The UniSQ study, an observational analysis of 1,766 participants from the Longitudinal Study of Australian Children using instrumental variable mediation, reveals that frequent social media use correlates with three times higher odds of cannabis experimentation, mediated roughly 25% by externalizing behaviors like impulsivity rather than internalizing distress. This observational design, lacking RCT controls, cannot prove causation but strengthens inference by isolating effects from confounders such as phone access. The finding challenges assumptions in prior work, such as a 2023 Addictive Behaviors review linking social media to substance use primarily via mental health pathways, and echoes NIDA's longitudinal data on adolescent brain vulnerability to reward-driven platforms. Missed in initial coverage is the policy timing with Australia's under-16 ban, which may overlook how platform algorithms amplify peer normalization of risk-taking over simple screen limits. Parents should prioritize self-regulation training and offline engagement, as intensity of use—not mere access—drives outcomes.