Improved calibration of neurophysiological measures tools

Abstract

To carry out more reliable UX tests, it is necessary to analyze how neurophysiological measurements are treated. In particular, the use of baseline, the neutral task that compares a subject to itself, has so far been the subject of little study in the literature. The purpose of this paper is to focus on using the baseline in a UX test context. Our main result is that we can improve the baseline procedure, which currently consists of doing one baseline task at the beginning. Indeed, in our research, the validity of this procedure was studied during the calibration of the EDA (Electrodermal Activity). We then discovered that a majority of our subjects had an increasing or decreasing trend throughout the experiment. To answer this problem we propose to move the location of this one baseline or add a second baseline at the end of the experiment.