How are drivers' stress levels and emotions associated with the driving context? A naturalistic study
DOI: 10.1016/j.jth.2023.101649
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 association between drivers' stress levels, emotions, and specific driving context factors to identify environmental triggers for negative states. While previous research has focused on detecting driver stress, there is a gap in understanding the specific environmental causes—such as road objects or traffic dynamics—that induce these states. Understanding these triggers is critical for designing interventions to improve road safety and human-centered vehicle systems. The authors hypothesize that specific environmental events, such as encountering large vehicles or intersections, and changes in car-following distance are significantly correlated with increases in heart rate (HR) and negative facial emotions. The researchers utilized the HARMONY naturalistic driving dataset, which includes multimodal data from 15 participants, comprising in-cabin and on-road video recordings and psychophysiological metrics collected via smartwatches. To analyze the driving context, the study employed computer vision techniques to detect seven categories of environmental perturbations: stop signs, speed limit signs, traffic signals, trucks, buses, motorcycles, bicycles, and pedestrians. For sign detection, a YOLO V5 model was retrained on merged datasets (LISA, COCO, and Balali et al.) and integrated with Optical Character Recognition (OCR) to distinguish specific sign types. Driver stress was quantified by detecting abrupt increases in HR using a Bayesian Change Point detector, while facial emotions were analyzed using the Affectiva module to measure "valence" (positivity/negativity) and "engagement." The study also estimated the relative distance to lead vehicles to assess the impact of car-following behavior. The results indicate that different road objects are associated with varying levels of stress and emotional response. Larger vehicles, specifically trucks and buses, were associated with the highest increases in drivers' HR and the highest proportions of negative facial emotions. Shorter distances to the lead vehicle and higher standard deviations in car-following distance were correlated with a higher frequency of abrupt HR increases, suggesting elevated stress levels. Conversely, driving at higher speeds, typical of highway environments, was associated with more positive emotions, lower facial engagement, and fewer abrupt HR changes. These findings suggest that highway driving, when avoiding certain stress-inducing objects, promotes a calmer driver state compared to urban environments with complex traffic interactions. The significance of this research lies in its objective, data-driven identification of stress triggers in real-world driving conditions, moving beyond subjective self-reports. By linking specific environmental attributes to physiological and emotional responses, the findings can inform the design of personalized driving interventions, such as route planning that avoids high-stress objects or adaptive car-following assistance. Furthermore, these insights are valuable for automated driving systems, potentially improving take-over control mechanisms by anticipating driver stress levels based on the surrounding context.
Key finding
Larger vehicles on the road, such as trucks and buses, are associated with the highest increases in drivers' heart rate and negative emotions, while shorter distances to lead vehicles correlate with abrupt stress-related heart rate spikes.
Methodology
naturalistic
Sample size: 15
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 | — | — | — | 1 | 2026-05-07 |
| archive | success | canonical_url | — | — | 7 | 2026-06-06 |
| extract | success | cached | — | — | 3 | 2026-06-10 |
| clean | success | clean | — | — | 1 | 2026-06-04 |
| chunk | success | chunk | — | — | 1 | 2026-06-04 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-04 |
| enrich | success | — | — | — | 1 | 2026-05-07 |
| promote | success | — | — | — | 1 | 2026-05-07 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 2 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 15 | 2026-06-11 |
| verify | partial | — | — | — | 2 | 2026-06-10 |
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.
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: physiological data, behavioral performance data
- Methodological Resource: dataset resource