Back to search

Hellenic Open University

Human-computer interaction: evaluating the user experience of interactive information systems based on bio-signals analysis

Abstract

dc:description

Stress constitutes a problem of growing importance in our society. Although it may have positive effects on certain aspects of life (eustress), the majority of users of interactive systems have associated stress with negative experiences (distress). It has been proved that prolonged exposure to stressful life conditions (e.g. poverty, family problems, job strain etc.) reduces users’ productivity significantly and affects their overall health. More specifically, it is estimated that the cost of work-related stress is € 617 billion per year (at European level). Therefore, it is important to identify stress periods during users’ interaction with a system in a reliable and efficient way. The present doctoral thesis proposes the development of the REACTION methodology. This methodology is supported by an observation analysis software named PhysiOBS and can be applied during the evaluation of user experience (User eXperience, UX). In particular, REACTION focuses on the automatic detection of problematic interaction periods (design and/or operational errors) associated with stress by using bio-signals (e.g. skin conductance, heart rate, skin temperature etc.). Bio-signals have been adopted by many researchers and practitioners during UX evaluation in order to address the challenge of an efficient and effective way to detect stress. The objectivity of bio-signals (the typical user cannot easily manipulate this type of data, as they could possibly do with facial expressions) and the ability to record them in the background of an interaction are the two key features which, in combination with the evolution of recording sensor technology, have contributed to their wider acceptance. Conversely, the use of bio-signals requires specialized knowledge on data analysis, special handling of the recording equipment (e.g. proper placement), and appropriate experimental conditions in order to collect reliable datasets. In addition, their recording should be executed by users who meet the necessary conditions for participation in the experimental process (e.g. the measurement of skin conductivity should not be applied to someone who has a medical condition related to sweating). Skin conductivity is a reliable indicator of stress and it is the signal on which this research was based. More specifically, skin conductivity represents the ability of the skin to conduct electricity as a result of the sympathetic activation of the sweat glands of the skin (in each activation of the sweat glands, the dermal conductivity of the skin increases). Apart from skin conductance, additional data sources, such as user videos and comments - data recorded during the interaction - can also be deployed by the UX evaluator in the context of REACTION implementation, thus contributing to a multifaceted evaluation approach. In the context of the present doctoral thesis, a series of experiments were conducted with the aim of creating an appropriate stress recognition mechanism through bio-signals as well as the operation and the integration of the above mechanism in PhysiOBS. The results of the research showed that the proposed evaluation method REACTION can reliably identify periods of stress. REACTION can be applied either independently or in combination with other traditional UX evaluation methods (e.g. user observation, speech subject protocol). It should be noted that this thesis does not posit to replace traditional UX evaluation methods with REACTION. On the contrary, REACTION is proposed as a supplementary technique allowing researchers and practitioners to be more effective while managing the available UX data sources.

Degree

thesis:*
Grantor dc:publisher
Hellenic Open University
Year dc:date
2020

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Liapis, Alexandros
  • Λιάπης, Αλέξανδρος

Subjects

dc:subject × 22

Rights

dc:rights
Statement dc:rights
  • BY_NC_ND
Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/47774

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Liapis, Alexandros; Λιάπης, Αλέξανδρος. Human-computer interaction: evaluating the user experience of interactive information systems based on bio-signals analysis. Hellenic Open University, 2020. http://hdl.handle.net/10442/hedi/47774