Background: Quantitative autistic traits are common, heritable, and continuously distributed across the general human population. Patterns of autistic traits within families suggest that more complex mechanisms than simple Mendelian inheritance-in particular, parent of origin effects-may be involved. The ideal strategy for ascertaining parent of origin effects is by half-sibling analysis, where half-siblings share one, but not both, parents and each individual belongs to a unique combination of paternal and maternal half-siblings. While this family structure is rare in humans, many of our primate relatives, including rhesus macaques, have promiscuous breeding systems that consistently produce paternal and maternal half-siblings for a given index animal. Rhesus macaques, like humans, also exhibit pronounced variation in social functioning.

Methods: Here we assessed differential paternal versus maternal inheritance of social functioning in male rhesus macaque offspring (N = 407) using ethological observations and ratings on a reverse-translated quantitative autistic trait measurement scale. Restricted Maximum Likelihood mixed models with unbounded variance estimates were used to estimate the variance components needed to calculate the genetic contribution of parents as the proportion of phenotypic variance (σ²_P) between sons that could uniquely be attributed to their shared genetics (σ²_g), expressed as σ²_g/σ²_P (or the proportion of phenotypic variance attributable to genetic variance), as well as narrow sense heritability (h²).

Results: Genetic contributions and heritability estimates were strong and highly significant for sons who shared a father but weak and non-significant for sons who shared a mother. Importantly, these findings were detected using the same scores from the same sons in the same analysis, confirmed when paternal and maternal half-siblings were analyzed separately, and observed with two methodologically distinct behavioral measures. Finally, genetic contributions were similar for full-siblings versus half-siblings that shared only a father, further supporting a selective paternal inheritance effect.

Limitations: These data are correlational by nature. A larger sample that includes female subjects, enables deeper pedigree assessments, and supports molecular genetic analyses is warranted.

Conclusions: Rhesus macaque social functioning may be paternally, but not maternally, inherited by sons. With continued investigation, this approach may yield important insights into sex differences in autism's genetic liability.