Heritability of affectionate communication: A twins study

ABSTRACT

Using a twin study design, we explored the extent to which affectionate communication is a heritable behavioral trait. Participants (N = 928) were 464 adult twin pairs (229 monozygotic, 235 dizygotic) who provided data on their affectionate communication behaviors. Through ACE modeling, we determined that approximately 45% of the variance in trait expressed affectionate communication is heritable, whereas 21% of the variance in trait received affection was heritable. A bivariate Cholesky decomposition model also revealed that almost 26% of the covariation in expressed and received affection is attributable to additive genetic factors. These estimates were driven primarily by females and those 50 years of age and older. The results suggest the utility of giving greater attention to genetic and biological influences on communicative behaviors by expanding the scope of communication theory beyond consideration of only environmental influences.

KEYWORDS:

• Affectionate communication
• genetics
• twin study
• ACE model
• heritability
• affection exchange theory

Acknowledgements

The authors gratefully acknowledge the assistance of Ally Avery and the Washington State Twin Registry.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes on contributors

Kory Floyd (PhD, University of Arizona) is professor of communication and professor of psychology at the University of Arizona

Chance York (PhD, Louisiana State University) is assistant professor of mass communication at Kent State University.

Colter D. Ray (PhD, Arizona State University) is assistant professor in the School of Communication and associate director of the Center for Communication, Health, and the Public Good at San Diego State University.

Notes

1 These percentages sum to >100 because some participants reported multiple racial backgrounds.

2 When asked at what age they moved apart from their twin, 2.5% of participants said they still lived with their twin, whereas 0.2% moved apart before age 6; 0.2% between 6 and 10; 0.5% between 11 and 14; 10.3% between 15 and 17; 69.5% between 18 and 21; 13.0% between 22 and 24; and 3.8% at age 25 or older.

3 We selected interpersonal affiliation and prosocial behavior as constructs to use for the power analysis because these seemed closest conceptually to affectionate behavior among the traits for which heritability estimates existed, and also because heritability estimates for these constructs were based on meta-analyses instead of single studies.

4 An anonymized view of the OSF preregistration is available at https://osf.io/hdnmp/?view_only=2417667886be487bafc16b51c403804e.

5 For more information on the WSTR, see https://wstwinregistry.org/.

6 Other-sex pairs are automatically classified as DZ.

7 Estimates of the variance components (A), (C), and (E) always sum to 100% or 1.0 of the variance explained in a given observed variable. By definition, any variance not accounted for by twins’ shared genes (A) or shared environment (C) is accounted for by the unique environment factor (E). In this sense, (E) represents any experience idiosyncratic to one twin and not the cotwin, while at the same time (E) serves as the model’s error term.

8 Only 11.2% of our respondents reported no longer living together prior to age 18. The majority (82.5%) reported separating during ages 18 to 24. In addition, prior twin studies confirm that a common upbringing is a safe assumption. That is, it is widely safe to assume a twin and cotwin are born at roughly the same time, raised together in the same household by the same parents, and are socialized within the same culture. One practical objection to this assumption is that, even if twins are reared in the same home, they may have novel experiences (e.g., having a distinct peer). ACE models address this objection in part through the (E) factor, which is intended to capture environmental experience unique to each twin.

9 Floyd (2019) pointed out that nearly every study that has explored the effect of biological sex has reported that women both express and receive more affectionate communication than do men, and those studies that have not documented such a difference have reported null results. There is no documentation of men, in any culture or age group, scoring significantly higher than women on either TAS-G or TAS-R.

10 Empirical evidence for tend-and-befriend theory’s claim of a sex difference in oxytocinergic reactivity to stressors is inconsistent, however (see, e.g., Floyd et al., 2010).