UNIVERSE: UNobtrusIVE measuRement of mental workload and stress in uncontrolled environments
Authors/Creators
-
1.
Hasso Plattner Institute
Contributors
Data collector (3):
Other:
-
1.
Hasso Plattner Institute
Description
A Dataset on Unobtrusive Measurement of Cognitive Load and Physiological Signals (EEG, PPG, EDA) in Uncontrolled Environments
The dataset (approximately 315 hours in total) consists of physiological signals from wearable electroencephalography (EEG), electrodermal activity (EDA), photoplethysmogram (PPG), acceleration, and temperature sensors. The recorded dataset is curated from 24 participants following an eight-hour cognitive load elicitation paradigm. The mentioned consumer-grade physiological signals are obtained from the Muse S EEG headband and Empatica E4 wristband. The data is balanced across controlled and uncontrolled environments and high vs. low mental workload levels. During the study, participants worked on mental arithmetic, Stroop, N-Back, and Sudoku tasks in the controlled environment (roughly half of the data) and realistic home-office tasks such as researching, programming, and writing emails in uncontrolled environments. Data labels were obtained using Likert scales, Affective Sliders, PANAS, and NASA-TLX questionnaires. The completely anonymized data set and its publicly available features open a vast potential to the research community working on mental workload detection using consumer-grade wearable sensors. Among others, the data is suitable for developing real-time cognitive load detection methods, research on signal processing techniques for challenging environments, developing artifact removal techniques from low-cost wearable devices' data, or developing personal mental workload assistants.
Follow the dataset descriptor publication for more details:
Anders, C., Moontaha, S., Real, S., Arnrich, B.: Unobtrusive measurement of cognitive load and physiological signals in uncontrolled environments. Scientific Data 11(1), 1000 (2024).
Technical info
Due to its comprehensive nature, the dataset is uploaded in two .zip files, each containing data from 12 participants. UN_101, UN_102, and up to UN_124 refer to the participant IDs. To use the dataset, one can download each .zip file.
For example, each folder, UN_101, consists of three subfolders: Lab1, Lab2, and Wild. The first two refer to controlled sessions, and the latter refers to uncontrolled sessions. Each folder is divided into folders and files: Raw, Preprocessed, Labeled, Features, Notes.pdf, and Task_Labels.csv.
As the name suggests, the Raw folders contain raw data from two devices in two folders: Muse from the Muse S EEG headband and Empatica from the Empatica E4 wristband. Labeled data refers to the raw data split into different folders according to the timestamps of the labels. The names of each folder inside the Labeled folder for Lab1 and Lab2 indicate the defined low and high classes of the data based on tasks. For Wild folders, the Labeled folder names indicate the binary classes based on Likert scales of stress and mental workload given by the participants, and derived by applying a personalized threshold on the mean for the individual scales. Nevertheless, to extract the actual reading from different scales, the Task_Labels.csv file is provided in each folder. A multi-class classification model can be derived. The Preprocessed contains the labeled data after following extensively detailed preprocessing steps for the individual modalities. Features were extracted for 60-second windows derived from the preprocessed data.
For a more in-depth understanding of the preprocessing steps, refer to the respective publication. Furthermore, the GitHub repository is a valuable resource, offering codes for data loading, preprocessing, feature extraction, and a machine learning example method.
Files
UNIVERSE_UN_101_to_UN_112.zip
Files
(20.3 GB)
| Name | Size | Download all |
|---|---|---|
|
md5:64ff57fc16d9d7093bb1db9507662200
|
10.2 GB | Preview Download |
|
md5:9efe0aa556b6db4d83e94c07c5471fd1
|
10.1 GB | Preview Download |
Additional details
Software
- Repository URL
- https://github.com/HPI-CH/UNIVERSE
- Programming language
- Python , Jupyter Notebook
- Development Status
- Active