External seminars

We study the relationship between overconfidence and the political and financial behavior of a nationally representative sample. To do so, we introduce a new method of directly eliciting overconfidence of individuals that is simple to understand, quick to implement, and that captures respondents’ excess confidence in their own judgment. Our results show that, in line with theoretical predictions, an excessive degree of confidence in one’s judgment is correlated with lower portfolio diversification, larger stock-price forecasting errors, and more extreme political views. Additionally, we find that overconfidence is correlated with voting absenteeism. These results show that overconfidence is a bias that permeates several aspects of peoples’ lives.

Social information programs are widely used to nudge behavioural change. Their behavioral spillover beyond the targeted outcome and interaction with other nudges are often overlooked, leading to the possible mis-estimation of their cost-effectiveness and welfare impacts. Through an RCT with water clients of a multi-utility, we test the impact of a social information campaign to nudge water conservation. We exploit variability in whether customers also have gas and electricity contracts and receive other resources' reports, similar to the water one. We find that the water report significantly decreases water and electricity usage over the two post-treatment years, particularly for customers for whom the water report is the first one to be received. Gas and electricity customers receiving the program are significantly less likely to deactivate their contract.

Mindfulness-based meditation practices are becoming increasingly popular in the Western world, including in firms and in education. While the scientific literature has largely documented the benefits of mindfulness meditation on mental health, little is still known about potential spillovers of these practices on other important life outcomes, such as performance. We address this question through a field experiment in an educational setting.  We study the causal impact of mindfulness meditation on academic performance through a randomized evaluation of a well-known 8-weeks mindfulness meditation training delivered to university students on campus.  As expected, the intervention improves students’ mental health and non-cognitive skills. However, it takes time before students' performance can benefit from mindfulness meditation: we find that while the intervention significantly increases academic performance by about 0.4 s.d. in the long-run (ca. 6 months after the end of intervention), if anything it marginally decreases average grades in the short-run, i.e, during the exam period right after the end of the intervention. We discuss explanations for and implications of our results.

 From the growing popularity of authoritarian political leaders to the underrepresentation of women in boardrooms, leadership is an important theme in current human affairs. Leadership is also a prominent and widely studied research topic across the social, biological and cognitive sciences. However, these active literatures have evolved somewhat independently and there is an urgent need for synthesis. An evolutionary approach can integrate seemingly divergent perspectives first by suggesting that leadership and followership have evolved as adaptive strategies to solve to a range of group coordination challenges, from collective movement to internal peacekeeping and intergroup conflict. Some manifestations of leadership and followership are therefore phylogenetically old, and operate similarly across species. Second, these different challenges entail different risks for followers (e.g., fear of coordination vs. exploitation) and so they should therefore shape different mental models of ideal leader prototypes. Third, a review of the different literatures shows the importance of a distinction between two main leader prototypes, prestige and dominant leaders that have contrasting expressions, functions, histories, and neural-developmental pathways. The distinction may help to resolve various scientific puzzles such as: (i) contrasting philosophical views on the different traits and qualities of ideal leaders; (ii) the appeal of dominant leaders under threat; (iii) gendered perceptions of leadership, and (iv) the fundamental difference between leadership and management as two ways of organizing group activities.

We empirically assess whether hindsight bias has consequences on how citizens evaluate their political actors. Using an incentivized elicitation technique, we demonstrate that people systematically misremember their past policy preferences regarding how to best fight the Covid-19 pandemic. At the peak of the first wave in the United States, the average respondent mistakenly believes they supported significantly stricter restrictions at the onset of the first wave than they actually did. Exogenous variation in the extent of hindsight bias, induced through random assignment to survey structures, allows us to show that hindsight bias causally reduces trust in government.

Eruptions of anger and erratic mixing of cooperation and competition can make competitors unpredictable to outside observers and opponents. According to Rational Choice Theory (RCT), being unpredictability can be strategically advantageous. However, core assumptions underlying RCT have been questioned and cognitive science suggests that people have difficulty making themselves truly unpredictable. Here we examine the biobehavioral origins and consequences of unpredictability when aiming to outmaneuver one’s competitor, or to protect against being exploited. Meta-analyzing nine interactive contest experiments (N=650) shows that individuals are unpredictable especially during attack and this increases their probability of winning. In contrast to RCT, however, unpredictability emerges in part because individuals invest out-of-equilibrium, which is irrational from a payoff-maximizing perspective. Follow-up experiments (N=53 dyads) uncover that attacker (but not defender) unpredictability associates with elevated pre-contest testosterone and cardio-vascular stress reactivity. Being unpredictable originates in competitive arousal and increases the likelihood of winning at a cost to both victor and victim.

With the present paper, we challenge all too simplistic virtuous accounts of ethical leadership by showing that context matters. We specifically focus on preceding follower behavior and argue that such behavior critically qualifies how ethical subsequent leader behavior is perceived. Integrating scope of justice theory and accounts of moral emotions, we hypothesize that observers believe that disrespectfully (vs. respectfully) acting followers do (vs. do not) deserve disrespectful treatment. As a result, observers experience less (vs. more) anger following disrespectful (vs. respectful) leader behavior and consequently perceive the leadership as being less (vs. more) unethical. To test our hypotheses, we conducted an experiment in which participants ostensibly interacted with a fellow (dis)respectful follower and a (dis)respectful leader (Study 1), three experimental vignette studies (Studies 2a-c), and a critical incident study (Study 3). We find consistent support for the interaction rationale and the underlying processes. We discuss the resultant significant challenges for research on ethical leadership and the practice of the same.

We present the first meta-analysis on willful ignorance – when individuals avoid information about the negative consequences of their action to maximize own outcomes – covering 34,007 decisions made by 6,434 participants across 56 treatments. Results reveal that being able to remain ignorant drives people to act selfishly, even at the cost of harming others. We estimate that about 40% of the observed ignorance is driven by “reluctant altruists” who use it to excuse selfishness. The finding suggests that not all pro-social people are genuinely pro-social.