Rear Vehicle Matters: How Tailgaters Influence the Car-Following Behaviors of the Ego-Vehicle

Liu, Yang; Zhang, Jiahao; Ouyang, Yuxuan; He, Dengbo · 2025 · Proceedings of the Human Factors and Ergonomics Society Annual Meeting

DOI: 10.1177/10711813251358261

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 investigates the influence of following vehicles (FVs) on the car-following (CF) behaviors of ego-vehicles, addressing a gap in existing research that primarily focuses on the impact of lead vehicles (LVs). Motivated by the high prevalence of rear-end collisions and evidence that drivers perceive rear traffic information, the authors specifically examine how tailgating—defined as a time headway of less than two seconds between the FV and ego-vehicle—affects driver behavior. The research aims to quantify these "nudging effects" to improve traffic safety models and controller designs. The methodology utilized the highD naturalistic driving dataset, from which the authors extracted CF segments based on specific criteria: duration exceeding 15 seconds, following distance under 100 meters, no lane changes, and minimum speed above 10 m/s. This process yielded 2,999 tailgated CF events and 1,679 normal CF events. To isolate the effect of the FV from variations in traffic flow speed, the study employed the Dynamic Time Warping (DTW) algorithm to pair ego-vehicle trajectories with similar speed profiles across both event types, resulting in 477 matched pairs. The analysis compared speed fluctuation metrics (standard deviation, mean absolute deviation, coefficient of variation, and time-varying stochastic volatility) and safety metrics (mean time headway and maximum reciprocal time-to-collision) using paired t-tests. The results demonstrated that while speed fluctuation metrics showed no statistically significant differences between tailgated and normal events, confirming the effectiveness of the DTW pairing, there was a significant difference in safety metrics. Specifically, the mean time headway (meanHDW) was significantly lower in tailgated events (mean of 2.38 seconds) compared to normal events (mean of 2.62 seconds), representing a 10.8% reduction (p = .003). This indicates that when being tailgated, ego-vehicles maintain a smaller gap to their lead vehicles. Although the maximum reciprocal time-to-collision was higher in tailgated events, this difference was not statistically significant. The findings confirm that drivers adjust their CF strategies in response to rear-vehicle pressure, effectively being "nudged" to drive closer to the LV when tailgated. This suggests that drivers consider more than just LV information, adapting their behavior to the complexity of surrounding traffic. The authors conclude that future driver behavior models and traffic simulations must incorporate the influence of FVs to accurately represent real-world driving dynamics. They recommend further research using data-driven methods, such as inverse reinforcement learning, to quantitatively model these behaviors and validate findings across additional datasets.

Key finding

Drivers maintain a significantly smaller mean time headway to their lead vehicle when being tailgated compared to normal car-following conditions.

Methodology

dataset

Sample size: 4678

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-prismaquant on 2026-06-10; verification: verified.

Topics

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

• following distance
• traffic density
• aggressive driving
• secondary collisions

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