Search for a command to run...
Digital environments have become an integral part of our lives, with the global average screen time fast approaching 7 h per day. This rapid shift has sparked ongoing debates regarding its consequences on mental well-being. Existing research on this topic has produced inconsistent findings, partly because it attempts to answer an oversimplified question of whether digital environments are “good” or “bad” for mental health. This mini review aims to examine why such a simplistic binary perspective is inadequate. We conducted a narrative literature search of major scientific databases covering studies published between 1990 and 2025 and analysed the evidence using a tripartite Dose–Content–Disposition (DCD) framework. Dose refers to the quantity and pattern of digital exposure, content to the qualitative nature of digital environments, and disposition to the users’ personality traits and characteristics. Our analysis reveals that high dose is not inherently harmful; rather, its impact is moderated by content (e.g., active versus passive engagement, socially connective versus comparative material) and disposition (e.g., vulnerability factors such as low self-esteem or pre-existing mental health conditions). The reviewed literature suggests that the mental health impact of digital environments is neither linear nor uniform but is shaped by a complex interplay between these DCD variables. This framework provides a tool to analyze the mixed results in the field and also highlights the need for more nuanced research methods, individualized analytical approaches, and ethical, well-being-oriented design principles. Future research should prioritize longitudinal and experience-sampling methodologies, identification of user profiles, and interdisciplinary collaborations to incorporate mental health considerations into platform architecture. As a mini review, this article is limited by its non-systematic approach, which may introduce selection bias, and the inherent challenges of drawing definitive conclusions from methodologically diverse studies.