machiavellianism-shares-genetic-variance-with-emotional-concern-but-not-perspective-taking
Machiavellianism Shares Genetic Variance With Emotional Concern but Not Perspective-Taking

Machiavellianism Shares Genetic Variance With Emotional Concern but Not Perspective-Taking

A population-based twin investigation published in Behavior Genetics has revealed that Machiavellianism shares both genetic and unique environmental influences specifically with affective empathy, while exhibiting no reliable genetic or environmental relationship with cognitive empathy or general social skills. Conducted by Albena G. Ruseva, Peter K. Hatemi, and Nicholas G. Martin, the study analyzed data from young adult twins to resolve a long-standing behavioral controversy regarding whether interpersonal manipulators lack the cognitive capacity to understand others or simply lack the emotional resonance that restrains exploitative action. By simultaneously modeling the genetic and environmental architecture of multiple empathy dimensions alongside Machiavellian traits through multivariate structural equation modeling, the research team demonstrated that Machiavellianism is characterized by a hereditary reduction in emotional concern for other people, leaving mentalizing ability etiologically uncoupled. Furthermore, the findings indicate that Machiavellianism remains an etiologically distinct personality construct, with approximately eighty percent of its additive genetic variance operating independently of the broader empathy domain.

Understanding individual differences in social orientation has long occupied evolutionary and psychological researchers, who note that sociality represents a core defining characteristic of human development and may have served as the evolutionary catalyst for the expansion of the human neocortex. Across the human social landscape, orientations range from deep emotional sensitivity toward others’ distress to cynical exploitation where peers are viewed merely as instruments for personal benefit. Within this spectrum, empathy and Machiavellianism represent opposing behavioral dispositions that directly shape social cohesion, altruism, and interpersonal conflict. Empathy, traditionally conceptualized as a vital behavioral brake on interpersonal aggression and exploitation, is subdivided into two distinct yet related faculties: cognitive empathy, which describes the intellectual capacity to identify and comprehend another person’s perspective or internal mental state, and affective empathy, which involves the vicarious emotional resonance elicited by observing another person’s feelings. Conversely, Machiavellianism, recognized alongside narcissism and psychopathy as part of the Dark Triad, is characterized by cynical disregard for conventional morality, strategic exploitation, and interpersonal deceit, distinguished primarily by a calculated preference for subtle manipulation rather than physical coercion.

The exact nature of the relationship between Machiavellian tendencies and empathic capacities has sustained an ongoing theoretical debate between two competing paradigms. The empathic deficit hypothesis posits that manipulative individuals suffer from a fundamental social impairment akin to autism spectrum conditions, arguing that high Machiavellianism reflects an inability to take others’ perspectives or comprehend the emotional harm being inflicted. In sharp contrast, the evil genius hypothesis asserts that successful exploitation inherently requires sophisticated social cognition, perspective-taking, and mentalizing abilities, which manipulative individuals deploy strategically to advance their own self-interested goals while remaining unburdened by emotional distress. Prior phenotypic research has increasingly aligned with this evil genius framework, showing that individuals with elevated Machiavellianism maintain intact perspective-taking alongside atypical affective responses, such as affective dissonance, wherein they experience positive affect in response to others’ misfortune. However, because phenotypic correlations alone cannot determine whether observed negative associations reflect shared underlying biology, common developmental environments, or separate developmental processes, the etiological origins connecting empathy and Machiavellianism remained unresolved prior to this behavioral genetic investigation.

Earlier behavioral genetic inquiries into these traits evaluated empathy and Machiavellianism in isolation rather than within an integrated multivariate design, leaving their phenotypic covariation unexamined at the biometric level. While classical twin studies had established that empathic faculties exhibit moderate heritability, previous genetic evaluations of Machiavellianism had yielded uncertain conclusions. For instance, a prior twin study by Vernon and colleagues reported an additive genetic heritability estimate of thirty-one percent alongside a thirty-nine percent shared environmental contribution, yet both parameter estimates had wide confidence intervals spanning zero, leaving the biometric architecture of Machiavellian tendencies an unsettled empirical question. Moreover, no prior research had decomposed Machiavellianism alongside differentiated empathy subscales to determine whether its negative phenotypic association generalizes across all empathic constructs or concentrates exclusively within the affective domain. To resolve these ambiguities, Ruseva and colleagues formulated a multivariate investigation to determine whether the association is genetically or environmentally mediated, whether it exhibits domain specificity, and whether Machiavellianism can be modeled as the negative tail of a common empathy continuum.

The study analyzed data from the Economic Games Study, a population-based twin cohort nested within the broader Brisbane Longitudinal Twin Study in Australia, designed to evaluate genetic and environmental factors underlying economic, political, and social behaviors. The primary analytical dataset comprised 583 individuals, incorporating 253 complete twin pairs—consisting of 97 monozygotic and 156 dizygotic pairs—alongside 77 participating individual twin singletons who were retained using full information maximum likelihood estimation. Participants ranged from 19 to 31 years of age, with an overall mean age of 25.17 years, and included 340 females, representing fifty-eight percent of the sample. Because monozygotic twins were marginally older on average than dizygotic participants, age was statistically incorporated as a definition variable covariate across all biometric models. Sex composition did not significantly vary across zygosities, though sex exerted measurable effects on phenotype scores: female participants exhibited significantly higher mean scores on affective empathy and cognitive empathy, whereas male participants displayed significantly higher average Machiavellianism scores. Across the measures, age and sex accounted for 5.2 percent of affective empathy variance and 4.5 percent of Machiavellianism variance.

To capture empathic dimensions, the researchers administered an 18-item subset of the original 60-item Empathy Quotient developed by Baron-Cohen and Wheelwright, selected during initial study design through preliminary testing in comparable demographic samples. Parallel analysis and exploratory factor analysis with oblimin rotation demonstrated that a three-factor solution, comprising cognitive empathy, affective empathy, and social skills, provided superior statistical fit over alternative structures. Subsequent confirmatory factor analysis supported this structure with acceptable fit indices, yielding internal consistency coefficients of 0.76 for the six-item cognitive empathy scale, 0.72 for the five-item affective empathy scale, and 0.66 for the four-item social skills scale. Machiavellianism was measured using five items retained from Allsopp and colleagues’ scale after removing one item due to inadequate factor loading, producing an internal consistency of 0.65. Because Machiavellianism scores displayed severe positive skewness, with 70.5 percent of participants scoring zero and a sample mean of 0.52, the investigators modeled Machiavellianism as an ordinal variable under a liability threshold model, which assumes an underlying standard normal latent liability and stabilizes threshold estimation across skewed distributions.

Univariate biometric twin modeling revealed that additive genetic influences play a moderate and statistically significant role across all evaluated traits, with no detectable contribution from shared environmental factors. Intraclass correlations showed that monozygotic twin pairs exhibited substantially higher phenotypic concordance than dizygotic pairs across all four constructs, with dizygotic correlations falling below half of the monozygotic values. Formal comparisons between saturated baseline models, ACE models, and ADE models indicated that non-additive genetic dominance effects were negligible and that dropping shared environmental parameters produced no meaningful deterioration in statistical fit. Consequently, parsimonious AE models were retained for each phenotype, yielding standardized additive genetic heritability estimates of 0.34 for affective empathy, 0.29 for cognitive empathy, 0.29 for social skills, and 0.29 for Machiavellianism, with the overall Empathy Quotient showing a heritability of 0.33. Testing sex-limitation models for affective empathy demonstrated that while absolute mean levels differed between men and women, the relative magnitudes of genetic and unique environmental parameters were statistically equivalent across sexes, indicating identical biometric architecture.

Cross-twin cross-trait correlations provided preliminary evidence of selective genetic mediation between Machiavellianism and empathic traits prior to multivariate decomposition. Specifically, the cross-twin cross-trait correlation between affective empathy and Machiavellianism was noticeably negative among monozygotic twin pairs at -0.20, whereas the corresponding dizygotic correlation stood near zero at 0.04. In contrast, cross-twin cross-trait correlations between cognitive empathy and Machiavellianism, as well as between social skills and Machiavellianism, displayed no notable divergence between monozygotic and dizygotic pairs, suggesting negligible shared genetic etiology. These observed patterns directly motivated the execution of a quadrivariate Cholesky decomposition, which evaluated affective empathy, cognitive empathy, social skills, and Machiavellianism simultaneously. By incorporating all three empathy subscales into the same structural model, the researchers successfully partitioned out the general variance shared across empathy subscales, isolating the specific genetic and environmental variance components unique to affective resonance.

The primary quadrivariate AE Cholesky model demonstrated excellent fit to the empirical data and confirmed that the etiological association between empathy and Machiavellianism is strictly confined to affective responsiveness. The additive genetic correlation between affective empathy and Machiavellianism was moderate, negative, and statistically significant at rᵍ = -0.30, with a ninety-five percent confidence interval ranging from -0.89 to -0.05. A parallel negative association emerged at the unique environmental level, where non-shared environmental influences yielded a significant correlation of rᵉ = -0.38, with a confidence interval between -0.56 and -0.17. Conversely, cognitive empathy demonstrated no reliable association with Machiavellianism at either the genetic level, where rᵍ was -0.09, or the non-shared environmental level, where rᵉ was approximately 0.00. Social skills displayed a negative genetic correlation point estimate of -0.27 whose confidence interval crossed zero while encountering boundary constraints during profile likelihood estimation, precluding substantive interpretation. These results illustrate that the genetic predispositions predisposing individuals toward Machiavellianism specifically diminish affective emotional responsiveness while leaving cognitive understanding unaffected.

To evaluate whether the observed covariation reflects a single unified dimension or independent etiological pathways, the investigators fitted alternative structural equation models against the Cholesky baseline. An independent pathway model, which specified a single common genetic factor and a single common environmental factor, yielded acceptable fit indices but fit significantly worse than the Cholesky solution. Within this model, the common genetic factor loaded strongly and positively on affective empathy at 0.46 and negatively on Machiavellianism at -0.50, whereas its loadings on cognitive empathy and social skills were negligible. Furthermore, a common pathway model that constrained all genetic and environmental influences to operate through a single latent phenotypic factor fit the data poorly, yielding a comparative fit index of 0.841 and a root mean square error of approximation of 0.047. Machiavellianism loaded weakly on this shared phenotypic factor at -0.19, demonstrating that Machiavellianism cannot be conceptualized merely as the opposite pole of an overarching empathy construct, as roughly eighty percent of its genetic variance remains entirely trait-specific.

The study authors contextualized several methodological limitations that temper the precision of their findings while providing clear directions for future behavioral genetics research. The Machiavellianism instrument was limited to a brief five-item dichotomous scale, which, combined with the substantial floor effect where 70.5 percent of participants scored zero, imposed a statistical ceiling on detectable trait variance. Test-retest reliability across an eighteen-month interval on two available items yielded a correlation of 0.46, which extrapolates via the Spearman-Brown formula to an estimated reliability of 0.68 for the five-item composite, thereby attenuating heritability point estimates and broadening the ninety-five percent confidence interval of Machiavellian heritability from 0.01 to 0.53. In addition, the sample size of 583 individuals provided limited statistical power to separate additive genetic variance from non-additive genetic dominance or to detect modest shared environmental effects. The authors also noted that cultural factors, such as Australia’s egalitarian social norms and tall poppy syndrome, may socially suppress overt Machiavellian behaviors, potentially restricting the expression of underlying genetic predispositions.

Despite these measurement constraints, the investigation establishes the most detailed behavioral genetic decomposition of the empathy-Machiavellianism relationship conducted to date, offering empirical support for the evil genius model over the empathic deficit hypothesis. The findings confirm that Machiavellian traits do not stem from a broad neurocognitive inability to read, interpret, or predict the psychological states of other individuals. Instead, individuals possessing genetic predispositions toward Machiavellian manipulation maintain intact cognitive perspective-taking capacities, selectively lacking the shared emotional distress that typically discourages antisocial exploitation. Because roughly eighty percent of Machiavellianism’s genetic variance and eighty-two percent of its environmental variance operate outside the empathy cluster, the trait functions as an autonomous behavioral strategy that intersects with empathy at a single affective junction. Future investigations utilizing longer multi-point response scales, expanded twin-family cohorts, and molecular genomic methods will be essential to map the broader neurobiological systems that sustain this strategic orientation.

Subject of Research: Genetic and environmental architecture linking Machiavellianism to affective and cognitive empathy

Article Title: Machiavellianism Shares Genetic Variance with Affective but Not Cognitive Empathy

Article References: Ruseva, A. G., Hatemi, P. K., & Martin, N. G. (2026). Machiavellianism Shares Genetic Variance with Affective but Not Cognitive Empathy. Behavior Genetics. https://doi.org/10.1007/s10519-026-10294-w

Image Credits: AI Generated

DOI: 10.1007/s10519-026-10294-w

Keywords: Machiavellianism, affective empathy, cognitive empathy, social skills, behavior genetics, twin study, Cholesky decomposition, Empathy Quotient, additive genetic variance, non-shared environment, Dark Triad, liability threshold model, Brisbane Longitudinal Twin Study, evil genius hypothesis, empathic deficit hypothesis, heritability, common pathway model