Abstract
Highlights Changes in the subjects’ emotional state were detected using a dynamic simulator. • Acceptance of a human vs. AD logic were qualitatively and quantitatively assessed. • The ECG is able to retrieve accurate information about the subjects’ emotional state. Abstract Advancements in autonomous vehicle technology continue to progress rapidly, but concerns regarding public acceptance persist. Traditional studies have used questionnaires to assess social acceptance, but these have limitations. To gather more reliable data, tracking physiological signals offers a promising alternative. This study aims to develop and test an experimental protocol to assess human acceptance of autonomous driving by using both objective measures - i.e.: Electroencephalogram (EEG), Electrocardiogram (ECG), Skin Potential Response (SPR), and Eye Tracking (ET) - and subjective measures, by means of surveys. The tests were conducted in a controlled yet realistic environment using the VI-Grade DiM400 dynamic driving simulator. The study involved 31 participants who acted as passengers in a vehicle driven either by a human driver or by different human-like driver models across three driving scenarios. Various indices were extracted from each biosignal, and a repeated measures ANOVA was performed to assess differences in participants’ emotional states between human and autonomous driving, as well as during specific driving events. The ANOVA results indicated that the setup effectively captured changes in emotional state during events, with ECG time domain indices being particularly sensitive. However, no significant differences were found in the cognitive state between human and autonomous driving, suggesting a consistently relaxed state across conditions — a finding also supported by the questionnaires. This study demonstrates the potential of using ECG signals to objectively measure emotional responses in driving scenarios, providing valuable insights into trust in autonomous vehicles and offering new possibilities for driving safety and monitoring applications. Keywords Physiological signals Dynamic driving simulator Autonomous driving Experimental protocol
Authors
Mengucci, G., Cioffi, A., Galli, M., Melzi, Tomasini, S. &., & G.
https://doi.org/10.1016/j.trip.2025.101546