The ability to make inferences about our own and other’s people mind is called theory of mind. Theory of mind consists of affective and cognitive component. The affective component encompasses processing of other emotional expressions while cognitive involves the attribution of cognitive mental states to others. The most prominent instrument for measuring the affective theory of mind is RMET (Reading the mind in the eyes test, Baron-Cohen, et.al, 2001). Deception detection can be considered as a component of theory of mind since one should be able to understand how others can create false beliefs in order to deceive others. Signs of deception should become more transparent in high-risk situations where an individual has a lot to lose if he is caught. How succesfull individuals will be in detecting others mental states and signs of deception, depends on their personality and levels of emotional intelligence. Therefore, the aim of this research was to examine the contribution of the theory of mind, psychopathy, and emotional intelligence in explaining the variability of deception detection. Including RMET (BaronCohen, et.al., 2001), Levenson's self-report psychopathy scale (Levenson et.al., 1995), 12 videos based on high-risk situations (ten Brinke et.al.,2014), and the Test of emotion analysis (Kulenović et.al., 2000) were used. The results showed that none of the variables were significantly correlated with the deception detection while subjects were not better than the level of chance in the detection of truth or lies. Those with more pronounced features of malicious and non-emotional psychopathic traits have achieved poorer results in recognizing other affective mental states and were less successful in understanding emotional blends. On the other hand, those with higher levels of antisociality traits (characterised by anxiety and negative emotionallity) were better in mentalizing. Of all the variables used in the research, emotional intelligence has proven to be the best predictor of the affective theory of mind.