Human Merging Behavior in a Coupled Driving Simulator: How Do We Resolve Conflicts?
archive: archived pipeline: cataloged verified
Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)
Summary
This study addresses the limitations of naturalistic data in understanding human merging behavior, specifically the inability to control initial vehicle kinematics and observe operational driver inputs. To investigate how drivers resolve conflicts in a controlled environment, the authors conducted an experiment using a coupled, top-down driving simulator with nine pairs of participants. The scenario involved two vehicles approaching a merge point where their lanes combined into one. Participants controlled acceleration via a steering-wheel game controller while maintaining their initial velocity to avoid collisions, with no right-of-way rules established. The experiment varied initial positions and velocities across 11 conditions to create specific projected headways and relative velocities, ensuring that maintaining initial speeds would result in a collision. The researchers analyzed both joint outcomes (who merged first and Conflict Resolution Time, or CRT) and individual operational behaviors (acceleration and velocity profiles). Statistical modeling revealed that initial kinematics significantly determined the outcome: a greater projected headway advantage increased the likelihood of merging first, while a higher relative velocity decreased it. The CRT, measuring the time required to resolve the conflict, was lowest when one driver had a clear headway advantage and highest in neutral conditions where no driver had a kinematic advantage. Crucially, analysis of velocity traces showed that drivers employed "intermittent piecewise-constant control," characterized by blocks of constant acceleration separated by sharp decision moments. This pattern indicates that drivers do not continuously optimize their inputs but rather select a plan and adhere to it until a specific trigger prompts a replanning event. Furthermore, drivers frequently initiated compatible actions (one accelerating, one decelerating) immediately upon gaining control, suggesting the use of a shared mental model to negotiate the merge outcome before active interaction began. The findings challenge prevailing assumptions in traffic modeling and autonomous vehicle design. The evidence that drivers use intermittent, piecewise-constant control rather than continuous utility maximization suggests that models assuming continuous optimization of reward functions are inconsistent with human behavior. Instead, driver behavior aligns more closely with satisficing strategies, where drivers seek adequate plans and adjust only when necessary. Additionally, the strong dependence of merging outcomes on initial kinematics, rather than individual driver preferences, implies that modeling individual differences in reward functions is only relevant for scenarios with minimal kinematic disparities. These results advocate for the development of interaction models based on intermittent control and highlight the importance of considering both projected headway and relative velocity in predicting merging behavior.
Key finding
Drivers resolve merging conflicts using intermittent piecewise-constant control driven by key decision moments rather than continuous optimization, with the outcome heavily influenced by initial projected headway and relative velocity.
Methodology
simulator
Sample size: 18
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. Discovered via author_sweep_intake on 2026-05-27.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | author_sweep | — | — | 2 | 2026-05-27 |
| archive | success | canonical_url | — | — | 5 | 2026-06-06 |
| extract | success | cached | — | — | 3 | 2026-06-10 |
| clean | success | clean | — | — | 1 | 2026-06-07 |
| chunk | success | chunk | — | — | 1 | 2026-06-07 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-07 |
| enrich | skipped | — | — | — | 4 | 2026-07-02 |
| promote | success | — | — | — | 1 | 2026-06-04 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 2 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 15 | 2026-06-11 |
| verify | success | — | — | — | 2 | 2026-06-10 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
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: computational model, theory or model