The evolutionary demise of a social interaction: social partners differ in the rate at which interacting phenotypes are lost

Phenotypic plasticity enables animals to adjust their behaviour flexibly to their social environment – sometimes through the expression of adaptive traits that have not been exhibited for several generations. We investigated how long social adaptations can usefully persist when they are not routinely expressed, by using experimental evolution to document the loss of social traits associated with the supply and demand of parental care. We allowed populations of burying beetles Nicrophorus vespilloides to evolve in two social environments for 48 generations in the lab. In ‘Full Care’ populations, traits associated with the supply and demand of parental care were expressed at every generation, whereas in ‘No Care’ populations we prevented expression of these traits experimentally. We then revived trait expression in the No Care populations at generations 24, 43 and 48 by allowing parents to supply post-hatching care, and compared these social traits with those expressed by the Full Care populations. We found that offspring demands for care decayed in the No Care populations more rapidly than a parent’s capacity to supply care. Furthermore, male care decayed before female care. We suggest that this reflects differences in the strength of selection for the expression of alternative traits in offspring, males and females, which can enhance fitness when post-hatching care is disrupted. Impact Summary Social interactions between animals are suggested to be increasingly vulnerable to breakdown in our changing world. Our experiments offer a rare insight into what happens next, by assessing in real time the durability of social behaviours that are no longer routinely expressed. Our results also have implications for conservation captive breeding programmes where compensatory husbandry techniques prevent trait expression and so could inadvertently induce rapid, irreversible trait loss. We investigated how long it took populations to lose the ability to express appropriate social behaviour when they had been prevented from doing so for many generations. We did this by evolving replicate populations of burying beetles Nicrophorus vespilloides in the laboratory for 48 generations. The burying beetle is a common insect that is well-known for caring for its larvae, although larvae can survive in the lab without any care at all. In two populations (“Full Care”), we allowed parents and offspring to interact during the supply of post-hatching care, as usual. In two other populations (“No Care”), parents were removed before offspring hatched and so could not interact socially with their young. Over the course of 48 generations of experimental evolution, we periodically revived social interactions between parents and offspring in the No Care populations. We assessed the extent to which larval begging behaviours, and parental care behaviours, had decayed by comparing their expression with those in the Full Care populations. We found that larval begging behaviour eroded rapidly in No Care populations, and more rapidly than the supply of care by parents. Furthermore, paternal care decayed to a greater extent than maternal care (which was largely unchanged relative to its expression in the Full Care populations). We suggest that these differences could be due to differences in the strength of selection on each family member for alternative traits to enhance fitness. ### Competing Interest Statement The authors have declared no competing interest.