Database Credentialed Access

Electrodermal Activity of Healthy Volunteers while Awake and at Rest

Sandya Subramanian Patrick Purdon Riccardo Barbieri Emery Brown

Published: Aug. 26, 2021. Version: 2.0


When using this resource, please cite: (show more options)
Subramanian, S., Purdon, P., Barbieri, R., & Brown, E. (2021). Electrodermal Activity of Healthy Volunteers while Awake and at Rest (version 2.0). PhysioNet. https://doi.org/10.13026/136t-7g98.

Please include the standard citation for PhysioNet: (show more options)
Goldberger, A., Amaral, L., Glass, L., Hausdorff, J., Ivanov, P. C., Mark, R., ... & Stanley, H. E. (2000). PhysioBank, PhysioToolkit, and PhysioNet: Components of a new research resource for complex physiologic signals. Circulation [Online]. 101 (23), pp. e215–e220.

Abstract

This dataset of electrodermal activity was collected from 11 healthy volunteer subjects who were awake and at rest in seated position. The activity was recorded from each subject's non-dominant hand for one hour at 256 Hz. Electrodermal activity measures changing electrical conductance of the skin as an indicator of sweat gland activity. Sweat glands are a primitive part of the fight-or-flight response. These data were collected as part of a larger study to understand and build computational models for autonomic nervous system activity (including electrodermal activity) with approval from the Massachusetts Institute of Technology Committee on the Use of Humans as Experimental Subjects (COUHES).  The second dataset of electrodermal activity was collected from 11 healthy volunteer subjects who were under computer-controlled propofol sedation. The activity was recorded from each subject's left hand for about 3 hours at 500 Hz. These data were collected with approval from the Massachusetts General Hospital Human Research Committee. 


Background

Electrodermal activity (EDA) is a direct read-out of body’s sympathetic nervous system measured as sweat-induced changes in the skin’s electrical conductance. It is controlled almost exclusively by the sympathetic branch of the autonomic nervous system as part of the "fight-or-flight" response [1]. The goal of our study was to understand the statistical properties of various physiological signals specifically with respect to autonomic nervous system control.

We seek to build computational models for these signals (including EDA) to accurately extract this autonomic information. To this end, we collected EDA data from 11 healthy volunteers while awake and seated at rest for one hour. We also collected data from 11 healthy volunteers (different volunteers) while under computer-controlled propofol sedation at increasing and then decreasing concentrations [7]. We would like to share these small databases and code to allow other researchers to also explore and understand EDA data and the rich information it contains. There is growing interest in using EDA to track physiological conditions such as stress levels, sleep quality, and emotional states.


Methods

With the approval of the Massachusetts Institute of Technology (MIT) Institutional Review Board, we collected EDA data from 12 healthy volunteers between the ages of 22 and 34 while awake and at rest. We used the FDA-approved Neurofeedback System developed by Thought Technology Ltd. Electrodes were connected to the second most distal phalange of the second and fourth digits of each subject’s non-dominant hand.

Approximately one hour of EDA data was collected at 256 Hz. Subjects were seated upright and instructed to remain awake. They were allowed to read, meditate, or watch something on a laptop or tablet, but not to write with the instrumented hand. We assumed skin and ambient temperature were constant for the duration of the experiment.

One subject's data were not included in the database because they informed us after the data collection that they occasionally experience a Raynaud's type phenomenon. This would affect the quality of their EDA data. Data from the remaining 11 subjects are included in this database without any further processing.

Code is also included to analyze these data, including detecting and removing artifact, separating the slow-moving (tonic) and faster (phasic) components of the data, extracting 'pulses' from the data representing discrete sweat release events, and analyzing the goodness-of-fit of the extracted pulses for various statistical models [2,3,4,5,6]. 

For the second dataset, with the approval of the Massachusetts General Hospital (MGH) Human Research Committee, EDA data were collected from 11 healthy volunteers while under computer-controlled propofol sedation. The BedMaster system was usedElectrodes were connected to each subject's left hand. Approximately three hours of EDA data were collected at 500 Hz. Skin and ambient temperature were assumed to be constant for the duration of the experiment.

We have also included the separated phasic and tonic components, and extracted pulse times and pulse amplitudes for each subject in both datasets.


Data Description

The data were collected with approval from the Massachusetts Institute of Technology Committee on the Use of Humans as Experimental Subjects (COUHES) and the Massachusetts General Hospital Human Research Committee. All subjects provided written informed consent. The dataset comprises:

  • 22 distinct subjects
  • For the at rest dataset, per subject: 
    • Single electrodermal activity (EDA) time series
    • Phasic EDA data
    • Tonic EDA data
    • Extracted EDA pulse times
    • Extracted EDA pulse amplitudes
    • Each time series is one hour of data at 256 Hz
  • For the propofol sedation dataset, per subject:
    • Extracted EDA pulse times
    • Extracted EDA pulse amplitudes
    • Single electrodermal activity (EDA) time series for 3 subjects
    • Phasic EDA data for 3 subjects
    • Tonic EDA data for 3 subjects
    • Each time series is approximately 3 hours of data at 500 Hz

A simple set of code has been provided for processing the at rest data files:

  • 4 files of Matlab code
  • Main file: eda_analysis.mat

Usage Notes

For the at rest dataset, the data are purely observational; no specific stimuli were applied. We did not control what the subjects were reading or watching during the data collection. For the propofol sedation dataset, the propofol concentration was changed over time and is described in [7].

Format: 

  • The datasets are separated by folder, 'At-Rest' or 'Propofol-Sedation'.
  • The raw data files are named eda_##.csv. The subject order is randomized.
  • The phasic and tonic EDA data files are named phasic_##.csv and tonic_##.csv respectively.
  • The pulse times and pulse amplitude files are named pulse_times_##.csv and pulse_amps_##.csv respectively.
  • Each data file is a .csv file that contains a single column of EDA values sampled at 256 Hz for the at rest dataset and 500 Hz for the propofol sedation dataset.
  • Each data file comprises roughly 60 minutes of data for the at rest dataset and between 2 and 3 hours of data for the propofol sedation dataset. This was the recording length for each subject.

Code format:

  • The main file to run for the at rest dataset is eda_analysis.mat (the user can modify the subject number and whether or not figures should be returned).

Release Notes

This is the second release of this database. The first release only included one dataset and the raw data alone, while this one includes two and processed data as well.


Acknowledgements

We would like to thank the Massachusetts Institute of Technology Clinical Research Center Staff. This work was partially funded by funds from the Picower Institute for Learning and Memory, the National Science Foundation Graduate Research Fellowship Program, and National Institutes of Health Award P01-GM118629.


Conflicts of Interest

Patent filed 07-15-2020 (PCT/US2020/042031).


References

  1. W. Boucsein, Electrodermal Activity. (Springer, 2012).
  2. Y. Pawitan, In All Likelihood. (Clarendon Press, Oxford, 2013).
  3. E. Brown, R. Barbieri, V. Ventura, R. Kass, L. Frank, The time-rescaling theorem and its application to neural spike train data analysis. Neural Comput. 14, 325–346 (2001).
  4. D. Daley, D. Vere-Jones, An Introduction to the Theory of Point Processes: Volume II: General Theory and Structure. (Springer, 2007).
  5. S. Subramanian, R. Barbieri, E. N. Brown, Point process temporal structure characterizes electrodermal activity. Proceedings of the National Academy of Science. (2020).
  6. S Subramanian, PL Purdon, R Barbieri, EN Brown. A Model-Based Approach for Pulse Selection from Electrodermal Activity. IEEE Trans Biomed Eng. 2021 Apr 6. doi: 10.1109/TBME.2021.3071366.
  7. PL Purdon et al., Electroencephalogram Signatures of Loss and Recovery of Consciousness from Propofol. PNAS. 110, E1142-1151 (2013).

Share
Access

Access Policy:
Only credentialed users who sign the specified DUA can access the files.

License (for files):
PhysioNet Credentialed Health Data License 1.5.0

Corresponding Author
You must be logged in to view the contact information.
Versions
  • 1.0 - Aug. 4, 2020
  • 2.0 - Aug. 26, 2021

Files