Social skew as a measure of the costs and benefits of group living in marmots

In group-living animals reproduction is often skewed such that some group members reproduce more than others. In addition to reproductive skew, group members may also exhibit social skew, where some individuals show particular behaviors more often than others. Significant social skew in behaviors such as anti-predator defense or social interactions may influence survival and reproduction. Therefore, social skew has the potential to translate into reproductive skew and affect group productivity. We measured social skew across groups in a population of yellow-bellied marmots (Marmota flaviventris). Several behaviors such as agonistic interactions, affiliative interactions, and first emergence were significantly skewed in most groups. Alarm calling, however, was infrequently skewed more than would be expected by random chance. Thus, marmot groups do not appear to have behavioral roles in terms of individuals acting like sentinels. Although significant social skew was present, it did not obviously affect fitness as measured by female reproductive success for each group. However, skew in individual-directed behavior (e.g. agonistic and affiliative interactions) did significantly correlate with the level of reproductive skew. Finally, the results were independent of the scale at which groups were defined. Behavioral variability appears to occur similarly across the entire marmot population. The results of this study illustrate that the quantification of social skew has potential to be a powerful tool for understanding the evolution of sociality.Introduction.Animal groups are characterized by variation in the degree to which individuals exhibit specific behaviors and variance across group members in the distribution of such variables has come to be called “skew” (Reeve Ratnieks 1993, Keller Reeve 1994, Ruzzante et al. 1995, Kokko Lindström 1997, Kokko et al. 1999, Nonacs 2000)