Electrodermal activity measurements: a review of artifacts

Electrodermal activity (EDA) is a general term used to indicate changes in the electrical parameters of the skin reflecting sympathetic nervous system activity. EDA has been widely used in various fields; however, the reliability of high-quality EDA measurements is frequently compromised by artifacts and environmental factors, particularly for wearable instruments. This review provides a comprehensive examination of the major sources of artifacts in EDA measurements and discusses approaches to mitigate their effects. To the best of our knowledge, this is the first review to address all sources of artifacts in EDA recordings alongside potential strategies for their removal. Physiological artifacts (respiration, movement, and speech), technical artifacts (electrode and operating systems), and environmental influences (temperature, humidity, sound noise, and air pressure and altitude) are the main sources of EDA signal changes. Various approaches and tools are recommended to detect and mitigate the effects of these artifacts. Some of them work for laboratory-controlled measurements, but are difficult to meet with ambulatory measurements. Therefore, the reliability of EDA measurements obtained from wearable devices continues to be challenged by artifacts. Ongoing developments in data processing, filtering, and instrumentation should address these concerns in EDA measurements.