Meta recently agreed to pay up to $17.1 billion in a settlement to resolve claims from 47 states that its social media products harmed children’s mental health. Importantly, as part of that agreement, Meta also agreed to implement a variety of product restrictions such as mandatory usage pauses, nighttime blocks, school time notification restrictions, and age assurance.
This settlement constitutes the most sweeping set of social media limits thus far applied on any major platform in the United States. Some will cheer that development; others will oppose it. Either way, legislators who have spent years looking for a template will treat these new rules as the new baseline that can be used to push laws that expand these terms to everyone.
Whether the Meta settlement was the right remedy is a different question than where we go from here. What is clear is that the broader and soon-to-be more aggressive push to regulate the entire social media industry remains supported by research that is inconclusive, mixed, and methodologically weak.
For years, policymakers have ignored, missed, or misunderstood that research. They often treat the academic literature as settled science and the harms as generally applicable. Table 1 presents 28 different studies that do not fit that story: studies that call for more research, educational programs, or otherwise recommend caution over bans and restrictions.
Table 1: Studies That Caution Against Social Media Regulation or Recommend More Research
Table 1: Studies That Caution Against Social Media Regulation or Recommend More Research by Pablo Garcia QuintDesigning policy that is not aligned with, or does not account for, the recommendations of the entire body of evidence on social media is a big issue. Not all papers in Table 1 suggest that social media does not cause harm. However, they do caution against social media regulation as the obvious next step. Two reviews conclude that bans are ineffective. Many of the papers that find only small—or even beneficial—effects also warn against regulation. One study finds that the effects are person-specific, not population-wide, which suggests broader regulation may be the wrong remedy.
That research debate is only half the problem. Researchers must also contend with the policy consequences of the policymaking that research inspires: age verification downsides, the simplicity of circumventing bans, and just general ineffectiveness. Those choices have consequences—for privacy, speech, and safety.
Fortunately, there is a better way forward. Unlike the mixed and weak social media literature, there is wider consensus on how school start times, family history of mental health problems, and socioeconomic status negatively affect mental health in teens. So do poorly designed academic standards: stress from chronic overload and high-stakes grade competition that can create a general fear of failure. None of that is an argument against quality standards, which are worth preserving. But taken too far, those demands risk imbalance that can lead to burnout and mental health issues.
Sleep time and removing distractions during school, both targeted in the settlement, hold promise to actually improve child mental health. Research psychologists Amy Orben and Andrew Przybylski have argued for years that sleep is a more important determinant of mental health than social media. And the developmental and quantitative psychologist Candice Odgers explained in a recent TED Talk that more attention should be placed on academic pressure and family environments, not on social media bans.
Theoretically, pauses on notifications during school hours could help relieve some peer pressure, and blocking social media apps during sleeping hours is a reform that could improve sleep. While these are important factors to consider, it is not clear from the academic literature that social media, rather than tech devices more generally, plays a major contributing role in either.
For sleep time, research has looked at long trends for adolescents and has been unable to establish a large causal effect of social media on how much children sleep. And for social media or phone use during school, researchers have not been able to prove their effectiveness. Perhaps these changes will lead to better studies and improved outcomes—but the current research does not support that thesis.
The bottom line: policymakers should spend more time on policy solutions to improve sleep, school start times, family conditions, and socioeconomic status. That means spending less time treating social media regulation as a silver bullet.
Pablo Garcia Quint is a technology policy analyst with the American Consumer Institute, a nonprofit education and research organization. Follow Pablo on X @gq_pablo