Machine Learning(s) in gaming disorder through the user-avatar bond: A step towards conceptual and methodological clarity

Vasileios Stavropoulos; Maria Prokofieva; Daniel Zarate; Michelle Colder Carras; Rabindra Ratan; Rachel Kowert; Bruno Schivinski; Tyrone L Burleigh; Dylan Poulus; Leila Karimi; Angela Gorman-Alesi; Taylor Brown; Rapson Gomez; Kaiden Hein; Nalin Arachchilage; Mark D Griffiths

doi:10.1556/2006.2024.00063

Machine Learning(s) in gaming disorder through the user-avatar bond: A step towards conceptual and methodological clarity

J Behav Addict. 2024 Nov 22;13(4):894-900. doi: 10.1556/2006.2024.00063. Print 2024 Dec 30.

Authors

Vasileios Stavropoulos¹, Maria Prokofieva², Daniel Zarate¹, Michelle Colder Carras³, Rabindra Ratan⁴, Rachel Kowert^{5

6}, Bruno Schivinski⁷, Tyrone L Burleigh¹, Dylan Poulus^{8

9}, Leila Karimi¹, Angela Gorman-Alesi¹⁰, Taylor Brown¹, Rapson Gomez¹, Kaiden Hein¹, Nalin Arachchilage¹¹, Mark D Griffiths¹²

Affiliations

¹ 1Department of Psychology, Applied Health, School of Health and Biomedical Sciences, RMIT University, Australia.
² 2Victoria University, Australia.
³ 3Bloomberg School of Public Health, John Hopkins University, Baltimore, USA.
⁴ 4Michigan State University, USA.
⁵ 5Psychology Research Institute, Faculty of Social Studies, Masaryk University, Brno, Czech Republic.
⁶ 6Department of Communication, University of Münster, Münster, Germany.
⁷ 7School of Media and Communication, RMIT University, Australia.
⁸ 8Physical Activity, Sport and Exercise Research Theme, Faculty of Health, Southern Cross University, Gold Coast, Australia.
⁹ 9Manna Institute, Southern Cross University, Gold Coast, Australia.
¹⁰ 10Catholic Care Victoria, Australia.
¹¹ 11School of Computing, RMIT University, Australia.
¹² 12International Gaming Research Unit, Psychology Dept, Nottingham Trent University, UK.

Abstract

In response to our study, the commentary by Infanti et al. (2024) raised critical points regarding (i) the conceptualization and utility of the user-avatar bond in addressing gaming disorder (GD) risk, and (ii) the optimization of supervised machine learning techniques applied to assess GD risk. To advance the scientific dialogue and progress in these areas, the present paper aims to: (i) enhance the clarity and understanding of the concepts of the avatar, the user-avatar bond, and the digital phenotype concerning gaming disorder (GD) within the broader field of behavioral addictions, and (ii) comparatively assess how the user-avatar bond (UAB) may predict GD risk, by both removing data augmentation before the data split and by implementing alternative data imbalance treatment approaches in programming.

Keywords: classification; diagnosis; gaming disorder; machine learning; user-avatar bond.

MeSH terms

Avatar
Humans
Internet Addiction Disorder*
Machine Learning*
Supervised Machine Learning
User-Computer Interface
Video Games