Colman Noctor: Protecting children means tackling social media's harms

Colman Noctor: A lot of harm social media causes stems from algorithmic recommendations. Picture: iStock 

The British government’s recent plans to restrict children’s access to social media ignited a familiar debate. On one side are those who argue that social media is causing significant harm to young people’s mental health and social development and needs to be banned. On the other side, others warn that blanket bans may create as many problems as they solve.

As an academic, a clinician, and a parent, I can see the argument from both sides.

Social media has transformed how children communicate, and many young people believe that online spaces are where friendships are maintained and communities are built.

Yet it is impossible to ignore that these same platforms expose children to harmful content and online exploitation, while also encouraging compulsive use. The challenge is what can be done about it.

The move towards social media bans reflects growing frustration that, despite years of highlighting the problems among young users, little or no progress has been made in addressing them. Well-intentioned recommendations that parents supervise more, schools educate more, and young people become more digitally resilient have proved futile against platforms whose profits rely on keeping users online for as long as possible.

It is understandable that governments are resorting to stronger measures.

However, Australia’s experience of implementing social media restrictions offers an important reality check. Six months on from the implementation of the ban, Australia’s eSafety Commission has found that the age-assurance systems of many platforms are ineffective, with significant numbers of underage users continuing to access services using VPNs (virtual private networks) to misrepresent their age.

This does not mean the policy is pointless. Rather, it highlights how difficult it is to regulate a digital environment where young people and profit-driven companies are more technologically adept than the adults seeking to regulate them.

Inconclusive evidence

Some academics have legitimate concerns that ‘bans’ oversimplify complex issues. Their arguments against bans focus on the limited evidence of their effectiveness and the lack of evidence that social media causes, rather than merely correlates with, mental health difficulties. They make the case for waiting until the evidence is conclusive.

However, parents cannot afford to wait while their children navigate a harmful environment that is evolving faster than research. While academics understandably seek certainty, governments and parents must often act in the face of uncertainty.

The decision to raise the minimum age risks exposing children who manage to bypass the restrictions to even more inappropriate online content than before, as it will be assumed they are over 16 years of age.

While these are valid concerns, to borrow a phrase from Dr Michael Ryan of the World Health Organization, during the covid pandemic, ‘perfection is often the enemy of the good’.

The fact that a social media ban may not work perfectly is not an argument for doing nothing. Every protective measure has limitations. Speed limits do not eliminate road accidents. Age restrictions do not stop all teenagers from drinking alcohol or vaping. Yet we implement these measures because reducing harm matters, even when eliminating it is impossible.

The question should not be whether a social media ban is perfect, but whether it will meaningfully reduce harm.

My central concern, however, is that we may be focusing on the wrong aspect of the issue. Many parents understandably focus on the harmful content on social media, but underestimate the impact of manipulating our attention. Yet, even if every piece of harmful content were removed from the internet tomorrow, we would still have the scenario of a 14-year-old spending nine hours a day watching videos of cats on skateboards.

There is nothing dangerous, offensive, or illegal about that content. No parent would be calling for it to be removed. Yet spending an entire day consuming an endless stream of ‘brain rot’ videos represents a very unhealthy use of a young person’s time and comes at great cost to their development.

So the issue is not always what children are watching; it is that they struggle to stop watching it, and what healthy activities they are not doing while they are watching it.

Modern social media platforms are not simply libraries of content; they are technological environments carefully designed to manipulate users’ attention.

Young people’s developing brains are particularly vulnerable to these manipulative algorithmic systems because the designers know that they are more sensitive to rewards, novelty and social approval. Most are unable to resist temptations engineered by teams of the world’s best behavioural scientists, data analysts, and software designers.

Remove the algorithm

A social media ban may reduce exposure, delay access, and create valuable developmental space for some young people. However, it will not teach them to manage their attention or disengage from a compulsive activity.

This is why I find myself increasingly drawn to a different solution.

Rather than focusing on banning social media, what if we insisted on removing the algorithmic component?

The algorithm is arguably the most powerful force in a child’s digital experience. It determines what they see, what they do not see, and what captures their attention. It learns their interests, exploits their vulnerabilities, and continuously manipulates their feed to deliver content designed to maximise and manipulate their engagement.

Two children can access the same platform, and one may be directed towards sport or music content, whereas the other may find themselves immersed in appearance-focused material and extreme content that hijacks their existing insecurities.

While the social media platform is the vehicle, the algorithm is the architect of dependency. And the longer the child spends on a platform, the greater the profits.

A lot of the online harm stems from recommendation engines, which learn what captures their attention and then relentlessly serve them more of it. They do not care whether that attention is for fitness videos, pornography, celebrity gossip, or cats on skateboards. Their objective is not wellbeing; it is retention.

Remove the algorithm, and social media looks very different. Endless scrolling becomes finite because the next video is no longer automatically generated, thereby reducing the endless stream of cats-on-skateboards described earlier. Turning the platform into a tool rather than a manipulation.

Would all problems disappear? Of course not.

Bullying would still exist, and harmful content would still be available, but it would have to be sought rather than delivered.

The current algorithmic system rewards outrage, comparison, and extremism because these factors keep users engaged.

Algorithms are not an accidental feature of social media; they are the engine of the product.

The industry’s extraordinary profitability depends on them.

Asking technology companies to abandon their algorithms could be like asking a casino to remove its slot machines, which is why government regulation that has teeth will be necessary. Not because bans are a perfect solution or because the evidence is complete, but because the risks facing children are immediate, and the regulation of social media companies is virtually nonexistent.

We should absolutely continue debating the merits of social media bans, scrutinise the evidence, learn from Australia’s experience, and remain cautious about simplistic solutions. However, we cannot make complexity an excuse for inaction.

The question is not whether children need protection online; it is whether we are willing to confront the systems that harm them.

A social media ban may not be the answer. They are difficult to enforce and easy to circumvent. Yet the growing appetite for such measures reflects a recognition that the current interventions are failing too many children.

Doing nothing is not a strategy.

Waiting for the perfect solution is not a strategy.

Trusting technology companies to place children’s wellbeing ahead of profit is definitely not a strategy.

But when it comes to our children’s digital lives, doing nothing makes us silent witnesses to the ongoing harm to their childhoods.

• Dr Colman Noctor is a child psychotherapist