Participation in social behaviors that enhance group-level fitness may be influenced by mutations that affect patterns of social epistasis in human populations. Mutations that cause individuals to not participate in these behaviors may weaken the ability of members of a group to coordinate and regulate behavior, which may in turn negatively affect fitness. To investigate the possibility that de novo mutations degrade these adaptive social behaviors, we examine the effect of paternal age (as a well-established proxy for de novo mutation load) on one such social behavior, namely religious observance, since religiosity may be a group-level cultural adaptation facilitating enhanced social coordination. Using two large samples (Wisconsin Longitudinal Study and AddHealth), each of a different US birth cohort, paternal age was used to hierarchically predict respondent's level of church attendance after controlling for multiple covariates. The effect is absent in WLS (β = .007, ns, N = 4560); however, it is present in AddHealth (β = - .046, p < .05, N = 4873) increasing the adjusted model R2 by .005. The WLS respondents were (mostly) born in the 1930s, whereas the AddHealth respondents were (mostly) born in the 1970s. This may indicate that social-epistatic regulation of behavior has weakened historically in the USA, which might stem from and enhance the ability for de novo mutations to influence behavior among more recently born cohorts-paralleling the secular rise in the heritability of age at sexual debut after the sexual revolution.
Keywords: Church attendance; Paternal age effect; Religious behavior; Social epistasis amplification model.