Exploring factors in Human Cognition that affect Human Performance during Takeover in Autonomous Driving

Liu, Zijing; Zhang, Jiayi · 2024 · Crossref

DOI: 10.54097/tr5g0e77

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 investigates the human cognitive and psychological factors influencing driver performance during mandatory takeovers in Level 3 autonomous vehicles. Motivated by safety concerns regarding automation-initiated, driver-controlled (AIDC) transitions, where nearly 89% of reported collisions occur, the research aims to identify specific workload factors that impair situational awareness and decision-making. The authors argue that existing literature lacks clarity on the precise manifestations and priority levels of these factors, hindering effective Human-Machine Interface (HMI) design. To address this, the study constructs a theoretical model categorizing driver workload into cognitive, emotional, and environmental loads, hypothesizing that specific subordinate factors within these categories significantly impact takeover success. The methodology involved a quantitative survey administered to 849 participants with experience in SAE Level-3 autonomous systems. After filtering for validity, 374 responses were analyzed. The survey utilized a five-point Likert scale to measure 16 observational variables across the three workload dimensions. Data analysis employed Kaiser-Meyer-Olkin (KMO) and Bartlett’s tests to confirm structural validity, followed by Exploratory Factor Analysis (EFA) to extract and prioritize influencing factors. The KMO value of 0.946 and significant Bartlett’s test results indicated the data was suitable for factor analysis. The EFA results identified three significant factors explaining 53.3% of the variance. Factor 1, representing cognitive workload, accounted for the largest variance (22.7%) and included variables related to the clarity of takeover messages, the driver’s arousal level, and the system’s transparency regarding real-time vehicle and traffic status. Factor 2, associated with emotional workload (20.8%), highlighted the impact of multi-modal cues, anthropomorphic messaging styles, and message duration. Factor 3, linked to environmental workload (9.8%), focused on the cognitive difficulty of comprehending text-based messages and the resulting distraction from external road conditions. The analysis revealed that the clarity of the driver’s awareness and the arousal level induced by the takeover message were the most critical determinants of workload. The study concludes that cognitive workload, specifically driven by message clarity and driver arousal, exerts the greatest influence on driver performance during AIDC transitions. These findings suggest that HMI designs for autonomous vehicles should prioritize interfaces that effectively arouse driver attention while providing clear, comprehensive, and humanized information. By optimizing these factors, designers can reduce mental burden and improve the safety and efficiency of human-automation handovers, addressing the "out-of-the-loop" performance problems inherent in conditional automation.

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-nvidia on 2026-08-10; verification: verified.

Topics

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

• automation
• takeover transitions
• situational awareness
• automation surprise
• automation complacency bias
• mode awareness

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: behavioral performance data
• Theoretical Contribution: conceptual framework, theory or model