The global outbreak of COVID-19 has caused severe physical and economic damage. The extent of damage differs across countries, and is relatively small in Japan. This study investigated the predictors of infection prevention behavior among Japanese people and attempted to understand why the damage in Japan was less compared to other countries. We explored the following predictions: (1) people who perceive higher COVID-19 risks will engage in infection prevention behaviors regardless of their perceived norms; and (2) people who perceive lower risks for COVID-19 will engage in infection prevention only when they perceive infection prevention behavior as a social norm. We conducted two studies by recruiting 1,588 and 339 participants for studies 1 and 2, respectively. In Study 1, as an indicator of the perceived infection risk, we measured whether participants had been vaccinated, assuming that unvaccinated people perceived COVID-19 to be low risk. In Study 2, we directly measured the perceived infection risk. The results were consistent with our predictions, suggesting that social norms promote infection prevention behavior, even among individuals who perceived COVID-19 as low risk. This may be one of the reasons for the relatively small COVID-19-related damage in Japan.

This study demonstrates a new approach for boosting peoples decision-making abilities. Previous studies have demon-strated that those who have less knowledge are more likely to make a scientifically biased decision. Therefore, we canexpect that providing them with more knowledge can reduce their biases. However, since people have difficulty changingtheir minds in response to knowledge that contradicts their opinions, it is unclear whether people accept and appropriatelyunderstand the provided knowledge. To more efficiently help these people, this study focused on the illusion of explanatorydepth, which means that the knowledge people think they have is greater than the knowledge they actually possess. Weconducted two experiments and demonstrated that (i) those who had a stronger illusion of explanatory depth were morelikely to make a logically biased decision, and (ii) by informing them of their illusion and providing them with objectiveknowledge, we could reduce their bias.

People apply more frequently when “apply” is the default choice (Apply Default architecture) than when “do not apply” is the default choice (Not-Apply Default architecture). However, Apply Default architecture might let them make choices inconsistent with their preferences as this architecture is counterintuitive. Those trying to apply might mistakenly choose to not apply under Apply Default architecture. In this study, we hypothesized that people's choices under No-Default architecture (i.e., a choice architecture without a default option) are less consistent with those under Apply Default architecture than those under Not-Apply Default architecture (Hypothesis 1). We also hypothesized that people who spent more time on making decisions would make choices consistent with their preferences because when people spend sufficient time to understand the construction of Apply Default architecture, they can make choices consistent with their preferences (Hypothesis 2). We recruited 997 participants and asked them to make decisions under No-Default and Default architectures (Apply Default or Not-Apply Default architecture). The results supported both Hypothesis 1 and Hypothesis 2. A method to help applicants make choices consistent with their preferences is finally discussed.