The COVID-19 pandemic presented an unprecedented global health emergency and socio-economic crisis. In Italy, a strict lockdown from March to May 2020 drastically altered work habits and daily life. Millions faced job losses and economic hardship alongside the health crisis. Prior research on risk-taking behavior after catastrophic events like natural disasters or wars showed conflicting results: some studies indicated increased risk aversion, while others found increased risk-loving behavior. However, these studies often used monetary tasks and collected data after the event, potentially missing real-time behavioral changes and neglecting the impact of ecological factors like health and employment status. This study aimed to address these gaps by using a novel task, the Cov19 Risk Task (Cov19-RT), which included monetary and ecological (health and employment status) conditions to assess risk-taking during the actual lockdown phase in Italy. The researchers hypothesized that participants would be more risk-averse in monetary conditions than in health and employment conditions, and that healthcare workers (whose employment remained stable) would be less risk-averse in employment-related scenarios compared to non-healthcare workers.

Existing research on risk-taking behavior following catastrophic events is contradictory. Studies of individuals affected by events such as tsunamis or civil unrest have shown increased risk aversion, suggesting a heightened perception of danger and a preference for safety. However, other research, such as studies following Hurricane Katrina, has found increased risk-taking behavior, potentially due to a sense of invincibility or desperation. This discrepancy in findings highlights the need for further research to understand the factors influencing risk perception and behavior in the wake of such events, and the limitations of using solely monetary stimuli. A key limitation of previous studies is the time lag between the event and data collection. The present study aims to address this gap by capturing immediate risk-taking behaviors during a major ongoing crisis.

The study employed a modified Holt-Laury Paired Lottery Task, forming the Cov19-RT. This included a monetary condition (Cov19-RT MC) and two novel ecological conditions: a Health Status Condition (Cov19-RT HSC) and an Employment Status Condition (Cov19-RT ESC). The task presented participants with a series of choices between two lotteries with varying probabilities and payoffs. In the MC condition, payoffs were monetary. In the HSC condition, payoffs involved hypothetical health outcomes (e.g., symptomatic COVID-19 infection, shoulder fracture), and in the ESC condition, payoffs represented hypothetical employment outcomes (e.g., paid leave, job loss). Each condition had two series with different payoff amounts (Series 1: lower payoffs; Series 2: higher payoffs) to analyze the incentive effect. Participants were recruited from the PsyCOVID study, a larger study investigating the psychological impact of the pandemic in Italy. After applying exclusion criteria (age, employment status), 140 healthcare workers and 140 matched non-healthcare workers were invited to participate in the Cov19-RT. 130 Participants (65 healthcare workers and 65 non-healthcare workers) fully completed the online survey administered via Google Forms. Data were analyzed using non-parametric tests to compare risk-taking behavior across conditions and groups. Spearman's Rank correlation assessed relationships between risk measures; Wilcoxon signed-rank tests examined the incentive effect within groups; Freedman Test and Wilcoxon post-hoc analyses investigated differences across conditions; and Mann-Whitney U tests compared groups (healthcare vs. non-healthcare workers).

The study's primary finding was that participants exhibited greater risk aversion in the monetary condition (Cov19-RT MC) compared to both the health (Cov19-RT HSC) and employment (Cov19-RT ESC) conditions. A significant positive correlation was found between the HSC and ESC conditions, implying a link between risk perceptions in these ecological domains. However, no significant correlation existed between the MC condition and the ecological conditions, suggesting that risk-taking in monetary domains is distinct from ecological risk-taking. The incentive effect (a shift toward risk aversion with increased payoffs) was observed in the MC condition as expected. However, a reverse incentive effect was observed in both the HSC and ESC conditions—increased payoffs led to increased risk-loving behavior. Healthcare workers showed significantly less risk aversion than non-healthcare workers in the ESC condition but not in the MC or HSC conditions. This could be related to the greater job security experienced by healthcare workers during the pandemic. The sample comprised mainly young adults (mean age 38, median 37) which might have contributed to the observed reverse incentive effect, as younger individuals tend to be more risk-prone.

The findings suggest that the context of risk significantly influences risk-taking behavior. The use of ecological stimuli, directly relevant to the participants' lived experiences during the pandemic, yielded results that differ from those obtained using traditional monetary lotteries. The reverse incentive effect in the health and employment conditions might reflect a prioritizing of outcomes with significant personal and societal impact over abstract monetary values. The difference in risk aversion between healthcare and non-healthcare workers further highlights the role of personal circumstances and security in shaping risk perception. These results emphasize the need for decision-making models that incorporate both monetary and ecological factors, especially in crisis situations. The ecological relevance of stimuli enhanced the validity of this study in capturing real-life behavior in a time of crisis.

This study demonstrates that experiencing the psychological and economic consequences of the COVID-19 pandemic can modulate individual risk-taking behaviors. The findings highlight the importance of considering both monetary and ecological factors when assessing risk-taking during crises. Future research should investigate the long-term effects of the pandemic on risk perception, expand the study's sample size, and potentially consider other factors such as personality traits and prior experiences. Further investigation is needed into the impact of the reverse incentive effect observed and whether it is a temporary phenomena linked to youth and pandemic-specific anxieties.

The cross-sectional nature of the study limits causal inferences about changes in risk-taking aptitudes. The sample size was relatively small and had an unbalanced gender ratio between healthcare and non-healthcare workers (due to the convenience sampling). Geographic location was not controlled for. These factors might affect the generalizability of the findings. Future studies should address these limitations using longitudinal designs, larger and more balanced samples, and geographic controls.