Route guidance attacks in cyber transportation networks: a user-centered study of behavioral sensitivity

The rapid adoption of digital navigation systems and connected vehicles has introduced new vulnerabilities in transportation networks, including the risk of Route Guidance Attacks (RGAs) that disseminate distorted travel times. Drawing on traveler psychology and decision theory, this study investigates how RGAs propagate congestion through three behavioral models: perfect rationality, logit-based stochastic choice, and bounded rationality with an indifference threshold. By systematically varying attack intensity and traveler indifference levels, we simulate user responses in Sioux Falls, Anaheim, and Chicago networks. The results show that user behavior and network structure jointly determine the severity and spatial distribution of congestion: dense networks can initially absorb low-intensity misinformation but undergo sharp overload once critical thresholds are crossed, whereas sparser networks succumb to traffic disruptions even under modest falsifications. Perfectly rational users exhibit collective behavior toward the shortest route, logit-based users disperse but remain susceptible at high intensities, and boundedly rational travelers alter their routes when a strategically induced benefit surpasses their indifference level. These findings underscore the necessity of coupling cybersecurity measures with interventions that account for trust, risk perception, and the cognitive heuristics travelers use to evaluate route choices. In particular, user-interface designs featuring reliability scores, timely alerts, or partial verifications can reduce blind compliance and mitigate the sudden mass switching that amplifies adversarial manipulation. This study enhances the interaction between traveler behavior models and network topology under RGAs, contributing to transportation psychology, informing safer route guidance design, and highlighting strategies for improving network resilience.