Modelling level 1 situation awareness in driving: A cognitive architecture approach
DOI: 10.1016/j.trc.2024.104737
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This research addresses the lack of computational models for Situation Awareness (SA) in driving, specifically focusing on Level 1 SA (perception of critical elements). While SA is widely studied empirically, existing approaches are descriptive rather than predictive, limiting their utility in early-stage system design and safety evaluation. The study aims to bridge this gap by developing a psychologically plausible computational model grounded in cognitive and perceptual mechanisms. The primary motivation is to enable researchers to simulate driver behavior and SA quantitatively, reducing reliance on costly and potentially risky human participant experiments while providing explanatory accounts of performance independent of modeller bias. The methodology involves the development of QN-ACT-R-SA, an integrated cognitive architecture that combines the Queueing Network-Adaptive Control of Thought-Rational (QN-ACT-R) framework with the dynamic visual sampling model (SEEV). This model simulates realistic attention allocation and information retrieval processes, allowing it to interact directly with a driving simulator. The model’s validity was assessed through three studies. Study I validated the model against probe-based SA measures (SAGAT and post-experiment questionnaires) under easy and complex driving conditions. Study II evaluated the model’s ability to simulate Brake Perception Response Time (BPRT) for on-road versus roadside hazards. Study III tested the model’s predictive power by comparing its outputs to empirical data regarding the effects of Adaptive Cruise Control and cell-phone use on SA. The results demonstrate that QN-ACT-R-SA accurately simulates human Level 1 SA across all three studies. In Study I, the model achieved a Mean Absolute Percentage Error (MAPE) of 5.02% and a Root Mean Square Error (RMSE) of 3.47 for SAGAT scores, and a MAPE of 6.73% and RMSE of 6.13 for post-experiment questionnaires. Study II showed that BPRT was significantly shorter for on-road hazards than roadside hazards, with the model achieving a MAPE of 9.4% and an RMSE of 0.13 seconds against empirical data. Study III confirmed the model’s predictive capability, yielding a MAPE of 5.6% and an RMSE of 4.9 when predicting the impacts of automation and distraction. These low error margins indicate a strong fit between the simulated and empirical results. The significance of this work lies in providing a robust, real-time programmable tool for predicting collective driver SA. By validating the model against established empirical measures, the research establishes a quantitative foundation for human factors engineering in driving. This approach allows for the testing of hypotheses and evaluation of design trade-offs, such as interface designs or roadway configurations, prior to formal experimentation. Furthermore, the study lays the groundwork for future extensions into modeling higher levels of SA (comprehension and projection) and applications in other complex domains, advancing the field from descriptive analysis to predictive simulation.
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.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | Crossref | — | — | 1 | 2026-06-17 |
| archive | success | unpaywall | — | — | 2 | 2026-06-25 |
| extract | success | pdftotext | — | — | 2 | 2026-06-26 |
| clean | success | clean | — | — | 1 | 2026-06-26 |
| chunk | success | chunk | — | — | 1 | 2026-06-26 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-26 |
| enrich | failed | — | — | — | 5 | 2026-07-05 |
| promote | success | — | — | — | 1 | 2026-06-17 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-25 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-26 |
| verify | partial | — | — | — | 1 | 2026-06-26 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-25; verification: verified_with_issues.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
- situational awareness
- mental model of traffic
- situation awareness theory
- anticipation
- attention
- cognitive capacity variation
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).
- Theoretical Contribution: computational model, theory or model, conceptual framework