Mechanisms of kin recognition

Abstract

Although kin recognition mechanisms are necessary neither for the operation of kin selection nor for optimal mate selection (e.g. inbreeding avoidance), once established, such mechanisms may accelerate the evolution of kin-directed behaviors. Aside from spatially-based recognition, in which organisms adjust their behavior principally in response to their immediate location, several behavioral mechanisms originally suggested by Hamilton recently have generated discussion and controversy. Familiar kin can be identified individually, but individual recognition mechanisms alone cannot serve to distinguish between closely and more distantly related kin, or to identify novel relatives. These kin can be identified through group recognition mechanisms that evaluate the extent of trait overlap among individuals to determine their probable genetic relatedness. Precisely whom an organism recognizes as kin using either type of mechanism may be dependent on its past social experience. Individual recognition permits discrimination of previously encountered conspecifics, whereas group recognition generally leads to discrimination of individuals sharing traits with previously encountered conspecifics. Individual and group recognition mechanisms are not mutually exclusive, and they may operate concurrently. Furthermore, they are not distinct, separable processes. Mistaken individual identifications become more likely as conspecifics show increasing phenotypic resemblance, and hence kin discrimination can result directly from individual recognition mechanisms. The extent to which kin are identified depends on a criterion rule which may fluctuate in response to social, spatial, and temporal factors. When favored by natural selection, kin recognition may be facilitated by the process of stimulus generalization, as trait matching is achieved within wider tolerance ranges. But such generalization may occur even in the absence of selection per se . Recognition effected through the action of hypothetical “recognition alleles” does not constitute a logical alternative to individual or group kin recognition mechanisms. In common with those processes, recognition allels must operate by the effective matching of phenotypes. Unlike kin recognition mechanisms, which assess similarity by generalized phenotypic comparisons, the phenotype compared by a recognition allele (or linkage group) is that it itself generates, leading to the possibility of intragenomic conflict. As suggested by Hamilton, alleles might be expected to induce their bearers to favor behaviorally conspecifics that share thier copies, regardless of the overall genetic relatedness of those conspecifics. Hence, recognition alleles, if they exist, would not invariably lead to kin identifications. Moreover, such alleles need not act in a manner that necessarily excludes learning. Through inbreeding experiments, a genetic component to the labels that elicit kin recognition has been demonstrated in some invertebrates. Genetically determined kin recognition templates, with which the labels are compared, have yet to be established. A “genetic recognition system” cannot be safely inferred when experiments fail to demonstrate experiential effects on recognition abilities. Kin recognition mechanisms can be characterized through detailed examination of the ontogenetic and sensory processes underlying recognition abilities, and of the ecological and social contexts in which kin-directed behaviors are expressed.