Assessment selection in human-automation interaction studies: The Failure-GAM2E and review of assessment methods for highly automated driving

Grane, Camilla · 2018 · Crossref

DOI: 10.1016/j.apergo.2017.08.010

archive: archived pipeline: cataloged verified

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

Summary

This paper addresses the challenge of selecting appropriate assessment methods for human-automation interaction studies, specifically within the context of highly automated driving. As vehicles become more automated, the driver’s role shifts from manual control to supervision, rendering traditional driving performance metrics (e.g., steering angle, braking pressure) largely irrelevant except during take-over scenarios. The author argues that existing literature lacks a systematic approach to selecting assessment methods for this new domain, often relying on traditional measures that fail to capture emergent factors like trust, situation awareness, and mental workload. To address this gap, the paper proposes the Failure-GAM2E model, a structured method designed to guide researchers in identifying relevant failures, goals, actions, and corresponding subjective and objective assessment methods. The study employs a two-part methodology. First, the author conducts a review of 35 journal papers investigating human behavior in highly automated vehicles. The review categorizes assessment methods used for key human factors, including mental workload, situation awareness, trust, acceptance, fatigue, and take-over performance. The analysis reveals that while mental workload and situation awareness are frequently studied, there is significant variability in the methods used, with NASA-TLX being the most common subjective measure for workload. Second, the author develops the Failure-GAM2E model by combining elements from Hazard Analysis (ISO 26262) and Goal-Operators-Methods-Selection (GOMS) rules. The model consists of three main steps: defining the situation, executing six sub-steps (identifying Failures, Goals, Actions, Subjective Methods, Objective Methods, and Equipment), and finalizing the selection based on resources. The application of Failure-GAM2E is exemplified through its use in the MODAS project, which developed an information and warning system for automated driving. The systematic application of the model resulted in a well-reasoned assessment plan that moved beyond traditional metrics. Notably, it led to the identification of novel measurement techniques, such as using foot movements to assess trust, and the proposal of an Optimal Risk Management Model. The review findings indicate that while many studies still rely on traditional methods, the Failure-GAM2E framework helps researchers systematically link potential system failures to specific driver goals and actions, ensuring that selected assessment methods are directly relevant to the human-automation interaction. The significance of this work lies in providing a practical, structured tool for researchers and vehicle developers exploring new areas of human-automation interaction. By forcing a systematic breakdown of the driving scenario from failure modes to specific assessment equipment, Failure-GAM2E reduces the complexity of study design and ensures that assessment methods are well-motivated and comprehensive. The paper concludes that this approach supports the creation of robust studies capable of capturing the nuanced behavioral changes associated with highly automated driving, thereby aiding in the development of safer and more effective human-automation interfaces.

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
• automation complacency bias
• automation surprise
• situational awareness
• mode awareness
• trust calibration

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: self report data
• Methodological Resource: measurement protocol
• Theoretical Contribution: conceptual framework