Over the past ten years, studies comparing monozygotic (identical) and dizygotic (fraternaal) twin pairs in terms of intelligence, personality, and social behavior have increasingly appeared in major research journals. Importantly, many of the variables examined are central constructs in communication theory. Assertiveness, social anxiety, self-monitoring, and empathy are but a few of the variables studied in the twins literature. In the present study, mean heritability estimates for three clusters (interpersonal affiliation, aggressiveness, and social anxiety) were computed. The studies selected for analysis focused on variables that have been frequently cited in mainstream reference texts and/or professional journals in communication. Conservative data analytic procedures were adopted. Specifically, (1) unless evidence of additive gene effects was absolute, nonadditive gene effects were assumed; (2) unless reliability coefficients were reported in a study, the highest reliability reported in the literature for the measure was employed in correction procedures; and (3) observer ratings, although imperfect, were not corrected for attenuation due to error in measurement. Results indicated that interpersonal affiliation was 70% heritable, aggressiveness was 58% heritable, and social anxiety was 65% heritable. Notably, heritability estimates were not larger for self-report measures than for behavioral observation. Criticisms of the twins design, possible moderators, and theoretical implications are discussed.