The Relationship Between the Dark Triad Personality Traits and Decision Making

Author: Sijia (Scarlett) Dong
Foxcroft School
May, 2020


Decision making has always been an important role in human life, especially in economics. In the market of oligopoly competition, a few companies set the price interdependent from each other, competing using the game theory. Each decision they make on the price determines their gains and losses. In economics, people serve as rational agents who make decisions that maximize their utility; however, this is not normally the case. Individuals’ decisions are affected by various factors (e.g. personality, educational background and also psychological state characteristics such as mood), and people do not always choose the profit-maximizing decision. The present study examines the relationship between personality and economic decision making. 

Personality is closely associated with behaviors, cognition, and emotions. Assessing personality can be used to predict behavior tendencies. One way to measure personality is by using the Big Five personality test. The Big Five personality traits includes conscientiousness (disorganized / impulse vs. careful / disciplined), agreeableness (ruthless / suspicious vs. soft-hearting / trusting), neuroticism (calm / secure / self-satisfied vs. anxious / insecure / self-pitying), openness (practical / conforming vs. imaginative / independent), and extraversion (retiring / reserved vs. sociable / affectionate) (McCrae & Costa, 2008). The Big Five personality factors are relatively stable in adulthood and appear to be present in every culture. It also clearly depicts the whole picture of individual’s traits and as a result, studies mostly apply the Big Five personality test for predicting behaviors. 

The present study employed the Dark Triad personality measure as a form of personality assessment. The Dark Triad is a measure of malevolent personality developed by Paulhus and Williams (2002). People who score high on the Dark triad are more likely to commit crime and cause social distress. The three dimensions of the dark triad are Machiavellianism, Narcissism, and Psychopathy. Narcissism stands for strong ego, greediness, and a sense of superiority and dominance. Machiavellianism stands for the manipulative, calculating, and amoral personality and people with high levels of Machiavellianism typically focus on self-interest and self-gain. Psychopathy, on the other hand, is marked by cruelness, impulsivity, and anti-social behaviors. The Dark triad, in general, is related to antisocial behaviors, such as aggression and violence, and emotional deficit, such as a lack of empathy and theory of mind (Paulhus & Williams, 2002). It is also related to poor well-being such as loneliness and anxiety. People who score high on the Dark triad are more likely to score low on agreeableness in the Big 5 personality. 

Personality has been linked with cognitive functioning, such as attention, reaction time, but also decision making. One study found that personality measured using the Big 5 questionnaire influences decision-making during the Ultimatum Game, which measures the willingness to share (Fiori et al., 2013). In the Ultimatum Game, the first participant was asked to divide 10 dollars with the second participant. If the second participant rejects, then neither can get the money. According to the theory of rational agent, the proposer will always offer the smallest amount, and the receiver will always accept. However, the experiment findings do not agree with the prediction. It was found that participants who scored higher on conscientiousness and lower on extroversion were more likely to accept the offer in the game (Fiori et al., 2013). Individuals with high neuroticism seem to have decreased willingness to take risk. In another study, participants were asked to choose between a lottery and a certain amount, six times. The lottery is always the same but the fixed payment varies in each choice. The task was repeated four times. For the first two times, the lottery is positive, and the lottery can incur a loss of $1 for the third time and $5 for the last time. The researchers would record the number of times participants choose the lottery over a fixed payment and associated these with the person’s personality (Anderson et al., 2011). In the experiment of sequential prisoner’s Dilemma, two players each had $5, and the first player decided whether to send $5 or $0 to the second player. The amount would double after the transfer. Then the second player decides to transfer back a number between $0 and $5. The researchers recorded the money transfers and found that agreeableness increases the fraction of first moves transferring $5 and increases also the average amount transferring back. Cognitive skills and IQ also influenced the result (Anderson et al., 2011). Another study found that individuals with Dark Triad personality traits were more likely to engage in opportunistic decision-making, which is the actions assessed toward an immediate circumstance in favor of the individual or company (D’Souza & Lima, 2015). In business, people with Narcissism are visionary; they seek superior position to influence over others and make decisions that are beneficial to their reputation. On the other hand, narcissistic people are also innovative with prominent leadership with the ability to lead their company to glory. Evidence has shown that most managers and people in executive positions in a corporation score low in Psychopathy; they are not self-centered, cruel, or cynical, but can contribute to the longevity and gains of the companies. People with high levels of Machiavellianism seek management and leadership positions, which allow them to manage, control and manipulate people inferior to them. They focus on their personal power and can easily influence others (D’Souza & Lima, 2015). 

Most previous research has been done on the Big 5 personality measure in relation to decision-making, including willingness to share and take risks and the game theory, while the relationship between the Dark Triad and various decision-making games has not been assessed in detail. Moreover, previous research mostly focuses on adults rather than younger population groups. The present study aims to explore the relationship between the Dark Triad personality and economic decision-making, specifically, the wiliness to take risk, in a female sample of high school students. It was hypothesised that the Dark Triad personality traits would be associated with high school students’ decision in risks-taking contexts. 


Participants and sampling

Participants in the present study were female students from an all-girl high school in Virginia (US), with the age range from 14 to 20. Convenience sampling method was used: the questionnaire was sent to the school email, so every student in the high school had access to the questionnaire and could fill it out voluntarily. It was attempted to obtain a homogenous sample, with all participants being females with similar age range and similar education level. 


This study was based on a non-experimental cross-sectional design. The dependent variable was decision-making/ risks-taking and the independent variable was personality based on the Dark Triad. 


Personality was measured by the short Dark Triad questionnaire (Jones & Paulhus, 2014). It contains 27 questions: 9 questions measured Machiavellianism (e.g. “I like to use clever manipulation to get my way”), 9 questions measured narcissism (e.g. “People see me as a natural leader”), and 9 questions measured psychopathy (e.g. “I like to get revenge on authorities”). Each item was rated on a 5-point Likert-type scale (strongly disagree-strongly agree; Jones & Paulhus, 2014). Sum scores for each of the sub-scales and the overall global scales were computed and higher scores indicated higher attributes of the respective trait.

Decision-making was measured by Iowa Gambling Task on the online simulation on Psytoolkit (Stoet, 2010; Stoet, 2017). Participants would be given $2,000 at the start and asked to choose from one of the four card decks (A, B, C, and D) for 50 trials. Deck A and B yield $100 each time and deck C and D yield $50. Each time a participant chooses a deck of card, there is a 50% chance to get a penalty. The penalty is $250 for choosing deck A and/or B and $50 for choosing deck C and/or D. Participants need to gain as much money as possible. Without knowing how much each deck of card yields and the amount of penalty it incurs, participants would have to make a decision each time. The online experiment recorded participants’ reaction times (used to determine cognitive skills), their choice, whether there was a penalty and money they gained and lost. This study focused on how many times participants chose deck A/B versus deck C/D to determine the willingness to take risk: the more deck A/B was chosen, the more willing the individual to take risk. An index was calculated to determine decision-making based on A/B choices divided by C/D choices and values >1 indicate risk taking and <1 indicate risk aversion.

Further, math ability was assessed with rating from 1 to 5 (very low – very high); math liking was assessed as 3 ratings (Yes/No/Neutral). Students reported whether they have studied Psychology or Economics (Yes/No) and also gave their mood ratings on a 3-scale (Low or negative mood/Neutral mood/High or positive mood).

Statistical analyses

All data analyses were performed on SPSS (IBM, version 25). Normality tests were performed on the outcome measure and revealed that decision-making was highly positively skewed (Skewness: 4.00 and Kurtosis: 17.70) and the Shapiro-Wilk test (p<.001) suggested violation of normality. Log-transformation (ln method) was applied and normalised the distribution (Skewness: 0.03 and Kurtosis: 2.70; Shapiro-Wilk test, p=.080) and hence the log-transformed data was used with parametric testing. Robustness checks were conducted prior to analyses. To explore whether individuals who dropped out prior to the DM task differed in personality from individuals who completed the survey three independent-sample t-tests were conducted. Findings showed that the three sub-scales Machiavellianism, Narcissism and Psychopathy were not statistically significantly different between the non-completers and the completers (all p’s >.410), suggesting that the final sample seems to be representative of the whole sample that took part. 



Table 1 depicts the characteristics of the sample. The sample was aged on average 16.12 years (SD=1.30). The average maths ability was 3.31 (SD=0.84). Most students had not studied Economics (80.8%) or Psychology (80.8%). 

From the personality traits, the average score on Machiavellianism was 3.27 (SD=0.59); the average score on Narcissism was 2.77 (SD=0.60) and the average score on psychopathy was 2.15 (SD=0.70). The global score of the three traits was 8.19 (SD=0.87). 

The mean score of decision-making after applying log-transformation was 0.09 (SD=0.87), suggesting that participants on average were more risk-taking.

The effect of personality on decision making

Bivariate Pearson’s correlation analyses were conducted and showed that Machiavellianism was not correlated with DM, r=.217, p=.286. Narcissism was also not related to DM, r=-.214, p=.293. Psychopathy was also not related to DM, r=.082, p=.690. Further, the global score was also not related to DM, r=.035, p=.865. This suggests that the three personality traits were not related to DM. The other study variables, i.e. maths levels, maths liking and mood did not associate with DM either (all p’s >.392).

A multiple linear regression model was run to explore whether there is an effect of personality on DM, when controlling for the other potential confounding variables (i.e. maths levels, maths liking and mood) and the findings revealed that none of the variables predicted DM, F(6,19)=.507, p=.796. 

Table 1. Study variables for the full sample.

VariableFull sample (N=26) M(SD)/N(%)
Age16.12 (1.30)
Maths ability (1-5)3.31 (0.84)
Maths liking Yes No Neutral
12 (46.2) 8 (30.8) 6 (23.1)
Economics study No Yes
21 (80.8) 5 (19.2)
Psychology study No Yes
21 (80.8) 5 (19.2)
Mood Low/negative mood Neutral High/positive mood
4 (15.4) 20 (76.9) 2 (7.7)
Personality Machiavellianism Narcissism Psychopathy Global score
3.27 (0.59) 2.77 (0.60) 2.15 (0.70) 8.19 (1.42)
Decision making (ln)0.09 (0.87)


The present study aimed to explore the relationship between personality and decision-making in the context of risk taking. The study employed young females as a sample. It was hypothesised that Dark Triad personality traits (Machiavellianism, Narcissism, and Psychopathy) would influence decision-making on risk taking. Decision making was measured with the Iowa Gambling task which assesses the risk taking vs the risk aversive behaviour in terms of gaining financial rewards and paying penalties for wrong decisions. The findings revealed that there were no relationships between Dark Triad personality traits and risky decision-making in the present sample. 

The results are different from previous findings, which have consistently found that the Dark Triad personality traits are influencing decision-making. One previous study indicates that people with the Dark Triad personality traits (indicating higher levels of Machiavellianism, higher levels of Narcissism, and higher levels of Psychopathy) are more likely to make decisions that are in favour of themselves (D’Souza & Lima, 2015). However, the present study shows no relationship between these personality traits and decisions that make them better off (i.e. gaining more money and paying lower penalties). One explanation for the null findings is that the sample had relatively low scores within each of the personality sub-scales and low variability in the sample in those personality traits reduces statistical power and hence impedes potential findings. 

Previous studies focus on adults, while the present study has been conducted using high school students, a sample in which potentially financial risk taking behaviour is rather low (as they experience these risk taking behaviour in daily life to a smaller extent). A previous study had shown that personality influences acceptors’ decisions in the Ultimate Game but not proposers’ decisions, and more honest people gained more and more introverted people accepted more often. The study was conducted on a sample aged 18 – 44 (Fiori et al., 2013), suggesting that these relationships might occur in a wider age range, including adults but not in younger samples. One study of 91 undergraduate students (aged 18-28) from Ohio University consisting of 90% Caucasian explored the Iowa Decision-making Task and found a relationship between personality and deck selection, however, personality traits and mood have been shown to impact the performance of the participants (Buelow & Suhr, 2013). The BIS/BAS scale was used for measuring personality and also mood measurements were implemented. The results show that participants with high level of drive and impulsivity tend to focus on short-run gain by choosing more from deck A and B. It also found that individuals with high negative mood selected more Deck B than Deck C (Buelow & Suhr, 2013). However, the present study showed no impact of mood on decision making, which might attribute to the low variability in mood. 

One limitation of the study was the sample. The sample was small and only included female students. If the study included both females and males, the result might have been different. The sampling was done in only one high school due to inaccessibility to other high schools in other locations. This led to sample bias as the sample of female students could not represent the whole population of this age group. Moreover, the sampling method might have led to a sample bias; the questionnaire was sent through the school email, and might have neglected those who do not frequently check their email. 

Second, the presentation of the Iowa Decision Making Task in the questionnaire is a limitation of the study. The data collected showed that some students randomly choose the card decks (A, B, C, D) or choose them in a pattern, which might be attributed to their misunderstanding of the Iowa Decision Making Task or to their impatience in the whole study data collection period. The study was observational and could not control for many confounding variables including the participants’ environment that might influence their performance in the task. Some students might perceive the whole decision-making game as too long so they randomly choose the deck instead of trying to gain more money. The introduction of the game might not be conspicuous or clear enough for some students to pay close attention to and understand the game. The Iowa Decision Making Task was not the first choice for measuring decision-making and the initial intention was to measure economic decision-making. However, due to inaccessibility of online simulation for other economic games and time limitation, the Iowa Decision Making Task was chosen. The task could not measure specifically economic decision making and is largely used to investigate participants’ cognitive skills and mental health (Bechara et al., 1994). The 50%-chance payment on deck A and B is $250 while money gained is $100 and payment on deck C and D is $50 while money gained is $50, so clearly the better way to ensure gaining money is to always choose deck C and D. Moreover, the participants were untold about the amount of payment, so they needed to figure this out by themselves from the first few trials they engaged in (Bechara et al., 1994). As a result, the Iowa Decision Making task might not be the best measurement for risk-taking in this context. One cross-section study has shown that as age increases, people are more likely to choose deck C and D to ensure money gained. Childred and adolescents are more likely to choose risky deck A and B (Buelow & Suhr, 2009). The present study agrees with these findings as the study showed that young participants are more willing to take risk. Some previous studies have used different versions of the Iowa Decision Making task specifically adapted to a younger adult sample, which might be more useful to explore their decision making behaviour (Buelow & Suhr, 2009).

Future research should aim to explore the Dark Triad personality traits in relation to economic decision-making. Economic decision-making would be measured for example with the prisoner’s dilemma or the Ultimatum game, which were tasks used in various previous research regarding decision-making, and these studies found correlations between the Big Five personality traits and economic decision-making (with e.g. higher levels of neuroticism relating to lower risk taking). If risk-taking is to be measured, the study could apply the lottery game utilized in the previous study (Anderson et al., 2011). The lottery game is designed to incur gains or losses in different trials, with the amount of incurred losses varying in different trials, which makes it a better for measuring risk-taking as it considers more aspects (Anderson et al., 2011), while the Iowa Decision Making Task only incurs fixed losses. The sample should include both female and male high school students in a larger range of locations and also a wider age width. The study could take place physically instead of providing questionnaires/ tasks online, in order to control some external factors that might influence participants’ performance. 

The study investigated the relationship between personality and decision making. The study was conducted in high school female students with the use of an online questionnaire and a decision making task. Personality was measured by the Dark Triad personality test by Paulhus and Williams (2002) and decision-making in a risk-taking context was measured with the online simulation Iowa Decision Making Task (Stoet, 2010; Stoet, 2017). The results showed no relationships between the Dark Triad personality traits (Machiavellianism, Narcissism, and Psychopathy) and decision making in female high school students, which provides no support for the present hypothesis that Dark Triad personality traits would influence decision-making. Other variables that were assessed in the present study included math-liking, math-ability, previous knowledge, and mood also no relationships emerged between these and decision-making. The study is subject to several limitations, which include the sample and the choice of the decision-making task. Future research should apply different economic games (e.g. the prisoner’s dilemma or the Ultimatum Game) and should be physically conducted on a larger sample in order to provide more reliable findings. 


Mentor: Dr. Bianca Serwinski, Northeastern University

Anderson, J., Burks, S., DeYoung, C., & Rustichini, A. (2011, January). Toward the integration of personality theory and decision theory in the explanation of economic behavior. In Unpublished manuscript. Presented at the IZA workshop: Cognitive and non-cognitive skills. 

Bechara, A., Damasio A.R., Damasio H., Anderson S.W. (1994). Insensitivity to future consequences following damage to human prefrontal cortex. Cognition, 50, 7-15.

Buelow, M. T., & Suhr, J. A. (2009). Construct validity of the Iowa gambling task. Neuropsychology review, 19(1), 102-114.

Buelow, M. T., & Suhr, J. A. (2013). Personality characteristics and state mood influence individual deck selections on the Iowa Gambling Task. Personality and Individual Differences, 54(5), 593-597. 

D’Souza, M., & Lima, G. A. S. F. D. (2015). The dark side of power: the dark triad in opportunistic decision-making. Advances in Scientific and Applied Accounting, São Paulo, 8(2), 135-156. 

Fiori, M., Lintas, A., Mesrobian, S., & Villa, A. E. (2013). Effect of emotion and personality on deviation from purely rational decision-making. In Decision making and imperfection (pp. 129-161). Springer, Berlin, Heidelberg. 

Jones, D. N., & Paulhus, D. L. (2014). Introducing the Short Dark Triad (SD3): A brief measure of dark personality traits. Assessment, 21, 28-41.

McCrae, R. R., & Costa, P. T., Jr. (2008). The Five-Factor Theory of personality. In O. P. John, R. W., Robins, & L. A. Pervin (eds.), Handbook of personality: Theory and research (3rd. ed.). New York: Guilford.

Stoet, G. (2010). PsyToolkit – A software package for programming psychological experiments using Linux. Behavior Research Methods, 42(4), 1096-1104.

Stoet, G. (2017). PsyToolkit: A novel web-based method for running online questionnaires and reaction-time experiments. Teaching of Psychology, 44(1), 24-31.

About the author

Sijia Dong