Assessing Training Methods for Advanced Driver Assistance Systems and Autonomous Vehicle Functions: Impact on User Mental Models and Performance

Murtaza, Mohsin; Cheng, Chi‐Tsun; Fard, Mohammad; Zeleznikow, John · 2024 · OpenAlex-citations

DOI: 10.3390/app14062348

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study addresses the critical need for effective training methodologies for Advanced Driver Assistance Systems (ADAS) and Autonomous Vehicle (AV) technologies. With road traffic accidents remaining a major public health challenge, largely driven by human factors, ADAS and AVs offer significant safety potential. However, low adoption rates for mature functions like Adaptive Cruise Control (ACC) and Lane-Keeping Assist (LKA) suggest a gap in user understanding. The research investigates how different training methods influence drivers’ mental models—defined as cognitive frameworks for understanding system capabilities and limitations—and subsequent performance. The authors argue that accurate mental models are essential for safe interaction, preventing misuse or excessive reliance on automation. To evaluate this, the researchers conducted a comparative analysis of text-based and video-based training methods using a driving simulation. The experiment utilized the York driving simulator (version 6.61) integrated with a Logitech G27 racing wheel, chosen for its ability to simulate human factors and driver interaction. Participants were tasked with activating specific ADAS and AV functions, including Lane-Keeping Assist (LKA), Collision Avoidance (CA), Adaptive Cruise Control (ACC), and Autopilot, within a realistic three-dimensional virtual environment featuring varied traffic patterns and road conditions. Performance was measured by the accuracy of function activation and reaction times during simulated driving scenarios. The study aimed to determine which training approach better facilitated the development of robust mental models and improved operational proficiency. The findings indicate that video-based training yielded superior outcomes compared to text-based methods. Participants who received video-based instruction demonstrated better performance metrics, including higher accuracy and faster reaction times when interacting with ADAS and AV functions. Furthermore, these participants developed more accurate mental models, exhibiting a deeper understanding of the functionalities, capabilities, and limitations of the systems. The study highlights that video-based learning aligns with cognitive load theory, suggesting that multimedia instruction enhances information absorption and meaningful comprehension in complex technological contexts. The significance of these results lies in their implications for policy makers, automotive manufacturers, and educational institutions. The study underscores the necessity of developing tailored, video-centric training programs to facilitate the proficient and safe operation of increasingly complex automotive technologies. By improving driver education through effective training methodologies, the research suggests that adoption rates for ADAS and AV functions can be increased, thereby enhancing road safety and reducing accident rates. The findings provide empirical evidence supporting the shift from traditional text-based manuals to immersive, visual training approaches in the automotive sector.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• simulator training transfer