Driver emotion recognition framework based on electrodermal activity measurements during simulated driving conditions
DOI: 10.1109/iecbes.2016.7843475
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 persistent issue of road accidents in Malaysia, where accident rates and associated financial damages remain high despite advancements in vehicle safety technologies like traction control and anti-sleep systems. The authors identify the driver as the most fragile component of the road-vehicle-driver system, suggesting that human emotional states, specifically stress and anger, are primary contributors to accidents. The research aims to develop a driver emotion recognition framework using electrodermal activity (EDA) measurements to detect these critical emotional states during simulated driving conditions. To achieve this, the researchers conducted an experiment involving 20 subjects who participated in a simulated driving assignment designed to elicit specific emotional responses. The simulation included preset scenarios intended to induce neutral, stress, and anger states. EDA signals were recorded from the participants during these scenarios. The raw EDA data underwent preprocessing, including bandpass filtering between 0.5 Hz and 2 Hz, followed by a short-time Fourier transform. From this transformed data, the mean, median, and variance of the power spectral density were extracted as key features. These parameters were then subjected to statistical analysis using a two-sample f-test to determine significant differences between the emotional states. Additionally, the dataset was split into two equal groups of ten subjects for training and testing a support vector machine (SVM) classifier, utilizing a 10-fold cross-validation approach to assess classification performance. The results demonstrated statistically significant differences in EDA readings across the neutral-stress, neutral-anger, and stress-anger simulated driving scenarios, confirming that EDA is a viable indicator for distinguishing these emotional states. In terms of machine learning performance, the SVM classifier achieved an accuracy of 85% for distinguishing between neutral and stress states, as well as for distinguishing between neutral and anger states. However, the accuracy for differentiating between stress and anger states was lower, at 70%. These findings indicate that while EDA is effective for detecting the presence of negative emotions compared to a neutral baseline, distinguishing between specific negative emotions like stress and anger presents a greater challenge. The significance of this work lies in its contribution to the development of intelligent vehicle systems that can monitor driver well-being in real-time. By validating EDA as a reliable metric for detecting stress and anger, the study supports the integration of emotion recognition frameworks into modern vehicles. This could lead to proactive safety interventions, such as alerting drivers or adjusting vehicle systems when high-stress or anger states are detected, potentially reducing accident rates caused by human emotional factors. The study highlights the potential of physiological monitoring to address the human element in road safety, complementing existing technological improvements in vehicle stability and road infrastructure.
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 | OpenAlex-citations | — | — | 1 | 2026-06-25 |
| archive | success | semantic_scholar | — | — | 6 | 2026-06-26 |
| extract | success | cached | — | — | 2 | 2026-06-26 |
| clean | success | clean | — | — | 1 | 2026-06-25 |
| chunk | success | chunk | — | — | 1 | 2026-06-25 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-25 |
| promote | success | — | — | — | 1 | 2026-06-25 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-26 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-25 |
| verify | success | — | — | — | 1 | 2026-06-26 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; 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: physiological data
- Methodological Resource: validation psychometrics, tool software