Related Experiment Video
Updated: Feb 24, 2026
07:15
Tactile Vibrating Toolkit and Driving Simulation Platform for Driving-Related Research
Published on: December 18, 2020
5.1K
A multimodal dataset for various forms of distracted driving
Salah Taamneh1, Panagiotis Tsiamyrtzis2, Malcolm Dcosta3
1Computational Physiology Laboratory, University of Houston, Houston, Texas 77204, USA.
Scientific Data
|August 16, 2017
Summary
This study presents a multimodal dataset from 68 volunteers driving under various distractions. The data aids research into distracted driving behaviors and physiological responses, crucial for understanding car crashes.
Area of Science:
- Human-computer interaction
- Cognitive psychology
- Automotive engineering
Background:
- Distracted driving is a major cause of road accidents.
- Understanding the impact of different distractions on driving behavior is critical.
- Multimodal physiological data can provide insights into driver state.
Purpose of the Study:
- To create a comprehensive multimodal dataset of driving behavior under controlled distraction.
- To enable research into the effects of cognitive, emotional, and sensorimotor distractions.
- To facilitate studies on physiological responses during driving and startle events.
Main Methods:
- Controlled driving simulator experiment with 68 participants.
- Acquisition of multimodal data including vehicle dynamics and physiological signals.
- Four distraction conditions: no distraction, cognitive, emotional, and sensorimotor.
- Inclusion of a startle stimulus (unintended acceleration) under varying distraction levels.
Main Results:
- Continuous recording of key response variables (speed, steering, lane position) and explanatory variables (EDA, heart rate, facial expressions, eye tracking).
- Dataset captures nuanced driving behaviors across different distraction types.
- Physiological data correlated with driving performance under stress.
Conclusions:
- The dataset provides a valuable resource for studying distracted driving and its physiological correlates.
- Enables research into driver behavior under abstracted distracting stressors.
- Applicable for physiological channel benchmarking and multispectral face recognition.
