The available evidence from anthropology, economics, and psychology suggests that sensitivities to the emotions of shame and guilt vary across cultures. So does (over)confidence in ability and skills. Is there a connection between these observations? We address this question theoretically and empirically. Theoretically, we explore the socially optimal combination of psychological incentives and the emergence of different cultural equilibria. Empirically, we find significant evidence of a negative relationship between individual confidence and the cultural importance of shame versus guilt. The relationship holds across countries, and for U.S. immigrants relative to their culture of origin, suggesting a causal effect still significant after more than eight years.
Do people correctly perceive the limitation of their attention when dealing with multiple tasks? We develop a simple model that predicts that, under rational inattention, individuals can correctly perceive their attention on a future task and possible attention spillover in a dual-task setting. We test our theoretical predictions in an online experiment and find that individuals generally overestimate their future attention to a scheduled, incentivized task, and thus report exaggerated valuation of the task. We also document that dual tasks have positive spillover effects on each other, improving baseline attention level for task completion, and people can indeed anticipate such positive spillover effects.
Do people value their attention optimally? Existing findings suggest that individuals systematically undervalue by how much attention-increasing technologies, in particular reminders, can boost their chance of completing future tasks. In a theory-driven experiment, we revisit this question and elicit a measure of individuals' valuation of reminders that is free from arbitrary risk preferences, under an incentive scheme of accumulating probability points to win a binary lottery. We find that even under such incentive structure, individuals still do not fully value the effectiveness of reminders. The violation of optimality cannot be explained by potential probability weighting.
This study analyzes the impact of trading- and non-trading-hour opinions on returns using data collected from an online stock forum in China. We find that non-trading-hour opinions have a stronger influence on returns than trading-hour opinions. However, a return reversal is observed during the subsequent trading periods based on non-trading-hour opinions, suggesting a tug-of-war between individual investors and arbitrageurs. Additionally, the effect of non-trading-hour opinions on returns is higher when firms announce important events overnight. These opinions also attract more investor attention. We propose that the announcement of such events exposes investors to high levels of uncertainty, leading them to seek advice through online forums. Our analysis suggests that investor sentiment and value-relevant information contained in online articles are likely factors that contribute to the return predictability of these opinions.
