ITSC 2025 Paper Abstract

Paper TH-LM-T29.2

Zhang, Junjie (Southeast University), Ma, Yongfeng (Southeast University), Xing, Guanyang (Southeast University), Zhang, Ziyu (Southeast University), Kang, Kai (Southeast University), Gao, Lu (Southeast University)

The Impact of Human–Machine Interaction Transparency on Driver Trust in Autonomous Driving

Scheduled for presentation during the Regular Session "S29a-Human Factors and Human Machine Interaction in Automated Driving" (TH-LM-T29), Thursday, November 20, 2025, 10:50−11:10, Currumbin

2025 IEEE 28th International Conference on Intelligent Transportation Systems (ITSC), November 18-21, 2025, Gold Coast, Australia

This information is tentative and subject to change. Compiled on October 18, 2025

Keywords Human-Machine Interaction Systems for Enhanced Driver Assistance and Safety, Trust, Acceptance, and Public Perception of Autonomous Transportation Technologies, User-Centric HMI Design for Autonomous Vehicle Control Systems

Abstract

In autonomous driving environments, drivers' trust in the system directly influences the effectiveness of human-machine interaction and driving safety, while interaction transparency is regarded as a key factor in enhancing trust. This study designed a simulated lane-change scenario in a work zone using the Carla driving simulator. Four levels of human-machine interaction transparency were implemented: no notification, a single notification, two notifications, and a single notification with an explanation of the lane-change reason. Trust questionnaire data and eye-tracking indicators were collected from 58 drivers to evaluate the influence of transparency on the dynamics of trust and attention distribution. The results show that as interaction transparency increases, drivers' trust levels significantly improve, with the highest trust observed when the reason for the lane change is explained. Furthermore, high transparency conditions were associated with reduced pupil diameter variability, longer average fixation durations, and lower saccade frequencies, suggesting that drivers experienced lower cognitive load and more efficient attention allocation under high transparency. These findings underscore the importance of system transparency in human-machine co-driving design and provide theoretical support for enhancing system explainability and driver acceptance in autonomous vehicles.