Jealousy was 29% heritable, and non-shared environmental influences explained the remaining variance (71%); no transmission of relationship models from parents to children

Why are some people more jealous than others? Genetic and environmental factors. Tom R. Kupfer et al. Evolution and Human Behavior, August 27 2021. https://doi.org/10.1016/j.evolhumbehav.2021.08.002

Abstract: Research on romantic jealousy has traditionally focused on sex differences. We investigated why individuals vary in romantic jealousy, even within the sexes, using a genetically informed design of ~7700 Finnish twins and their siblings. First, we estimated genetic, shared environmental and nonshared environmental influences on jealousy, Second, we examined relations between jealousy and several variables that have been hypothesized to relate to jealousy because they increase the risk (e.g., mate-value discrepancy) or costs (e.g., restricted sociosexuality) of infidelity. Jealousy was 29% heritable, and non-shared environmental influences explained the remaining variance. The magnitude and sources of genetic influences did not differ between the sexes. Jealousy was associated with: having a lower mate value relative to one's partner; having less trust in one's current partner; having been cheated by a previous or current partner; and having more restricted sociosexual attitude and desire. Within monozygotic twin pairs, the twin with more restricted sociosexual desire and less trust in their partner than his or her co-twin experienced significantly more jealousy, showing that these associations were not merely due to the same genes or family environment giving rise to both sociosexual desire or trust and jealousy. The association between sociosexual attitude and jealousy was predominantly explained by genetic factors (74%), whereas all other associations with jealousy were mostly influenced by nonshared environmental (non-familial) factors (estimates >71%). Overall, our findings provide some of the most robust support to date on the importance of variables predicted by mate-guarding accounts to explain why people vary in jealousy.

Keywords: JealousyTwinsMate value discrepancyTrustInfidelitySociosexualityIndividual differencesGenetics

4. Discussion

The current research aimed to shed light on why people differ in romantic jealousy. Our findings suggest that people differ in jealousy partly because of genetic influences, but mostly because of nonshared environmental influences. We did not detect an influence of the shared environment on jealousy. We also examined associations between jealousy and specific variables that have been hypothesized by mate-guarding accounts to influence jealousy proneness. Our findings provide some of the most robust evidence to date that mate value discrepancy, trustworthiness of a mate, and sociosexuality are associated with romantic jealousy.

Overall, 29% of variation in jealousy was attributable to genetic factors, with the remainder attributable to the nonshared environment. This genetic contribution to variation is on the low side compared to other psychological traits, including measurements of personality and emotions, for which the heritability is typically closer to 50% (Polderman, Benyamin, de Leeuw, et al., 2015). However, our finding is in line with those of Walum, Larsson, Westberg, Lichtenstein, and Magnusson (2013), who reported that sexual and emotional jealousy were 32% and 26% heritable, respectively. Also in line with Wallum et al., we found no evidence for sex differences in the magnitude of genetic and environmental influences on jealousy, or for different genetic or non-shared environmental influences operating in men or women. In other words, even though women reported higher jealousy than men, individual variation in jealousy within the sexes was influenced similarly by genetic and environmental factors.

The finding that familial environmental influences did not influence jealousy has theoretical implications. According to influential accounts of attachment theory, mental models of relationship expectations are transmitted from parents to children, through learning during infancy (Fonagy & Target, 2005; Van IJzendoorn, 1995; Verhage et al., 2016; c.f., Barbaro et al., 2017), and these mental models later determine emotion reactions, including jealousy, towards perceived relationship threats in adulthood (Mikulincer & Shaver, 2005; Sharpsteen & Kirkpatrick, 1997). Our finding that variation in jealousy is not influenced by familial environmental factors, which includes parenting, is inconsistent with these accounts. An implication is that research that seeks to understand variation in – and the development of – jealousy should attend more to genetic and nonshared environmental influences than to shared environmental factors such as parenting behavior. However, one caveat is that a limitation of twin studies is that they do not control for genetic and environmental interplay (for example, parental genes shaping the twin's family environment) which can confound the estimate of the influence of the family environment (Keller, Medland, & Duncan, 2010). Therefore, it is safest to say that we found no influence of the family environment ‘independent of genetic factors’ (Turkheimer, D'Onofrio, Maes, & Eaves, 2005).

In contrast to attachment theory's parental transmission account, mate-guarding perspectives hypothesize that jealousy should be primarily influenced by factors that increase the risk of infidelity by one's mate (Buss, 2013). These will often be socio-ecological variables (e.g., the attractiveness of one's mate, or the number of rivals in one's environment) which presumably derive more from the nonshared environment than the shared environment. Our finding of a substantial nonshared environmental influence on variation in jealousy is therefore consistent with mate-guarding accounts (though not uniquely consistent with those accounts). Note, however, that the estimate of the nonshared environment also includes measurement error.

The second aim of our study was to examine three of the variables predicted by mate-guarding accounts to influence jealousy: mate value discrepancy, cues to a mate's likelihood of infidelity (namely trust and actual experiences of infidelity), and sociosexuality. The strongest predictors of jealousy were more restricted sociosexual attitude and desire. Further, these relations were stronger for people in a relationship and for women. More sociosexually-restricted individuals may be more invested in fewer relationships and more motivated to protect them and, hence, experience more jealousy in response to cues to infidelity threats (Brase et al., 2014; Buss, 2013; Russell & Harton, 2005). Previous studies have most often not detected associations between jealousy and sociosexual orientation (Harris, 2003; Peters et al., 2014; Russell & Harton, 2005), but our findings were based on a much larger sample of individuals (N > 7000) than previous studies. The current finding was further strengthened by the discordant-twin design analysis within monozygotic twin pairs, which also detected a negative association between sociosexual desire and jealousy. This approach suggests that the association does not arise merely because sociosexual desire and jealousy emerge from the same genetic or family environmental sources. Results from the bivariate twin analyses were in accordance because they showed that nonshared environmental factors instead of familial factors explained the majority of the association (71%). Although it is therefore possible that restricted sociosexual desire causes higher jealousy, co-twin-control analyses cannot guarantee causal relationships or rule out reverse causation (McGue et al., 2010). Reverse causation (or bi-directional causation) is plausible, if, for example, individuals with higher jealousy pursue more exclusive relationships to reduce the possibility of infidelity by their mate.

Consistent with previous findings, people who reported being cheated on in the past, and those cheated on in their current relationship, also reported greater jealousy (Bendixen et al., 2015; Burchell & Ward, 2011; Edlund et al., 2006; Murphy et al., 2006; Sagarin et al., 2003). Additionally, having lower trust in one's partner was associated with higher jealousy (both on individual level and when we compared monozygotic co-twins discordant on trust in their partner). Therefore, findings suggest that variables assessing cues to a mate's likelihood of infidelity (trust and actual experiences of infidelity) relate to jealousy.

Also consistent with previous studies (Buss & Shackelford, 1997; Sidelinger & Booth-Butterfield, 2007), individuals who reported having a lower mate value than their partner reported higher jealousy. When examining associations within monozygotic twins only, those associations were non-significant (unlike associations between jealousy and sociosexual desire and mate trustworthiness), so the possibility that the association is due to similar genes influencing both mate value discrepancy and jealousy cannot be ruled out. However, the regression betas in the discordant twin design did not decrease in size, suggesting that the sample size of monozygotic twins may have been underpowered to detect an association. Moreover, the bivariate analyses did not detect familial influences on the association between jealousy and mate value discrepancy, indicating that the association between mate value discrepancy and jealousy is unlikely to be explained by similar genes or shared familial influences.

The study has some limitations. First, all measurements were based on self-reports. While the use of self-report questionnaires is common in psychology research (including most of the jealousy literature), they can be prone to measurement bias due to factors such as social desirability, which could, for example, have contributed to the low prevalence of cheating reported in our sample. Nonetheless, our self-report jealousy scale, which used 11 items describing jealousy-eliciting situations of varying severity, was likely to be a more sensitive measure of jealousy than measures commonly used. Many previous studies (e.g., Walum et al., 2013) have assessed jealousy with only two items asking participants how upset they would be in response to their partner's sexual infidelity and their partner's emotional infidelity. Another limitation was that the sample of discordant twin pairs contributing to the discordant-twin design analyses was much smaller than the sample in the overall regression analyses. Therefore, non-significant effects of sociosexual attitude, being cheated on in the past, and mate value discrepancy on jealousy within monozygotic twins could be due to lower power in these analyses. There are other potentially influential environmental variables that we were not able to assess in the current research. For example, perceived number and quality of rivals has been hypothesized to increase jealousy by increasing risks of cuckoldry or mate poaching (Buss, 2013; Pollet & Saxton, 2020), and perspectives other than the mate-guarding account propose that variables such as self-esteem are associated with jealousy (DeSteno, Valdesolo, & Bartlett, 2006). Future research on these factors using genetically informed studies would be valuable. Additionally, future research would benefit from using children-of-twins or nuclear twin designs that allow for the estimation of interplay between sources of variance that are impossible to disentangle using classical twin designs and might bias estimation of shared environmental influences (Keller et al., 2010).

In summary, this study confirms that people differ in jealousy partly because of genetic influences, but mostly because of nonshared environmental influences. Our findings provide some of the most robust evidence in support of several factors that have been hypothesized by mate-guarding accounts to influence jealousy proneness, and show that these factors similarly influence both men and women. Discerning the causes of variation in jealousy is an important step towards tackling the socially harmful consequences of jealousy, such as domestic violence and homicide.