*Most studies investigating risky decision-making in humans and animals typically focus on events with probabilities exceeding 10%. Our  study aims to investigate how we react to events that are both rare (probabilities around 1%) and extreme in their consequences (positive, Jackpot JP - or negative, Black Swan BS). In a four-armed bandit task, participants were exposed to stochastic gains and losses, including exceptionally rare and extreme events (REE). Our prior research involving rats highlighted their sensitivity to REE, with avoidance of extreme losses (BS) and attraction to extreme gains (JP). In this study, we introduce two tasks for humans, differing in their entry method: one through training and the other through a partial description. Preliminary findings on pilot participants suggest that the introduction of REE led participants to deviate from the traditional stochastic dominance of first order. Participants employed similar strategies or reactions to cope with REEs in the two tasks. They avoided BS and seek JP, with individual differences in the strength of these strategies. Individuals that exhibited differences in decision-making phenotypes and sensitivity had also different exposures and reactions to rare and extreme events.

**Imperfect take-up of social benefits has long been a major policy concern for modern welfare states. Few interventions have achieved a sizeable increase in the take-up rate. In this paper, I study the effects of a large and salient welfare reform of the French EITC (“prime d’activité”). Using exhaustive and granular administrative data, I show that the reform led to a remarkably large take-up response, resulting in a 50% cut of the non-take-up rate.  Leveraging the heterogeneous exposure to the benefit increase, I find that raising the net benefit of claiming had limited effects, with an estimated take-up elasticity of about 0.1. Instead, I provide evidence indicating that information frictions play a significant role, with the majority of the surge in take-up attributed to heightened awareness and visibility of the program following the reform's announcement. Additionally, to examine the targeting effects, I analyze the characteristics of the marginal enrollee. Drawing on the empirical findings, I develop a theoretical framework to assess the welfare implications of the reform and discuss how accounting for imperfect take-up affects the optimal design of welfare programs.