This study was aimed at testing the prediction about the correlation between understanding of verbal comic texts (jokes) and level of gelotophobia in individuals with different levels of psychometric intelligence (IQ). The object of study was the process of verbal humor understanding in subjects with different levels of IQ. At the first stage of the study, the J. Raven' Standard Progressive Matrices test was administered to participants to (N = 69; 45 women; 18-55 years; M = 31.8; SD = 12.6) in order to control their level of intelligence. According to the results, a number of subjects (N = 40, 25 women; 18-55 years old; M = 34; SD = 11) were selected to participate in the second stage of the study. These subjects were divided into 2 groups (with lower and higher IQ levels) in according to the median (M = 122; SD = 5). Next, the subjects were administered a set of 7 verbal jokes of various levels of complexity (Shcherbakova, 2009) that were a starting point for deep semi-structured interviews. Two independent experts used interview protocols to assess the levels of cognitive and emotional understanding of each joke. The level of participants' gelotophobia was assessed using the PhoPhiKat <30> questionnaire (Ivanova et al., 2016). To test the consistency of the experts' scores, the Kendell coefficient of concordance was used, and the Pearson's correlation coefficient was used to test the main hypothesis. It was shown that the magnitude of the correlation between the level of gelotophobia and the levels understanding of verbal humor did not differ in groups of subjects with different IQ levels. It was revealed that difficulties with understanding of verbal humor in young adults.