Abstract
Choice of romantic partner is an enormously important component of human life, impacting almost every facet of day-to-day existence, however; the processes underlying this choice are remarkably complex and have so far been largely resistant to scientific explanation. One consistent finding is that, on average, members of romantic dyads tend to be more alike than would be expected by chance. Selecting for self-similarity is at least partially driven by phenotypic matching wherein couples share similar phenotypes, and preferences for a number of these traits are partly genetically influenced (e.g., education, height, social attitudes and religiosity). This suggests that genetically influenced preferences for self-similarity might contribute to phenotypic matching (and thus assortative mating), but it has never been studied in actual couples. In the present study, we use a large sample of twins to model sources of variation in self-similarity between partners. Biometrical modelling revealed that very little of the variation in the tendency to assortatively mate across 14 traits was due to genetic effects (7 %) or the shared environment of twins (0 %).
