Leveraging Protection Motivation Theory to support secure online behaviours: Two experimental studies with vulnerable age groups

Cyber security represents a growing societal challenge, with different population groups facing distinct vulnerabilities while often receiving little actionable guidance from public authorities and media outlets. This study investigates Protection Motivation Theory (PMT)-informed risk communication in two target-group-specific cyber security contexts. In two preregistered randomised posttest-only experiments, we tested PMT-based messages in different media formats, text versus video, among adolescents and young adults aged 14 to 21 years (N = 1193) and older adults aged 60 to 75 years (N = 1085). Each experiment introduced a target-group-specific threat scenario and tested different variants of PMT-based risk messages. The studies assessed post-treatment behavioural intention and experiment-specific in-survey link-click behavioural proxies. In Experiment 1, no statistically significant differences in post-treatment behavioural intention were detected across the message conditions. Individual post-treatment perceptions of coping-appraisal elements were associated with intention in supplementary exploratory analyses. In Experiment 2, participants in several PMT-based message conditions reported higher post-treatment behavioural intention than participants in the neutral condition. However, coping-enhanced messages and messages emphasising individual coping components did not significantly outperform threat-only communication. Neither experiment showed statistically significant message-condition differences in its respective in-survey link-click behavioural proxy. The primary format comparisons also provided no statistically significant evidence that text and video differed in their effects on post-treatment intention. The experiments produced different context-specific patterns. The findings therefore support cautious conclusions about post-treatment protective intention within the examined contexts but do not demonstrate isolated effects of individual PMT components or actual cyber security behaviour change.