Prediction of Driver's Cognitive Workload using Cognitive Architecture : ACT-R
DOI: 10.7232/ieif.2012.25.2.187
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 challenge of predicting driver cognitive workload during Human-Vehicle Interaction (HVI) tasks, such as operating navigation or audio systems. Traditional methods for evaluating HVI usability rely on empirical data from human participants, which is costly, time-consuming, and potentially unsafe due to the risk of accidents during secondary task performance. The authors propose using the ACT-R (Adaptive Control of Thought–Rational) cognitive architecture to simulate driver behavior and quantitatively predict both performance times and cognitive workload in the early stages of system development, thereby reducing reliance on physical prototypes and human subjects. The researchers developed a driver model based on ACT-R 6.0, incorporating production rules for lateral and longitudinal vehicle control as well as secondary task execution. The model utilizes a threaded cognition approach, allowing for the simultaneous processing of driving and secondary tasks while accounting for resource competition between modules. To validate the model, an experiment was conducted with 15 participants who performed 10 different HVI tasks (including static/dynamic menu selection, destination search, and music selection) under two driving conditions: straight and curved roads. The study measured empirical performance times and subjective workload ratings using the NASA-TLX instrument. The ACT-R model’s predicted workload was calculated based on the activation time of its cognitive modules, weighted by their functional complexity and adjusted for resource competition errors. The results demonstrated a strong correlation between the model’s predictions and empirical data. For performance time, the model showed a high coefficient of determination ($r^2 = 0.950$) and a low root mean square error (RMSE = 1.253). Similarly, the predicted cognitive workload closely matched the subjective NASA-TLX ratings ($r^2 = 0.846$, RMSE = 6.837). When analyzed separately by road condition, the model’s accuracy improved further, with $r^2$ values of 0.952 for straight roads and 0.982 for curved roads. The study noted that while the model generally predicted lower workload values than human subjects, this discrepancy was attributed to the model’s idealized attention switching compared to the less efficient multitasking of untrained participants. The significance of this research lies in the validation of ACT-R as a robust tool for the early-stage evaluation of HVI designs. By accurately predicting driver performance and cognitive load without the need for physical prototypes or human trials, this method offers a cost-effective and safe alternative for usability testing. The findings support the application of cognitive architectures in human factors engineering to optimize interface design and enhance driving safety by identifying high-workload scenarios before system deployment.
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 | 2026-08-09 |
| archive | success | canonical_url | — | — | 1 | 2026-08-09 |
| extract | success | cached | — | — | 3 | 2026-08-10 |
| clean | success | clean | — | — | 1 | 2026-08-09 |
| chunk | success | chunk | — | — | 1 | 2026-08-09 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-08-09 |
| enrich | success | semantic_scholar | — | — | 1 | 2026-08-09 |
| promote | success | — | — | — | 1 | 2026-08-09 |
| summarize | success | llm | qwen3.6-27b-nvidia | summ-v5 | 2 | 2026-08-10 |
| tag | success | vector_similarity | — | — | 10 | 2026-08-11 |
| verify | partial | — | — | — | 2 | 2026-08-10 |
Summary generated by qwen3.6-27b-nvidia on 2026-08-10; verification: verified_with_issues.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
- mental demand
- workload measurement
- cognitive capacity variation
- situational awareness
- mental model of traffic
- attention
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).
- Methodological Resource: tool software
- Theoretical Contribution: computational model, theory or model