The average person now spends between seven and ten hours a day on screens. That is not a statistic about leisure. It is a description of the primary environment in which human cognition now operates.

The mental health consequences of this shift are beginning to be understood with more precision. A 2025 study published in Nature: Humanities and Social Sciences Communications found significant increases in anxiety, depression, and conduct problems in US children between 2016 and 2020, a period that maps closely onto the mass adoption of smartphones and algorithmically curated content. CDC data from the same period found that roughly one in four teenagers with four or more hours of daily screen time reported symptoms of anxiety or depression in the preceding two weeks. These are not marginal associations. They are patterns large enough to show up in nationally representative data.

Understanding why requires looking at what is happening inside the brain, how platforms are designed to exploit it, and what this means at a population level for those building in mental health.

What screens are doing to the brain

The neuroscience here has sharpened considerably in recent years. A 2025 editorial published in Cureus by neurologist Shaheen Lakhan draws on emerging functional neuroimaging studies to argue that heavy social media use, particularly among adolescents, is linked to measurable functional and structural changes in brain regions governing emotional regulation, impulse control, and social cognition. The neural patterns, he argues, resemble those seen in addiction, ADHD, and mood disorders.

The mechanism is well documented. Social media platforms are engineered to trigger dopamine release in the striatum, the same pathway activated by substance use. Notifications, likes, and the variable reward structure of an infinite scroll are not accidental design choices. They are the architecture of compulsion. A systematic review found that prolonged exposure leads to reward desensitisation: the brain requires more stimulation to register the same response, making ordinary life feel flat and unrewarding by comparison.

Lakhan introduces a term worth sitting with: digital anhedonia. The diminished capacity to find pleasure in real-world experiences after prolonged digital saturation. It is not addiction in the traditional sense. It is something quieter and, in some ways, harder to treat. A person who cannot feel fully present in an unmediated moment. A threshold for stimulation that the physical world can no longer meet.

Research published in JISTM (2025) adds a structural dimension. Studies using neuroimaging have found correlations between excessive screen time and reductions in grey matter volume in areas responsible for attention and decision-making. These are not temporary functional shifts. They are changes to architecture.

How platforms are built to keep you there

The neuroscience does not arise in a vacuum. It is the downstream consequence of product decisions.

A 2025 paper in Perspectives in Public Health identifies dopamine-scrolling as a distinct behavioural pattern driven by a specific set of design features: auto-play, infinite scroll, pull-to-refresh, algorithmic suggestions, and short-form video. Each of these features is designed to reduce friction, extend sessions, and reward continued engagement. The paper's authors describe the outcome as fragmented attention at population scale, with most teenagers now reporting being almost constantly online.

The World Economic Forum's 2025 analysis uses direct language: manipulative algorithms and addictive design features are not enhancing human potential. They are exhausting it.

This is the context in which digital mental health products are being built and used. The same devices, the same behavioural loops, the same reward architecture. A CBT-based app and a dopamine-maximising social platform share a substrate. Founders building therapeutic tools are working inside an environment that has been deliberately engineered to compete for the same attention their product requires.

A 2024 scoping review in PLOS Digital Health examined persuasive design frameworks in mental and behavioural health applications specifically, and found that deploying certain persuasive design strategies can yield counterproductive outcomes, particularly when the design principles borrowed from consumer apps are applied without clinical grounding.

What this looks like at population level

The clinical picture that emerges from this research is not uniform. A British Journal of Clinical Psychology study published in 2025 found that the relationship between screen time and mental health outcomes is stronger for maladaptive social media use than for passive consumption or gaming, and more pronounced in adolescent females. The evidence on total screen time and mental health outcomes remains heterogeneous, with associations that are real but smaller than public debate often implies.

What is clearer is the mechanism: problematic patterns of engagement, not raw hours, drive the most significant outcomes. This matters for how the problem is framed and addressed.

What builders need to take from this

The digital environment is not a neutral delivery channel. It has a neurological fingerprint. Products built for mental health are being used by brains that have been shaped, in measurable ways, by the platforms that preceded them. That is not a reason for fatalism. It is a reason for rigour.

The design decisions that maximise engagement are not the same as the decisions that support wellbeing. A meta-analysis of 92 RCTs of mental health apps published in npj Digital Medicine in 2025 found limited evidence that persuasive design features drive clinical efficacy, despite their prevalence in the field. Engagement and therapeutic benefit are not synonyms. A product that keeps people coming back is not automatically a product that helps them.

The people arriving at mental health tools are often already experiencing the attentional and emotional consequences of digital overstimulation. Design that adds to that burden, even in small ways, is not neutral. And the concept of digital anhedonia deserves clinical attention: a user who cannot feel present, who finds real-world interaction flat, who requires escalating stimulation to register reward, is not simply disengaged. They are presenting with something that has a neurobiological basis.

Building carefully in this environment is harder than it looks. It is also more important than the industry has so far acknowledged.

🪺