Classification of Gamers Using Multiple Physiological Signals: Distinguishing Features of Internet Gaming Disorder

Effects of Game-Related Tasks for the Diagnosis and Classification of Gaming Disorder

Gaming disorder (GD) is an addictive behavior characterized by an insatiable need to play video games and shares similar symptoms with the failure of self-control due to a decline in cognitive function. Current GD diagnostic and screening tools rely on questionnaires and behavioral observations related to cognitive functions to assess an individual's capacity to maintain self-control in everyday life. However, current GD screening approaches rely on subjective symptoms, and a reliable diagnosis requires long-term clinical follow-up. Recent studies have measured biosignals along with cognitive functional tasks to provide objectivity to GD diagnosis and to acquire immediate results. However, people with GD are hypersensitive to game-related cues, so their responses may vary depending on the type of stimuli, and the difference in response to stimuli might manifest as a difference in the degree of change in the biosignal. Therefore, it is critical to choose the correct stimulus type when performing GD diagnostic tasks. In this study, we investigated the task dependence of cognitive decline in GD by comparing two cognitive functional tasks: a continuous performance task (CPT) and video game play. For this study, 69 young male adults were classified into either the gaming disorder group (GD, n = 39) or a healthy control group (HC, n = 30). CPT score, EEG signal (theta, alpha, and beta), and HRV-HF power were assessed. We observed differences in the left frontal region (LF) of the brain between the GD and HC groups during online video game play. The GD group also showed a significant difference in HF power of HRV between CPT and online video gaming. Furthermore, LF and HRV-HF significantly correlated with Young's Internet Addiction Test (Y-IAT) score, which is positively associated with impulsivity score. The amount of change in theta band activity in LF and HRV-HF-both biomarkers for changes in cognitive function-during online video game play suggests that people with GD express task-dependent cognitive decline compared with HC. Our results demonstrate the feasibility of quantifying individual self-regulation ability for gaming and underscore its importance for GD classification.

Multiple-Kernel Support Vector Machine for Predicting Internet Gaming Disorder Using Multimodal Fusion of PET, EEG, and Clinical Features

Internet gaming disorder (IGD) has become an important social and psychiatric issue in recent years. To prevent IGD and provide the appropriate intervention, an accurate prediction method for identifying IGD is necessary. In this study, we investigated machine learning methods of multimodal neuroimaging data including Positron Emission Tomography (PET), Electroencephalography (EEG), and clinical features to enhance prediction accuracy. Unlike the conventional methods which usually concatenate all features into one feature vector, we adopted a multiple-kernel support vector machine (MK-SVM) to classify IGD. We compared the prediction performance of standard machine learning methods such as SVM, random forest, and boosting with the proposed method in patients with IGD (N = 28) and healthy controls (N = 24). We showed that the prediction accuracy of the optimal MK-SVM using three kinds of modalities was much higher than other conventional machine learning methods, with the highest accuracy being 86.5%, the sensitivity 89.3%, and the specificity 83.3%. Furthermore, we deduced that clinical variables had the highest contribution to the optimal IGD prediction model and that the other two modalities were also indispensable. We found that more efficient integration of multimodal data through kernel combination could contribute to better performance of the prediction model. This study is a novel attempt to integrate each method from different sources and suggests that integrating each method, such as self-administrated reports, PET, and EEG, improves the prediction of IGD.