Gaming tasks as a method for studying the impact of warning messages on information behavior

New and nontraditional approaches are required to effectively tackle the global problem of cybercrime. Online warning messages offer the unique potential to influence information behavior at the exact point of user decision-making. This research assessed the prevention effect of differing components of warning messages. Thirty-five male participants, aged 18–43, participated in a behavioral-compliance task comprising messages received when visiting websites likely to contain malware. Participants also rated messages on believability, severity, and effects on intention to comply. The components of messages tested were as follows: three "signal words" (warning, hazard, and stop), two levels of message explicitness (high, low), and two imagery conditions (eyes, no eyes). Contrary to expectations, explicitness was the only message component to yield a significant preventative effect on self-rated and behavioral responses. Participants not only perceived the explicit messages as more believable, severe, and likely to increase intention to comply but also demonstrated, through their behavioral-compliance data, a preventative effect from more explicit messages. The implications of these findings for designing messages to prevent cybercrimes are explored.