Taipei, Dec. 18 (CNA) A Taiwanese research team has developed a machine-learning model that analyzes electroencephalography (EEG) brain-wave patterns to distinguish individuals with internet addiction from healthy subjects with 86 percent accuracy, researchers announced Thursday.
The method's accuracy is significantly higher than that of self-report measures, Huang Hsu-wen (黃緒文), an assistant investigator at the National Health Research Institutes' National Center for Geriatrics and Welfare Research and one of the study's lead researchers, told a press event.
After analyzing 92 participants' (including 42 with internet addiction and 50 healthy controls) resting state EEG functional connectivity, the researchers found the addicted group showed elevated levels of phase synchronization, Huang said.
She believed it was because addiction disrupted neural systems in the inhibitory and reward pathways.
Huang said that changes in EEG patterns occur before addictive behaviors manifest, meaning that EEG testing combined with machine-learning classification models could identify early risk signals more efficiently and enable schools and medical institutions to intervene with greater precision.
Internet addiction refers to prolonged online engagement, inability to curb the urge to go online and discomfort when disconnected from the internet, according to the research article.
The article was published in Psychological Medicine, an international medical journal, in May 2025.
Other contributors to the research included Wu Shun-chi (吳順吉), a professor in the Department of Engineering and System Science at National Tsing Hua University; Huang Chih-mao (黃植懋), an associate professor in the Department of Psychology at the University of Hong Kong; as well as research institutions in Taiwan and overseas.
