James One was designed and released to be used by panelists. It has been doing great for the last 8 months having acquired biometric data in thousands of study participations.

It measures heart rate – the number of times your heart beats per minute – an important index in the study of emotions. In addition, James One can also extract other cardiovascular and respiratory metrics which can be even more informative than heart rate. But let’s leave this topic for later. The other biometric signal collected by James One is galvanic skin response (GSR), also known as electrodermal activity (EDA). EDA has been studied for more than 150 years, and is the most reliable indicator of emotional arousal, but it is not widely understood and rarely assumes relevance in applied setups. Here we uncover some of its potentials…

From a purely scientific perspective, EDA measures hand palm perspiration and reflects responses of the autonomic nervous system (ANS) to timely auditory, visual or somatosensory stimuli (e. g.  a sound or a picture).

Let’s be truly honest here: a definition such as this does not convince anyone unfamiliar with psychophysiology to adopt EDA recordings into their applications! We hope this post helps you better understand the capabilities and potential of EDA for media testing.

Understanding electrodermal activity might be useful!

EDA is usually divided into tonic (EDL – electrodermal level) and phasic (EDR – electrodermal response) phenomena. As illustrated in the picture below, the human skin is a complex structure. Although a thorough description of skin anatomy and physiology is far beyond the scope of this text, we’ll try to highlight the most important parts.

The main skin functions are to protect the organism from external threats such as temperature, chemical, mechanical and infectious agents and provide sensory information that can be understood as touch, temperature and/or pain. Among the several organs involved, sweat glands control skin moisture and are involved in temperature regulation.

Skin can be roughly divided into 2 main layers: the epidermis, which is a superficial layer composed of epithelial cells that become drier as they approach the skin surface; and the dermis, which is a deeper layer mainly comprised of fibrous tissue. The secretory part of sweat glands is located between the bottom part of dermis and hypodermis. The latter is an even deeper tissue and is mostly comprised of fat, blood vessels, and connective tissue.

                                                                                                                              © 2014 WEBMD, LLC. ALL RIGHTS RESERVED

There are two kinds of sweat glands: apocrine, which are mostly concentrated on armpits and perianal areas, and eccrine glands, which are spread over the whole human body, with the highest concentration on the palms of the hand and the soles of the feet. The last two are known to be optimal skin regions to measure emotional sweating. While just a few human apocrine glands are innervated by the ANS, the human skin has many sympathetic fibers (the sympathetic nervous system is part of the ANS) that innervate the eccrine glands. Considering that the sole of the feet is not one of the most comfortable places to collect measurements regularly, we guess you got the reason why we ask panellists to use James One on the palm of their hands.

How does MindProber look into EDA?

MindProber evaluates pictures, videos and sounds that brands, content producers or broadcasters want to be as emotionally engaging as possible to clients, viewers or fans. Thus, EDA is collected from dozens or hundreds of panellists while they are being exposed to the content, and panel-wise EDA records (with some data preprocessing steps in between) are shown. As in the example below, recently published during Halloween season, an electrodermal response (EDR) can be seen following the famous highly engaging moment in which Jack Torrance (from ‘The Shining’) appears through the bathroom door with his paranoid face. As expected, the EDR starts some seconds after the emotionally evoking video frames. The identification and the amplitude of the most emotionally engaging moments in a content are then used by brands, producers, and broadcasters to edit their content and make it even more impactful.

But we can do more. We have actually shown that by adding some smart AI to EDA signals, we can even build predictive models for audience disengagement during drama series. That’s right: EDA predicts audience ratings!