If teens can’t stop scrolling, should tech companies step in?

If we wouldn’t make seatbelt use optional for young adults, why do we let young people choose whether or not to limit their screen time?

Screen addiction has become a growing challenge of our times. Many use doomscrolling to self-soothe or unwind, but research suggests that too much screen time can be unhealthy.

Instagram CEO Adam Mosseri stated during a trial last month that only a small percentage of teenagers used Instagram’s “Take a Break” feature before the company made it a default setting for young users.

The feature, which encourages users to step away from Instagram after spending a set amount of time on the app, became a key point of discussion during an ongoing US trial examining whether Meta designed Instagram and Facebook in ways that could make children addicted to them.

“Most teens didn’t want it,” Mosseri said. “We decided to push forward with it anyway.”

This trial matters because concerns about teen development and screen addiction are now a big national issue. It also raises bigger questions: How much of the onus lies with teens? Is voluntary opt-out viable?

Dr. Christian Perrin, a psychology professor, at the University of Birmingham, says one of the main issues from the trial is how vulnerable teenagers are. Studies show that teens are especially sensitive to social rewards and feedback, while their ability to control impulses and make long-term decisions is still developing.

“That matters when those young people are using platforms built around likes, notifications, social comparison, and personalized recommendations,” Dr. Perrin says.

“Mosseri’s testimony illuminates a broader psychological problem with treating non-use of the ‘Take a Break’ feature as evidence that teenagers did not want it,” adds Dr. Perrin.

Behavioral research has long shown how strongly defaults can influence people’s choices. People tend to stick with options that have already been selected for them, particularly when changing those options requires effort. That effect can be even more relevant on platforms that are constantly competing for users’ attention.

Against that backdrop, Dr. Perrin says the low voluntary uptake of “Take a Break” is hardly surprising. An opt-in safeguard puts the burden on a young user to decide when to stop, while a default safeguard changes the environment around that decision.

The testimony also raises a broader question about where responsibility lies when a platform is designed to maximize engagement.

“Mosseri’s testimony reveals a telling imbalance. Features that keep teens scrolling- infinite scroll, autoplay, the like button, and algorithmic feeds, were on by default. Meanwhile, ‘Take a Break,’ meant to help teens step away, was optional, used by only 1–2% of teenagers, and not made the default until three years later, amid intensifying EU regulation and US lawsuits,” says Dr. Jais Adam-Troian, assistant professor at Heriot-Watt University Dubai.

“These elements could be interpreted as Meta knowingly designing platforms to keep young users engaged while offering protections it knew would go unused. This was only changed after the company faced legal pressure from regulators,” he adds.

NOT A NEUTRAL DESIGN CHOICE

When it comes to screen time, doomscrolling, and compulsive social media use, there are several layers of responsibility. Parents and schools have a role. Teenagers have agency. But the companies building these platforms make the decisions about how those environments work.

“Mosseri’s claim that ‘most teens don’t want it’ is precisely the point,” says Dr. Adam-Troian.

The American Psychological Association’s health advisory report explains that brain regions that drive the desire for peer attention and social reward become highly sensitive in early adolescence, while the regions responsible for self-control and impulse regulation continue to develop into adulthood.

“Expecting teenagers to voluntarily switch off features designed to exploit this developmental gap is not a neutral design choice; it shifts responsibility onto the very users least equipped to bear it,” says Dr. Adam-Troian.

Dr. Perrin argues that responsibility has to be considered alongside power.

“Responsibility should therefore be proportionate to power,” he says. Technology companies have access to vast amounts of behavioral data and control the architecture through which users interact with their products.

“A teenager deciding whether to stop scrolling and a company determining the conditions under which scrolling occurs occupy profoundly different positions of influence,” Dr. Perrin says.

Dr. Adam-Troian draws a comparison with gambling, where the effectiveness of voluntary safeguards has also been questioned.

“The online gambling industry faced a similar dilemma and learned the hard way that voluntary, opt-in tools consistently fail because those most at risk don’t use them,” he says, pointing to the 2024 Lancet Public Health Commission on gambling.

“‘Take a Break’ is Instagram’s version of a voluntary self-exclusion tool, and it failed for the same reason,” he says. “A business model that relies on maximizing engagement is not exclusive of built-in features to prevent or mitigate foreseeable problematic use for at-risk users.”

While the parent company, Meta, has denied the allegations and said its research found no clear link between adolescents’ social media use and a lack of well-being, the question of corporate responsibility remains central to the case.

DANGEROUS BY DESIGN?

Did Meta’s design decisions cross a line into knowingly creating risks for young users?

Perrin says that the question ultimately belongs to the courts. From a psychological perspective, however, he says the focus should be on what Meta knew, when it knew it, and what it did with that information.

“Companies understandably seek engagement with their products, and high engagement alone tells us relatively little about psychological harm,” Perrin says.

The more important evidence, he says, will be whether Meta’s own research identified patterns of problematic use or psychological risks among younger users and whether those findings influenced subsequent product decisions.

That puts the company’s internal research and communications under particular scrutiny. The timing of safety interventions and how those measures were implemented may also matter, Perrin says.

Dr. Adam-Troian has raised similar concerns about the gap between the way engagement features and safety tools were built into the platform.

Engagement-maximizing features were largely frictionless and switched on by default, he says, while “Take a Break” was optional and had very low uptake. He also points to evidence that Meta’s internal research had identified risks during the years the feature remained optional.

There are also questions about what the company knew internally compared with what it communicated publicly, including claims that more than 90% of users kept the feature on while only 1–2% of teenagers used it.

“The Lancet gambling framework applies here: when a company knows voluntary tools fail and still promotes them as protections while maintaining engagement-by-default design, the line between business decision and knowingly created risk narrows considerably,” Dr. Adam-Troian says.

For Perrin, transparency is another important part of the discussion. Social media companies have access to behavioral data on a scale that is difficult for independent researchers to replicate. When research points to potential risks for children and adolescents, he argues that those findings should be available for outside scrutiny.

The trial could therefore have implications beyond the question of Meta’s liability. It may also force a closer examination of what companies should disclose when their own research identifies potential risks to young users.

WHAT CHANGES?

The outcome of the trial could influence how technology companies approach safety features for younger users.

Dr. Adam-Troian points to other industries in which products changed after evidence of harm prompted regulatory or legal intervention. “Seatbelts weren’t voluntary,” he says. He also points to the American Psychological Association’s 2024 follow-up report, which argues that putting responsibility on parents, app stores, or young people does not address vulnerabilities built into platforms.

For social media, that could mean moving away from optional safeguards and towards protections that are built into products from the beginning.

Perrin sees another possible consequence: greater scrutiny of the research companies’ conduct behind closed doors. Greater access for independent researchers and more transparency around internal findings could help regulators, policymakers, and scientists better understand how platform design affects young users.

So, a debate that began with one Instagram feature is now much bigger: if a platform knows young users are vulnerable to its design, how much responsibility should it take for what happens next?