In this doctoral thesis, I present four studies aim at investigating the role of affective and emotional reactions in shaping people’s risk-taking behaviors in conditions of uncertainty. In the first two studies, I use the Iowa Gambling Task (IGT) in the attempt to mimic real-life conditions of uncertainty in a laboratory setting. In Study 1 (Chapter 3), I manipulate IGT decks by associating the hearing of a highly unpleasant sound to the moment of selection from Bad (Congruent condition) or Good decks (Incongruent condition) to make participants experience a negative affective reaction towards them. Drawing on the affect heuristic and the SMH, I expect that this unpleasant reaction will lead participants to avoid the manipulated decks, thus having a detrimental effect when it is associated with Good decks (i.e., lower selections from the long-term advantageous Good decks) and a beneficial effect when it is associated with the Bad ones instead (i.e., lower selections from the long-term disadvantageous Bad decks). In Study 2 (Chapter 4), I replicate Study 1’s design to detect a similar effect while using an emotional reaction of disgust, induced by a disgust-eliciting image, in line with assumptions on the role of discrete, incidental emotions in decision-making. In both Study 1 and Study 2, I also investigate the generation of somatic markers in terms of enhanced autonomic activation measuring anticipatory SCR as in classical IGT studies. Moreover, in the attempt to reach a deeper understanding of this mechanism, I also include measures of heart rate (Study 1 and Study 2) and pupil dilation (Study 2) as suggested in previous studies presented in the first chapter. In the last two studies instead, I investigate the effect of emotional reactions on health-related risk-taking behaviors during the COVID-19 pandemic. In Study 3 (Chapter 5), I explore how the frame used in the media (positive frame: number of recovered vs negative frame: number of dead) to communicate information about the pandemic and the comparison between the COVID-19 and the seasonal flu can influence citizens compliance with self-protective behaviors through the mediation of emotional reaction (i.e., worry) and risk perception as suggested by the affect heuristic and the risk as feelings framework. Moreover, I also compare results from three different European samples (Italian vs Austrian vs English). Last, in Study 4 (Chapter 6), I investigate the effect of six information formats (five numerical and one verbal) commonly used in the media to report COVID-19 mortality rates on citizens' emotional reactions, risk perception and intention to comply with recommended self-protective behaviors against the virus.