In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy
will defend his dissertation
Integrating Mobile Psychometrics with Wearable Physiological Sensing in Longitudinal Studies: Design, Testing, and Clinical Benefits
Traditionally, psychometric questionnaires have been given in paper form. Increasingly, questionnaires are delivered through web interfaces. Although, this is an improvement, web interfaces are not as flexible as mobile interfaces. We focus on the latter, because their ubiquity, can potentially cure the adherence problem. Indeed, in longitudinal studies, participants may be asked to fill out questionnaires several times a day over a period of time. Several problems arise in such studies, with adherence being the number one problem.
In this research we chose circumplex, a powerful but complex psychometric instrument, as a case study for evaluating the effect of mobile interfacing. Circumplex is a 2D psychometric, requiring careful consideration of design issues, as one tries to make it fit in a small smartphone screen. Therefore, we first ran a user interface study where we measured the goodness of several designs in the lab. Having selected the best designs out of the process, we ran a field study, to evaluate the goodness of these select designs in actual practice over a number of days. The next step was to use the winning design in a large longitudinal study with n=131 participants.
We have analyzed the data, keeping a keen eye on the issue of adherence. Initial results are promising. We have also collected concomitantly with the psychometric scores, physiological markers. These markers are electrodermal activity signals recorded by a wireless wrist band on the subject’s dominant hand. The ultimate goal is to mix mobile psychometrics with wearable sensing, thus delivering sustained multimodal responses that will paint a much more complete picture with respect to legacy approaches. These legacy approaches were not only suffering from adherence problems with regard to psychometric responses, but it was unthinkable to include complementary physiological responses, due to technical limitations.
Date: Friday, July 21, 2017
Time: 12:00 PM
Place: HBS 302
Advisor: Dr. Ioannis Pavlidis
Faculty, students, and the general public are invited.