Technology acceptance of passive brain-computer interfaces in vehicles for adaptive automation

Driver monitoring systems (DMS) enhance traffic safety by detecting critical driver states such as drowsiness and cognitive overload. Future brain–computer interfaces (BCIs) for driver monitoring and workload-adaptive automation promise direct access to neurophysiological signals, potentially offering higher sensitivity than conventional systems. Despite this potential, drivers' acceptance of BCIs in real-world driving remains largely unexplored. Because driving constitutes a complex sociotechnical system context, user, task, and system characteristics are likely to shape acceptance. This study thus examines the acceptance of non-invasive, wearable BCIs for workload-adaptive driver assistance from a sociotechnical perspective. We conducted an online experimental vignette study using a between-participant design ( N = 1201). Participants evaluated one of twelve animated driving scenarios in which a BCI detected high workload and initiated assistance. The scenarios varied three sociotechnical characteristics: driver status (able-bodied vs. disabled), the function within the assisted driving task (primary, secondary, or tertiary), and type of automation (adaptive vs. hybrid). Respondents evaluated acceptance using an adapted Unified Theory of Acceptance and Use of Technology (UTAUT2) framework. Acceptance of BCI-based adaptive automation varied systematically across conditions. Findings show that scenario evaluations indicated higher acceptance when BCIs supported driving-related functions and when the scenario depicted a driver with a disability. No differences were observed between the two activation modes tested. By integrating sociotechnical characteristics into an established acceptance framework, this study advances understanding of BCI acceptance in driving and informs the user-centered design of neuroadaptive driver assistance systems.