Road aggression in simulated and virtual driving and human-machine interaction: a systematic review and bibliometric analysis
Keywords:
aggressive driving; driving anger; virtual reality; driving simulation; human–machine interaction.Abstract
Road aggression is an important concern in traffic psychology, yet its study in virtual and simulated driving environments is complicated by inconsistent distinctions among aggression, anger, frustration, risky driving, and operationally defined aggressive driving styles. This systematic review and bibliometric analysis examined how aggression-related phenomena have been conceptualized, assessed, and investigated in virtual driving and human–machine interaction research. Searches of Scopus, Web of Science Core Collection, and APA PsycInfo yielded 134 records; after screening and eligibility assessment, 37 publications were included. Three findings were particularly relevant from a psychological perspective. First, driving anger, frustration, trait aggression, road rage, aggressive expression, and kinematically defined aggressive driving styles were related but non-equivalent constructs. Second, affective activation did not consistently correspond to risky or aggressive vehicle behavior: anger or trait aggression could be present without directly predicting behaviors such as risky passing or tailgating, supporting a distinction between internal state and behavioral expression. Third, several studies combined self-report and psychometric measures with behavioral, psychophysiological, and vehicle-dynamics indicators, suggesting the value of convergent multimodal assessment for construct validity. Virtual and simulated environments provide controlled conditions for examining aggression-related processes, but vehicle dynamics alone should not be interpreted as evidence of psychological aggression without supporting contextual, emotional, intentional, behavioral, or psychometric evidence. Future research should prioritize clearer construct definitions and validated multimodal assessment when developing adaptive human–machine systems.






