The BDNF Val66Met modulates the Dark Triad: empathic concern and aggression as mediators

ABSTRACT

Objectives

Empathy and aggression are features of the Dark Triad. They are linked to the BDNF Val66Met polymorphism. Here, we investigated the relationships between this polymorphism and the Dark Triad traits.

Methods

We genotyped the BDNF Val66Met of 698 college students and measured personalities of the Dark Triad, self-report empathy, and aggression with scales of the Short Dark Triad, Interpersonal Reactivity Index, and Buss and Perry’s Aggression Questionnaire, respectively.

Results

The Met/Met genotype of the BDNF Val66Met was related to higher scores on Machiavellianism and narcissism as compared with the Val carriers. The association between the BDNF Val66Met and Machiavellianism was mediated by aggression and empathic concern, while the association between this polymorphism and narcissism was only mediated by aggression.

Conclusions

This study highlights the importance of the BDNF Val66Met to the Dark Triad and reveals psychobiological pathways where aggression and empathic concern mediate the link between the BDNF gene and the Dark Triad.

KEY POINTS

What is already known about this topic:

(1) Twin studies have indicated that the Dark Triad has strong genetic predispositions.

(2) The BDNF Val66Met is associated with self-report empathy and aggression.

(3) Both impaired empathy and high aggression are features of the Dark Triad.

What is this topic adds:

(1) The Met/Met genotype of the BDNF Val66Met is related to higher Machiavellianism and narcissism as compared with the Val carriers.

(2) The Met/Met genotype of the BDNF Val66Met is related to higher aggression but lower empathic concern as compared with the Val carriers.

(3) Both empathic concern and aggression mediate the association between the BDNF Val66Met and Machiavellianism, while only aggression mediates the association between this polymorphism and narcissism.

KEYWORDS:

• Dark Triad
• empathy
• aggression
• brain-derived neurotrophic factor

Introduction

The traits that make up the Dark Triad are aversive features of human personalities (Paulhus & Williams, 2002). Individuals who score highly on the Dark Triad are often perceived as braggarts, manipulators, and downright aggressors (Edwards et al., 2017; Paulhus & Williams, 2002). Studies have indicated that the Dark Triad is related to undesirable life styles (Gott & Hetzel-Riggin, 2018; Turnipseed & Landay, 2018) and antisocial behaviours (Baughman et al., 2012, 2014).

Psychological constructs of the Dark Triad

The Dark Triad consists of three conceptually distinct but empirically overlapping personalities (Paulhus & Williams, 2002). Machiavellianism is characterized by a lack of empathy, duplicitous interpersonal style, and manipulative behaviours (Wilson et al., 1996). Psychopathy is characterized by an aggressive interpersonal style with a lack of empathy, short-term gratification, and high antisocial dispositions (Hare & Neumann, 2008; Jonason et al., 2015). Narcissism describes traits such as self-centredness, grandiosity, and a lack of interpersonal warmth (Bushman & Baumeister, 1998; Emmons, 1987).

Genetic foundations of the Dark Triad

Twin studies have indicated that the Dark Triad is strikingly influenced by one’s genetic predispositions (Campbell et al., 2009; Julie & Daniel, 2020). Psychopathy and narcissism have moderate to large heritabilities (Campbell et al., 2009), which suggests a cluster of candidate genes underpin the Dark Triad traits. Two genetic-association studies have identified specific genes of the Dark Triad, such that the number of long alleles of the serotonin transporter promoter polymorphism is positively related to the level of narcissism and the CC genotype of rs6280 of dopamine receptor D3 is associated with higher Machiavellianism than the T allele carriers (Brammer et al., 2016; Montag et al., 2015). These findings suggest that the genes involved in brain functions play crucial roles in the biological basis of the Dark Triad. Due to the little progress in the follow-up genetic studies, we further explored a new candidate gene of the Dark Triad.

The BDNF as candidate gene underlying the Dark Triad

Brain-derived neurotrophic factor (BDNF), an endogenous protein in mammalian brain, is related to the growth and differentiation of the nervous system (Fritsch et al., 2010; Fukumitsu et al., 2006). The BDNF Val66Met is a common single-nucleotide polymorphism in the BDNF gene. This polymorphism produces an amino acid changed from valine (Val) to methionine (Met) at codon 66 in BDNF protein, which regulates the BDNF secretion and synaptic plasticity (Egan et al., 2003; Xu et al., 2000). Studies have indicated that the BDNF Val66Met is associated with several types of personalities (Montag et al., 2010; Tsai et al., 2004). Inspired by these findings, we investigated whether this polymorphism is related to the Dark Triad. Given that the Dark Triad is influenced by growth environments (Birkas et al., 2018; Guo & Feng, 2017; Jonason et al., 2014), the variables of parental styles, childhood trauma, and family relationship were controlled when we investigated the associations between the BDNF Val66Met and the Dark Triad traits.

The Dark Triad and empathy

The Dark Triad is featured by a lack of empathy (Jonason et al., 2015; Jonason & Krause, 2013; Turner et al., 2019). Empathy refers to the capacity to substitute oneself into situations and generate feelings synchronously (Davis, 1983). According to the psychological constructs, it can be specifically described as cognitive empathy (ability to recognize and understand others’ feelings) and emotional empathy (ability to share others’ emotions). Studies have revealed that the two kinds of empathic traits are significantly connected with the Dark Triad, but are related inconsistently to the personalities of the Dark Triad. For instance, an early study has shown that all the Dark Triad traits are associated with low cognitive empathy, while psychopathy and Machiavellianism are correlated with low emotional empathy (Jonason & Krause, 2013). Similarly, several additive studies indicated that Machiavellianism is associated with low cognitive and emotional empathy, while narcissism and psychopathy hardly reach a constant relationship with empathy (Jonason et al., 2015; Pajevic et al., 2018; Turner et al., 2019). Due to these mixed findings, we further investigated the relationships. Moreover, a study has suggested that self-report empathy is modulated by the BDNF Val66Met (Taschereaudumouchel et al., 2016). The numbers of the Met allele are positively linked to the ability of fantasy (i.e., individuals immerse themselves into the feelings of characters in fictional situations), but negatively related to perspective taking (i.e., individuals’ cognitive propensity of adopting others’ view). Inspired by the links between empathy and the Dark Triad traits, we investigated whether empathy mediates the association between the BDNF Val66Met and the Dark Triad.

The Dark Triad and aggression

The Dark Triad is characterized by a high trait of aggression. Machiavellianism is associated with high hostility (Jones & Neria, 2015); psychopathy is positively related to verbal and physical aggressive traits (Jones & Neria, 2015); and narcissism is positively related to high aggression such as being excluded and insulted (Bushman & Baumeister, 1998; Webster et al., 2016). Studies have indicated that the BDNF Val66Met is associated with aggression. The larger number of Met alleles is associated with increased aggression (Kretschmer et al., 2014; Spalletta et al., 2010). Considering the relationships between the Dark Triad and aggression, we investigated whether aggression mediates the associations between the BDNF Val66Met and the Dark Triad traits.

Aim of the current study

In summary, this study investigates the association between the BDNF Val66Met and the Dark Triad. Considering that the Dark Triad is featured with a lack of empathy and high aggression, we further investigated whether empathy and aggression mediate the association between this polymorphism and the Dark Triad. Given that individuals with low empathy and high aggression show the high Dark Traid (Jonason et al., 2015; Jonason & Krause, 2013; Jones & Neria, 2015; Turner et al., 2019), and the larger number of Met alleles is associated with increased aggressive behaviour and low ability of perspective taking, we predict that the number of Met alleles is positively associated with the levels of Dark Triad personalities, in which empathy and aggression mediate the associations.

Methods

Participants

Power analysis indicated that a sample of 526–787 individuals is required (two-tailed α = 0.05, 1-β = 0.80) if the regression coefficient of approaches 0.01–0.015 (i.e., coefficient of determination R² = 1.0 − 1.5 %). We recruited participants through an advertisement on the university internet site. Six hundred and ninety-eight college students (456 females, mean age = 24.16 ± 2.1 years) completed the assessments. The students who entered our assessments got a ¥10 reward. This study was in accordance with the Declaration of Helsinki and approved by the Ethics Committee of College of Life Science, Northwest University, China.

Psychological dispositions assessments

Short Dark Triad (SD3)

The Dark Triad was measured with the Chinese version of SD3 (Jones & Paulhus, 2014; Zhang et al., 2020). This scale consists of three subscales. The Machiavellianism subscale taps manipulative interpersonal style, with less concern for other’s well-being and emotional coldness (e.g., “I like to use clever manipulation to get my way”).The psychopathy subscale measures destructive behavioural patterns along with high retaliatory (e.g., “I like to get revenge on authorities”). The narcissism subscale taps self-centredness and grandiosity (e.g., “People see me as a natural leader”). Each item was scored with a 5-point Likert scale (“1” = strongly disagree, and “5” = strongly agree). The internal consistency (Cronbach’s α) for the subscales of Machiavellianism, psychopathy, narcissism were 0.80, 0.85, 0.72, and 0.70, respectively.

Interpersonal Reactivity Index (IRI)

Self-report empathy was measured with the Chinese version of the 28-item IRI (Davis, 1983; Rong et al., 2010). This scale includes the subscales of Perspective Taking, Fantasy, Empathic Concern, and Personal Distress. The Perspective Taking subscale evaluates individuals’ cognitive propensity of adopting others’ view (e.g., “I sometimes try to understand my friends better by imagining how things look from their perspective”); the Fantasy subscale assesses the extent to which individuals immerse themselves into the feelings of characters in fictional situations (e.g., “I really get involved with the feelings of the characters in a novel”); the Empathic Concern subscale measures one’s compassion and concern in response to others’ misfortune (e.g., “I often have tender, concerned feelings for people less fortunate than me”); the Personal Distress subscale taps personal anxiety and discomfort when observing others’ pain (e.g., “When I see someone who badly needs help in an emergency, I go to pieces”). For each item, the respondent scored on a 5-point Likert scale (“0” = do not apply to me at all, “4” = “describe me very well”). The scoring procedure followed Davis’ suggestion (1983). The internal consistencies of the subscales of Perspective Taking, Fantasy, Empathic Concern, and Personal Distress were 0.71, 0.62, 0.80, and 0.70, respectively.

Buss and Perry’s Aggression Questionnaire (BPAQ)

The trait of aggression was measured with the Chinese version of the self-reported scale of BPAQ (Buss & Perry, 1992). This scale measures one’s dispositions for physical aggression, verbal aggression, anger, and hostility. The response to each item was scored on a 5-point Likert scale (“1” = extremely uncharacteristic, “5” = extremely characteristic). Cronbach’s α of this scale was 0.87.

Growth environmental variables

Considering that the Dark Triad is influenced by growth environments (Birkas et al., 2018; Guo & Feng, 2017; Jonason et al., 2014), the psychological distress, parental style, childhood trauma, and family functions of the participants were assessed for analysis, which are demonstrated in the Supplementary Materials.

Genotyping

We collected oral mucosa cells and extracted genomic DNA from with the Chelex-100 method (De Lamballerie et al., 1994). The BDNF Val66Met was amplified with the upstream primer, 5ʹ-GCAAACATCCGAGGACAA-3ʹ, and the downstream primer 5ʹ-TACTGAGCATCACCCTGG-3ʹ. Polymerase chain reaction (PCR) mixture included 2.50 µL reaction Mix, 0.50 µL DNA template, 1.50 µL ddH₂O, 0.50 µL (25 pmol/µL) primers. The PCR was performed with initial denaturation at 95°Cfor 5 min, 34 cycles at 94°C for 30 s, 60°Cfor 30 s, an extension at 72°Cfor 30 s, and a terminal extension at 72°Cfor 10 min. Genotyping was performed with single strand conformation polymorphism method with 13% polyacrylamide gel electrophoresis. On genotyping, six samples were randomly selected from each genotype group for sequencing to determine the specific genotypes. The genotypes were deviated from the Hardy-Weinberg equilibrium (Val/Val = 163, Val/Met = 392, and Met/Met = 143), χ² = 10.755, p = 0.001.

Statistical analysis

The Hardy-Weinberg equilibrium was tested with the FINETTI software (Sasieni, 1997). Power analysis was performed with the G*Power 3.1 (Faul et al., 2007). The effects of the BDNF Val66Met on the psychological dispositions were examined with linear regression analysis, a method for exploring genetic association of complex traits (Eberly, 2007; Lourenco et al., 2011). Due to the mixture of the genetic models of the personalities (Campbell et al., 2009), the associations between the BDNF Val66Met and Dark Triad were examined with three models: the dominant model (Met carriers vs. Val/Val), codominant model (Met/Met vs. Val/Met vs. Val/Val), and recessive model (Met/Met vs. Val carriers). In the dominant model, the numbers of the Met allele range from 0 to 1 (Val/Val = 0, Met/Val & Met/Met = 1); in the codominant model, the numbers of the Met allele range from 0 to 2 (Val/Val = 0, Met/Val = 1, Met/Met = 2); in the recessive model, the numbers of the Met allele range from 0 to 1 (Val/Val & Met/Val = 0, Met/Met = 1). To test the robustness of the results, the subsamples of 600, 500, 400, and 300 were selected from total sample 20 000 times to estimate the 95% confidence intervals of regression analysis. The mediations were examined by bootstrapping 1000 times with the SPSS version of INDIRECT macro (http://www.afhayes.com; Preacher & Hayes, 2008).

Results

Correlations between the Dark Triad traits and empathy and aggression

The scores on Machiavellianism, psychopathy, and narcissism were significantly positively inter-correlated. Moreover, the score on Machiavellianism was significantly negatively correlated with the score on Empathic Concern; the score on psychopathy was significantly negatively correlated with the scores on Empathic Concern and Perspective Taking; the score on narcissism was not significantly correlated with the empathic subscales. Additionally, the subscale scores on the Dark Triad were significantly positively related to the score on the BPAQ (Table 1).

Table 1. Descriptive statistics of the subscales of the IRI, SD3, and BPAQ

		Correlation coefficient (r)
Subscales	Mean ± SD	1	2	3	4	5	6	7
1. Perspective Taking	19.36 ± 4.40
2. Fantasy	16.19 ± 4.71	0.104**
3. Empathic Concern	21.65 ± 3.80	0.228**	0.173**
4. Personal Distress	12.32 ± 4.77	−0.227**	0.146**	0.118**
5. Machiavellianism	25.34 ± 5.18	−0.036	0.017	−0.120**	0.034
6. Psychopathy	16.97 ± 5.90	−0.090*	−0.035	−0.117**	0.030	0.300**
7. Narcissism	24.82 ± 3.74	−0.012	−0.038	0.024	−0.074	0.314**	0.319**
8. BPAQ	68.75 ± 14.84	−0.273**	0.145**	−0.123**	0.309**	0.237**	0.212**	0.137**

Note: IRI- Interpersonal Reactivity Index, SD3- Short Dark Triad, N- number of items, * p < 0.05, ** p < 0.01

The relationships between the BDNF Val66Met and the Dark Triad traits

Linear regression analyses indicated that the BDNF Val66Met was not significantly associated with the scores on the Dark Triad in the dominant model (Table 2). By contrast, this polymorphism in the codominant model was significantly associated with the scores on the subscales of Machiavellianism, and narcissism, but not with psychopathy. The number of Met alleles was significantly positively related to the scores on the Machiavellianism and narcissism (Table 2). Similarly, this polymorphism in the recessive model significantly predicted the scores on Machiavellianism and narcissism, but not psychopathy (Table 2). Compared with the Val carriers, the Met/Met genotype was related to higher scores on Machiavellianism, t (696) = 2.981, p= 0.003, and narcissism, t (696) = 3.145, p= 0.002 (Table 3). Bootstrapping on subsamples indicated the effects of the BDNF Val66Met on Machiavellianism and narcissism were stable (Table S 1 of the Supplementary Materials). Moreover, hierarchical regression analysis showed that the genetic effects were significant when we controlled sex, mental distress, parental styles, childhood trauma, and family relationship: Machiavellianism, β = 0.112, adjusted R² = 0.040, t (697) = 2.999, p = 0.003, and narcissism, β = 0.113, adjusted R² = 0.032, t (697) = 3.035, p = 0.002 (Table S 2 and Table S 3 of Supplementary Materials).

Table 2. The relationships between the BDNF Val66Met and the scores on the IRI, SD3, and BPAQ

	Dominant model				Codominant model				Recessive model
	β	R^2	t (697)	p	β	R^2	t (697)	p	β	R^2	t (697)	p
Dark Triad
Machiavellianism	0.062	0.004	1.651	0.099	0.108	0.012	2.879	0.004	0.112	0.013	2.981	0.003
Psychopathy	−0.003	<0.001	−1.089	0.929	−0.025	0.001	−0.672	0.501	−0.038	0.001	−1.009	0.313
Narcissism	0.029	0.001	0.769	0.442	0.091	0.008	2.407	0.016	0.118	0.014	3.145	0.002
Empathy
Fantasy	−0.019	<0.001	−0.506	0.613	−0.052	0.003	−1.367	0.172	−0.065	0.004	−1.710	0.088
Empathic Concern	0.006	<0.001	1.167	0.868	−0.070	0.005	−1.855	0.064	−0.122	0.015	−3.232	0.001
Perspective Taking	0.004	<0.001	0.114	0.909	−0.019	<0.001	−0.496	0.620	−0.035	0.001	−0.932	0.351
Personal Distress	−0.002	<0.001	−0.064	0.949	−0.004	<0.001	−0.106	0.916	−0.004	<0.001	−0.107	0.915
Aggression
BPAQ	0.048	0.002	1.270	0.204	0.077	0.006	2.030	0.043	0.075	0.006	1.993	0.047

Table 3. The scores (Mean ± SD) on the Dark Triad, empathy, and aggression of each genotype group

	Dominant model	Codominant model			Recessive model
	Met carriers	Met/Met	Val/Met	Val/Val	Val carriers
Dark Triad
Machiavellianism	25.25 ± 5.34	26.49 ± 5.44	25.17 ± 5.26	24.76 ± 4.78	25.05 ± 5.07
Psychopathy	16.96 ± 5.85	16.53 ± 5.64	17.12 ± 5.93	17.01 ± 6.07	17.09 ± 5.97
Narcissism	24.88 ± 3.88	25.69 ± 4.74	24.58 ± 3.48	24.62 ± 3.23	24.59 ± 3.41
Empathy
Fantasy	16.16 ± 4.82	15.95 ± 5.09	16.34 ± 4.71	16.35 ± 4.33	16.34 ± 4.60
Empathic Concern	21.66 ± 3.87	20.74 ± 4.44	22.00 ± 3.58	21.61 ± 3.58	21.88 ± 3.58
Perspective Taking	19.37 ± 4.42	19.06 ± 4.32	19.48 ± 4.46	19.33 ± 4.03	19.44 ± 4.33
Personal Distress	12.32 ± 4.80	12.29 ± 5.05	12.33 ± 4.72	12.35 ± 4.66	12.33 ± 4.70
Aggression
BPAQ	69.14 ± 15.01	70.95 ± 16.72	68.48 ± 14.31	67.46 ± 14.22	68.18 ± 14.28

Note: The score of the Val/Val group in the dominant model is as same as the Val/Val group in the codominant model, and the score of the Met/Met group in the recessive model is as same as the Met/Met group in the codominant model

The relationship between the BDNF Val66Met and empathy

In the dominant model and codominant model, the BDNF Val66Met was not significantly related to the subscale scores of IRI (Table 2). By contrast, in the recessive model, this polymorphism was significantly associated with the score on Empathic Concern (Table 2). The Met/Met genotype was related to lower scores on Empathic Concern than the Val carriers (Table 3). The bootstrapping test indicated that the genetic effect was stable (Table S 1 of Supplementary Materials). Considering the significant correlations between the Empathic Concern and growth environments (Table S 4), we examined the association between the BDNF Val66Met and empathic concern when the variables were controlled. The regression analysis indicated that the association was still significant when we controlled the variables (Table S 5).

The relationship between the BDNF Val66Met and aggression

The BDNF Val66Met was not significantly associated with the score on the BPAQ in the dominant model (Table 2). By contrast, this polymorphism was significantly associated with the score on the BPAQ both in the codominant model and recessive model. The number of Met alleles was significantly positively related to the score. The bootstrapping on the analysis with subsamples also indicated that the relationship between this polymorphism and the score on the BPAQ was stable (Table S 1 of Supplementary Materials). Considering correlations between aggression and variables of childhood trauma, and parental overprotection (Table S 6), we examined the association between the BDNF Val66Met and the score on the BPAQ when the variables were controlled. The analyses indicated that the genetic effect was still significant, β = 0.078, adjusted R² = 0.179, t = 2.256, p = 0.024 (Table S 7).

Mediations of empathy and aggression

For the potential mediations of empathic concern and aggression in associations between the BDNF Val66Met and the Dark Triad personalities (Figure 1), a multiple mediation indicated significant mediating effects of empathic concern, the mediating effect estimate = 0.1290, SE = 0.0862, 95% bias corrected confidence interval = [0.0075, 0.3627], and aggression, the mediating effect estimate = 0.2127, SE = 0.1178, 95% bias-corrected confidence interval = [0.0120, 0.4942]. Similarly, the association was significantly mediated by aggression (Figure 2), the mediating effect estimate = 0.0944, SE = 0.0572, 95% bias-corrected confidence interval = [0.0084, 0.2455], but not by empathic concern, the mediating effect estimate = −0.0613, SE = 0.0503, 95% bias-corrected confidence interval = [−0.2199, 0.0044].

Figure 1. Mediations of empathic concern and aggression in the association between the BDNF Val66Met and Machiavellianism. Coefficients are derived from the equations: Y = cX + e1, M1 = a1X + e2, M2 = a2X + e3, Y = c’X + b1M1 + b2M2 + e4. Y: Machiavellianism; X: the genotype of the BDNF Val66Met (0 = Val/Val = 0, Met/Val &Met/Met = 1); M1: empathic concern, and M2: aggression; * p < 0.05 and **, p < 0.01

Figure 2. Aggression, but empathic concern, mediates the association between the BDNF Val66Met and narcissism. Coefficients are derived from the equations: Y = cX + e1, M1 = a1X + e2, M2 = a2X + e3, Y = c’X + b1M1 + b2M2 + e4. Y: narcissism; X: genotypes (0 = Val/Val = 0, Met/Val &Met/Met = 1); M1: empathic concern, M2: aggression; *p < 0.05, and **p < 0.01

Discussion

This study reveals that the BDNF Val66Met is associated with the Dark Triad. The Met/Met genotype is related to higher Machiavellianism and narcissism than Val carriers. The association between the BDNF Val66Met and Machiavellianism is mediated by empathic concern and aggression, and the association between this polymorphism and narcissism is only mediated by aggression. Given that the Dark Triad is notable for criminal versatility and violent crime (Edwards et al., 2017), these findings provide new insights into the biopsychological foundations of aversive features of human personalities.

Associations between the BDNF Val66Met and the Dark Triad

We identify the BDNF Val66Met as a specific gene underlying the traits of Machiavellianism and narcissism, but not psychopathy. The findings suggest that, at least in part, the genetic foundations of Machiavellianism overlap with narcissism, while the genetic foundations of Machiavellianism disassociate from psychopathy. Considering that psychopathy has a moderate to large heritability (Campbell et al., 2009), we infer that a cluster of candidate genes underpin the genetic foundation of psychopathy, except for the BDNF Val66Met. Moreover, the BDNF Val66Met is associated with the functions of prefrontal cortex (Kim et al., 2013; Wang et al., 2014). The reduced brain functions of prefrontal cortex are found in individuals with high Machiavellianism and narcissism (Cohen-Zimerman et al., 2017; Mao et al., 2016), while the decreased activations of posterior medial frontal cortex were found in females scoring highly on psychopath (Overgaauw et al., 2020). Therefore, the brain functions possibly mediate the associations between this polymorphism and the Dark Triad.

A previous view believed that Machiavellianism is a feature of psychopathy, in which psychopathy is described as a broad construct such as manipulation, callousness, impulsion and risk-taking (Glenn & Sellbom, 2015). However, most studies tend to treat Machiavellianism and psychopathy as two distinct constructs (Jones & Paulhus, 2014; Paulhus & Williams, 2002; Zhang et al., 2020). Specifically, Machiavellianism indexes the psychological properties of manipulative tactics and cynical worldview (e.g., cynicism, coalition building, and preserving one’s own reputation), while psychopathy taps one’s antisocial behaviours and erratic life style (Jones & Paulhus, 2014; Zhang et al., 2020). Given that the BDNF Val66Met specifically underpins the trait of Machiavellianism, but not psychopathy, the findings suggest that destructive behaviours and manipulative interpersonal styles have different genetic foundations.

Mediations of empathic concern in the association between the BDNF Val66Met and Machiavellianism

A previous study indicates that the Met carrier of BDNF Val66Met is related to higher fantasy, but less perspective taking (Taschereaudumouchel et al., 2016). By contrast, this study reveals that the Met/Met genotype is associated with lower empathic concern than Val carriers. Considering that this study was conducted in a larger sample, we think our findings may be more robust. Moreover, this study reveals empathic concern mediates the association between the BDNF Val66Met and Machiavellianism. Individuals with the Met/Met genotype show less empathic concern for others’ misfortune. These findings suggest a link between empathy and the Dark Triad at the genetic level and provide a new perspective on the development of aversive features of human personalities.

Mediations of aggression in the associations of the BDNF Val66Met and Machiavellianism and narcissism

Consistent with the findings that the Met/Met of the BDNF Val66Met is related to higher aggression than the Val allele (Kretschmer et al., 2014; Spalletta et al., 2010), we found that the Met allele is related to higher aggression. These findings confirm the roles of the BDNF Val66Met in aggression. The Dark Triad is characterized by a high aggression (Jones & Neria, 2015). Individuals who score highly on Machiavellianism tend to behave in an aggressive manner to get what they want (Jones & Neria, 2015), and narcissistic individuals are more likely to lash out at others who threaten their views (Bushman & Baumeister, 1998). Similarly, this study shows that Machiavellianism and narcissism are positively related to aggression. More interestingly, the associations between the BDNF Val66Met and Machiavellianism and narcissism are mediated by aggression, which reveals a pathway where the BDNF affects the Dark Triad.

The influences of growth environments and the BDNF Val66Met

Consistent with previous studies indicating that growth environments influence the Dark Triad (Birkas et al., 2018; Guo & Feng, 2017; Jonason et al., 2014), this study indicates that individuals who experience more childhood trauma show higher Machiavellianism, and vice versa, ones who have better family relationships and parental care show lower Machiavellianism. More interestingly, the significant association between the BDNF Val66Met and the Dark Triad remains significant when we controlled such variables. These findings further confirm that the Dark Triad is influenced both by growth environments and genetic predispositions.

Limitations and implications

Several limitations should be noticed. First, this study was performed in Chinese college students. Given that a study on country-level correlates of the Dark Triad indicates that narcissism and Machiavellian are sensitive to the variables of gender inequality and modern democracies (Jonason et al., 2020), the interactions between genes and cultural environments in the Dark Triad should be investigated in future. Second, some studies showed that the Met allele was related to a reduced serum BDNF level as compared with the Val/Val (Cattaneo et al., 2010; Colle et al., 2017), while a few studies reported an opposite pattern (Moreira et al., 2015; Ryan et al., 2018). The mixed findings impaired the explanations of the functions of BDNF in the Dark Triad.

The Dark Triad traits have undesirable and antisocial implications in all nations. Exploring specific genes of the Dark Traid would promote our understanding of the biological basis of aversive features of human personalities. In this study, we identify that the Met/Met genotype of the BDNF Val66Met is related to high Dark Triad personalities, which highlights the roles of the BDNF gene in psychobiological foundations of immoral behaviours. Given that the genetic effects are stable after we controlled the effects of growth environments (Jonason et al., 2020), the findings could possible extend to other populations such as Australians and East Asians. Of note, the BDNF Val66Met only contributes to a small part of the individual differences in the Dark Triad, and the larger part of the variances in the personalities are influenced one’s growth environments (Campbell et al., 2009; Julie & Daniel, 2020). The individuals with the high risk alleles of the personalities could also show low dark triad traits if they have better family relationships.

Conclusions

This study indicates that the BDNF Val66Met is associated with the Dark Triad. The association between this polymorphism and Machiavellianism is mediated by empathic concern and aggression, while the association between this polymorphism and narcissism is only mediated by aggression.

Author contributors

All authors had full access to all the data. Performed the experiment: Rui Quan, Wenxuan Guo, Linlin He, Quanhe Wang, Rui Zhang, Yuhe Fan, Mengfei Zhang. Analyzed the data: Pingyuan Gong and Quan Rui. Wrote the paper: Pingyuan Gong, Rui Quan, and Ruliang Pan. Designed the study: Pingyuan Gong.

Acknowledgments

We thank Mr. Guomin Shen, Shoumin Xi, and Miss Yong Huang for their assistance.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplemental data for this article can be accessed https://doi.org/10.1080/00049530.2021.1898916.

Additional information

Funding

This study was supported by the National Social Science Fund of China [20BSH165] and the Ministry of Education of Humanities and Social Science Foundation [19YJAZH021] to Pingyuan Gong.