Overview
Most research on digital media and mental health measures problematic use through checklists of addictive symptoms. This study takes a different angle: it focuses on how adolescents perceive themselves as addicted to social media or gaming. That self-perception is a distinct psychological experience — more common than clinical symptom thresholds, and potentially important in its own right as something that shapes how young people think about and respond to their own behaviour over time.
Using five waves of data collected over one year from the ABCD cohort, we examined whether perceived addiction predicted subsequent emotional problems (depression and anxiety), and whether the reverse was also true. We also looked at whether these patterns differed between boys and girls. A statistical approach called random-intercept cross-lagged panel modelling was used, which separates stable personality-level differences between people from actual change happening within the same person over time.
What we found
The main finding is that, within a person, perceived addiction and emotional problems did not consistently predict each other over time — even though the two were clearly related across individuals. In other words, when a young person's sense of being addicted went up, their depression or anxiety did not reliably follow, and vice versa.
Across people, those who reported higher perceived social media addiction also tended to report more depression and anxiety on average. But this appears to reflect stable differences between people rather than a dynamic back-and-forth over time. Perceived gaming addiction showed no such cross-person associations.
These findings raise interesting questions. Perceived addiction to social media may be more tied to who someone is — stable traits, their social context, how they tend to interpret their own behaviour — than to how their mood fluctuates from month to month. This is a meaningfully different picture from what symptom-based measures show, and highlights the value of studying these constructs separately.
Study design
Data came from the ABCD cohort study, a large Dutch study that has followed participants from birth. Starting in 2021, we used five survey rounds collected three months apart over the course of one year, giving us a detailed picture of how things changed within each person over time — rather than just a one-off snapshot.
Participants were asked about depression and anxiety symptoms using established questionnaires, and about perceived addiction to social media and gaming using a simple, direct self-report question. The study was preregistered before data collection began, and ethics approval for the ABCD cohort was granted by the Medical Ethics Review Committee at the Academic Medical Centre Amsterdam.
What this means
Self-perceived addiction to social media and gaming appears to be relatively stable over time and largely independent of month-to-month changes in depression and anxiety. This matters practically: programmes targeting emotional problems as a route to reducing perceived addiction may not produce the expected change at the individual level. The cross-person associations, however, remain useful for identifying groups who may benefit from support. Future research should look more closely at what drives the sense of being addicted — whether it reflects stable personality traits, appraisal styles, the social environment, or something else entirely.