Research aims: The possibility of damaging factors such as risk aversion and past behavior when making investment choices is very high. Therefore, it is crucial to identify and measure risk aversion and the influence of people's past behavior in different conditions of certainty and uncertainty on investors' decisions in the market. The primary goal of this research, which has remained hidden from the perspective of many researchers, is to investigate the moderating effect of past behavior (experiences) on the relationship between risk aversion and investment choice among investors of the Tehran Stock Exchange.Design/Methodology/Approach: Considering the unlimited population size, Morgan's Table was used for sampling, and the sample size for the unlimited population according to Morgan's Table is 384 people. Therefore, 384 community members were selected as samples. The collected data were then analyzed through SPSS and SMART-PLS software, and research hypotheses were investigated using structural equation modeling.Research findings: The results obtained from the research demonstrated that the risk aversion factor yielded a positive and statistically significant effect on investment choice. In contrast, the factor of past behavior (experiences) generated a negative and statistically significant effect on the relationship between risk aversion and investment choice.Theoretical contribution/Originality: This study provides information on how to earn returns for investors and increase the efficiency of securities markets. In addition, there has been minimal research into this field in Iran.Practitioner/Policy implication: The results of this research provide for investors to pay attention to the impact of market efficiency and the optimal and desired use of available resources.Research limitation/Implication: Among the most important limitations, three things can be mentioned: (a) society's lack of familiarity with academic research, (b) failure to provide more variables, and (c) due to some characteristics of the respondents, generalizing the results to other populations must be approached with caution.