Driving behavior experiments using functional magnetic resonance imaging: A systematic review

Abstract

Driving a vehicle requires complex brain functions, including attention, motor coordination, and decision-making. Human factors are the main cause of 93% of crashes. Nowadays, it is possible to scan the brain using functional magnetic resonance imaging (fMRI) while performing driving-related tasks, often in simulated environments. This study tries to provide a comprehensive review of the application of brain fMRI in the field of driving behavior. Scopus, PubMed, and Google Scholar were searched until September 2025. Original articles that studied driving using fMRI were included. Screening and selection followed PRISMA guidelines. Seven major topics were identified. For each topic, a unique extraction table was designed. The objectives, participant characteristics, task design, outcome measurements, and behavioral findings were extracted. The final 80 original studies were included. Seven major topics, including distracted driving, intoxicated driving, risk-taking behavior while driving, simulated driving with brain impairment, drivers' physiological state, general brain activity while driving, and simulated driving environment and simulator characteristics, were identified. fMRI studies suggest the involvement of dynamic brain networks underlying driving behavior, influenced by cognitive load, impairment, and environment. Across domains, driving impairment appears to emerges as a network-level phenomenon characterized by disrupted executive control, altered reward valuation, and reduced visuospatial predictive processing. We interpret these findings within Menon's triple network framework, in which driving impairment is characterized as a disruption in the dynamic interactions among the central executive, salience, and default mode networks. Future studies should work on driving environment effects and overcoming methodological constraints of fMRI for more realistic simulation.