From social media to artificial intelligence: improving research on digital harms in youth

Abstract

In this Personal View, we critically evaluate the limitations and underlying challenges of existing research into the negative mental health consequences of internet-mediated technologies on young people. We argue that identifying and proactively addressing consistent shortcomings is the most effective method for building an accurate evidence base for the forthcoming influx of research on the effects of artificial intelligence (AI) on children and adolescents. Basic research, advice for caregivers, and evidence for policy makers should tackle the challenges that led to the misunderstanding of social media harms. The Personal View has four sections: first, we conducted a critical appraisal of recent reviews regarding effects of technology on children and adolescents' mental health, aimed at identifying limitations in the evidence base; second, we discuss what we think are the most pressing methodological challenges underlying those limitations; third, we propose effective ways to address these limitations, building on robust methodology, with reference to emerging applications in the study of AI and children and adolescents' wellbeing; and lastly, we articulate steps for conceptualising and rigorously studying the ever-shifting sociotechnological landscape of digital childhood and adolescence. We outline how the most effective approach to understanding how young people shape, and are shaped by, emerging technologies, is by identifying and directly addressing specific challenges. We present an approach grounded in interpreting findings through a coherent and collaborative evidence-based framework in a measured, incremental, and informative way.