A cognitive process approach to modeling gap acceptance in overtaking

Mohammad, Samir H.A.; Farah, Haneen; Zgonnikov, Arkady · 2023 · arXiv

DOI: 10.48550/arxiv.2306.05203

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 modeling human driver behavior during overtaking maneuvers in mixed traffic environments containing automated vehicles (AVs). While existing gap acceptance models provide insights into human decision-making, they often fail to capture the dynamic, high-speed nature of overtaking interactions or translate effectively to human-AV scenarios. The authors aim to bridge this gap by employing a cognitive process approach, specifically using drift-diffusion models (DDMs), to describe how human drivers dynamically accept or reject gaps when an oncoming vehicle is present. The study begins with a conceptual analysis of overtaking, identifying implicit communication and dynamic vehicle motion as critical factors. The authors assess various modeling approaches, concluding that cognitive models, particularly DDMs, are best suited for capturing response times and dynamic interactions. To validate this, they conduct a proof-of-concept study using a dataset of 1,758 overtaking decisions from a driving simulator experiment. The data includes decision outcomes and response times under varying initial distances to oncoming vehicles and initial ego-vehicle velocities. The researchers tested eight variations of the DDM, modifying components such as the drift rate, decision boundary, and initial bias to account for the ego vehicle’s initial speed, a factor previously ignored in low-speed gap acceptance models. The results demonstrate that incorporating an initial decision-making bias dependent on the ego vehicle’s initial velocity allows the DDM to accurately replicate observed human behavior. Specifically, Model M6, which includes a velocity-dependent bias and links the drift rate and decision boundary to time-to-arrival and distance, successfully captured all six qualitative patterns of human overtaking decisions, including how probability and response times vary with speed and gap size. Models lacking this velocity-dependent bias failed to predict the decrease in response times for accepted gaps at higher velocities. The significance of this work lies in its demonstration that cognitive process models can effectively predict human overtaking behavior in dynamic, high-speed scenarios. This approach offers a pathway for developing safer and more efficient interaction strategies for AVs, enabling them to anticipate human decisions and adjust their motion dynamics accordingly. The findings suggest that DDMs can enhance the training and validation of interaction-aware controllers, potentially reducing head-on collision risks and improving traffic flow efficiency in mixed traffic environments.

Key finding

Standard drift-diffusion models of gap acceptance cannot reproduce empirical overtaking response-time patterns, but adding a velocity-dependent initial decision bias to the DDM accurately captures the qualitative pattern of accept/reject decisions and response times observed in human drivers during high-speed overtaking against an oncoming vehicle, supporting cognitive-process (rather than static gap-threshold) accounts of human-AV overtaking interactions.

Methodology

simulator

Sample size: N=25 participants (re-analysis of 2,097 overtaking decisions from Sevenster et al. driving-simulator dataset)

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 tag_papers on 2026-05-30 (2 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified_with_issues.

Topics

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

• gap acceptance
• mental model of traffic
• anticipation
• risk taking
• following distance
• pedestrian behavior perception

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