Vole population cycles: Kin-selection or familiarity?

Abstract

Listen

Translate article

The recent paper by Charnov and Finerty (1980) on Vole population cycles offers a unique view of population fluctuations in these rodents. These authors invoke kin-selection as a factor that may contribute to vole cycles, stressing the point that the coefficient of relationship, r, plays an important role in the elaboration of aggressive behavior among individuals, and that average r changes as populations cycle. They argue that when the population is low (from which there is little dispersal), close relatives probably interact most frequently, and because average r is high among these individuals, there are fewer aggressive encounters than when non-relatives interact (among whom average r would be lower). I believe that there may be a more parsimonious proximate explanation of behavioral changes that accompany population cycles, one that is consistent with recent experimental findings on sibling and individual recognition in rodents. Specifically, it appears that familiarity among individuals and not r plays the more important role in mediating the nature of social interactions. When individuals are familiar with one another they tend to fight less and display more amicable (cohesive) behavior than when they are unfamiliar with one another (for review see Bekoff 1981). Indeed, there are recent experiments that have adequately distinguished between the relative influences of familiarity and r in influencing social interactions (Porter and Wyrick 1979); the results clearly show that familiarity is the more important factor influencing social interactions and that r can be ~ den" by cross-fostering procedures in which siblings and nonsiblings are reared together as if they were siblings. With respect to vole population cycles, then, one could argue that in low populations, individuals are more familiar with one another and therefore should engage in fewer aggressive encounters than they would if population size increased and familiarity (and r?) among individuals became diluted. It is also important to mention that the problem of individual recognition must be considered here, and that with one possible exception (Wu et al. 1980), all experiments to date (on vertebrates) have shown that the ability to recognize individuals is learned, and that familiarity develops early in life through exposure learning (Porter and Wyrick 1979). While it would be useful to know the precise relationship between familiarity and r in natural populations, available data favor the argument that familiarity and not r would be the more important factor influencing social interactions among voles (and other species). It would be relatively simple to test this idea in the field using methods that have proven effective in captive situations, the result of which would be clarification of the importance of familiarity and r. The above argument about familiarity does not in any way lessen the importance of the consequences of relatives interacting or does it provide an ultimate explanation. Rather, the familiarity argument de-emphasizes the role of r in directly influencing interaction patterns and removes the necessity for animals to make r-related decisions. Certainly, it would be easier to distinguish familiar from non-familiar than r from non-r, though the two might be tightly linked in certain situations.