This study tested the Spectrum hypothesis of problematic online behaviors (POBs) which posits that POBs (e.g., online gaming, gambling, sexual activities, and social networking) constitute distinct, yet related constructs. To overcome the limitations of previous research, we recruited a representative sample of young men and capitalized on a common assessment approach (Fifth version of the Diagnostic and Statistical Manual of Mental Disorders [DSM-5] and Eleventh version of the International Classification of Diseases [ICD-11] frameworks). The study was a single-center, observational study with a cross-sectional design (n = 2620). All targeted behaviors were assessed using the proposed criteria of the DSM-5 and the ICD-11. The correlation analysis showed that online activities were weakly correlated (0.01 ≤ r ≤ 0.19), suggesting that when people engage in one specific online activity to a problematic extent, they do not necessarily engage in several online activities in an excessive way. Using a network analysis, we showed that various POBs might reflect distinct constructs. The community detection analysis identified four communities of items corresponding to each POB. Our robust methodology and sampling strategy suggests that the construct of “Internet addiction” could be misleading. This epistemological shift might be necessary to develop adequate clinical care and public health measures to address POBs.