This project concerns the determinants of the propensity to adopt and follow the recommendations for prevention and containment in the face of the spread of Covid-19. The fundamental assumption is that this propensity is determined by the personal characteristics of the individual: one the one hand, by his risk and time preferences and his self-control, and on the one hand, by his social preferences (altruism, generosity, trust, cooperativeness). The main objective of the project is to identify the effect of these behavioral dimensions on the observance of containment measures and the adoption of barrier gestures. Their knowledge is a prerequisite for designing more effective non-coercive measures, such as monetary and non-monetary incentives (nudges), to better target communication during and after the crisis, and to increase its impact on behavior. To achieve these objectives, the project will combine several experimental economics' tools that will allow us to precisely measure various behavioral dimensions (risk aversion, impatience, altruism, trust, etc.) based on incentivized tasks. Some of the measures of behavioral dimensions, e.g. risk preferences, will be doubled by declarative and genetic measures. This precaution is intended to obtain converging evidence of the robustness of the main determinants. Finally, we will apply a discrete choice method to highlight the trade-offs that individuals agree to make to accept the restrictive measures and will test a “nudge” to encourage them to do so. Our results will be used to determine the right levers for effective communication with target audiences, with a view, for example, to adopting the observance of barrier gestures. The discrete choice experiment will provide an understanding of the trade-offs to which individuals agree to accept the measures. Finally, the processing variable "nudge" will identify the relevance of this type of measure under extreme conditions. Our work is also of longer-term interest for knowledge, thanks to the genetic block. The objective is to establish a link between the genetic profile and the individuals who transgress the instructions. If there is a pattern of transgression, it will be difficult to devise levers for action for this segment of the population, which would, therefore, represent an uncontrollable risk. Because disobedience is similar to risk-taking, it is likely that some of the SNPs that affect risk tolerance also affect obedience.