Electrodermal Activity as a Digital Biomarker

Electrodermal activity is the variation in the electrical conductance of the skin, caused by sweat gland activity under sympathetic nervous system control. It is one of the few measures that reflects sympathetic activity directly, without the parasympathetic contribution that complicates heart rate based measures, which is why it has a long history in psychophysiology.

The signal has two components. Tonic skin conductance level is the slow moving baseline, and phasic skin conductance responses are the rapid transient rises that follow a stimulus. They are analysed separately because they behave differently and mean different things.

The important caveat is the one the field states itself: electrodermal activity indexes arousal, and arousal is non specific. It rises with heat, exercise, effort and surprise as readily as with distress, so labelling it emotion attributes a specific cause to a signal that cannot distinguish causes.

Measurement passes a small current between two electrodes on the skin and records conductance. Laboratory work uses the palm or fingers, where sweat gland density is highest, which is impractical for daily wear. Wearable devices place electrodes on the wrist, where the signal is weaker and more affected by movement and electrode contact.

Published recommendations for electrodermal measurement specify electrode placement, gel, sampling and reporting, and adherence to them is what makes results comparable between studies. Systematic review of collection and signal processing innovations documents how much variation still exists in practice.

Processing separates tonic from phasic components, removes motion artefact, and counts responses per minute. Ambient temperature must be recorded, because sweating responds to heat and a warm room will raise conductance regardless of psychological state.

The most concrete clinical application is not psychological at all. Electrodermal activity contributes to wearable seizure detection, where a large sympathetic surge accompanies generalised convulsive seizures, and combining it with accelerometry improves detection over movement alone. That is the application with the strongest evidence and the clearest regulatory path.

In research it is used in stress and arousal studies, in autonomic function assessment, and increasingly in mental health work as a physiological correlate collected alongside validated questionnaires. It is reported with instruments such as the Perceived Stress Scale and the Hospital Anxiety and Depression Scale, precisely so that the physiological signal and the psychological state are recorded as separate things rather than conflated.

Wrist measurement is considerably noisier than palmar recording, and movement, electrode contact and ambient temperature all introduce variation that laboratory conditions control. A meaningful proportion of people show very few spontaneous responses, which limits what can be extracted from their data at all.

Specificity is the core limitation. The signal cannot distinguish psychological distress from physical exertion, heat or excitement, so any interpretation as emotion depends entirely on knowing the context, and free living recordings usually do not.

There is no standardised normal range, values are not comparable between devices or between people, and the measure is best used within a person against their own baseline under known conditions.

• Boucsein W, et al. Publication recommendations for electrodermal measurements. Psychophysiology. 2012. pubmed.ncbi.nlm.nih.gov
• Posada-Quintero HF, Chon KH. Innovations in electrodermal activity data collection and signal processing: a systematic review. Sensors. 2020. pubmed.ncbi.nlm.nih.gov
• Poh MZ, et al. Convulsive seizure detection using a wrist-worn electrodermal activity and accelerometry biosensor. Epilepsia. 2012. pubmed.ncbi.nlm.nih.gov
• Smets E, et al. Into the wild: the challenges of physiological stress detection in laboratory and ambulatory settings. IEEE J Biomed Health Inform. 2019. pubmed.ncbi.nlm.nih.gov