by which familiarity assists coordination are not well known. Prior social experience may simply allow individuals to learn the behavioral tendencies of familiar group-mates and coordinate accordingly. In the absence of prior social experience, it would be adaptive for individuals to develop strategies for coping with unfamiliar others to minimize the disadvantage of unfamiliarity. To explore the dynamics of familiarity in shaping group behaviors, we used a highly social catfish, Corydoras aeneus, that utilizes a distinctive, observable tactile interactions. Here we describe this tactile interaction behavior, physical “nudges” that are deployed to initiate group movements and maintain contact with group-mates during group movements. We then report the results of two experiments exploring the relationship between nudges and coordination. First, within triplets of two familiar and one unfamiliar individual, we found no individual differences in nudging rate based on familiarity. Despite all individuals interacting at similar rates, however, unfamiliar individuals failed to coordinate as well as their familiar group-mates, and were more frequently absent from group movements. Second, comparing pairs of familiar with pairs of unfamiliar fish, there was no difference in the level of coordination between pairs. Instead, we found that unfamiliar pairs exhibited significantly higher nudging rates, suggesting that unfamiliar pairs could compensate for their unfamiliarity by nudging more frequently. In contrast, familiar individuals coordinated with comparatively little nudging, presumably because they were experienced with each other. Overall, these results suggest that nudges can be used to improve coordination of group activities, but that their usage is reduced in the case of familiar individuals, implying that these potential signals may be costly.

One of the primary functions of animal aggregations is defence against predators. Many social animals enjoy reduced predation risk as a result of grouping, and individuals within groups can benefit from information transferred by their group‐mates about a potential predator. We present evidence that a tactile interaction behaviour we term “nudging” substantially modified group responses to a potential threat in a highly social catfish, Corydoras aeneus. These catfish deployed nudges during flight responses, and these nudges were associated with a greater likelihood of group cohesion following a threat event. Increased nudging behaviour also resulted in longer flight responses, a potentially costly outcome in natural contexts. In addition, individuals that perceived the threat first were more likely to initiate nudges, implying that nudges could be used to alert group‐mates to the presence of a threat. Taken together, our results suggest that tactile communication plays an important role in mediating anti‐predator benefits from sociality in these fish.

Social groups composed of familiar individuals exhibit better coordination than unfamiliar groups; however, the ways familiarity contributes to coordination are poorly understood. Prior social experience probably allows individuals to learn the tendencies of familiar group-mates and respond accordingly. Without prior experience, individuals would benefit from strategies for enhancing coordination with unfamiliar others. We used a social catfish, Corydoras aeneus , that uses discrete, observable tactile interactions to assess whether active interactions could facilitate coordination, and how their role might be mediated by familiarity. We describe this previously understudied physical interaction, ‘nudges’, and show it to be associated with group coordination and cohesion. Furthermore, we investigated nudging and coordination in familiar/unfamiliar pairs. In all pairs, we found that nudging rates were higher during coordinated movements than when fish were together but not coordinating. We observed no familiarity-based difference in coordination or cohesion. Instead, unfamiliar pairs exhibited significantly higher nudging rates, suggesting that unfamiliar pairs may be able to compensate for unfamiliarity through increased nudging. By contrast, familiar individuals coordinated with comparatively little nudging. Second, we analysed nudging and cohesion within triplets of two familiar and one unfamiliar individual (where familiar individuals had a choice of partner). Although all individuals nudged at similar rates, the unfamiliar group-mate was less cohesive than its familiar group-mates and spent more time alone. Unfamiliar individuals that nudged their group-mates more frequently exhibited higher cohesion, indicating that nudging may facilitate cohesion for the unfamiliar group-mate. Overall, our results suggest that nudges can mitigate unfamiliarity, but that their usage is reduced in the case of familiar individuals, implying a cost is associated with the behaviour.