You pick up your phone to check one thing. Twenty minutes later, you are still there. Nothing of particular value has happened. You are not happier or more informed in any meaningful way. And yet something kept you.
That feeling — that your phone has a kind of superglue on it — is not a failure of willpower. It is the intended outcome of a design process that has been refined over decades specifically to produce it. Understanding how it works is the first step toward having a different relationship with your device. And for a significant number of people, it is also the first step toward understanding something more personal: why a habit that feels compulsive does not yield easily to ordinary self-discipline.
The Gambling Industry Did This First
The features that make social media apps difficult to put down did not originate in Silicon Valley. They were developed, iteratively and methodically, in the casinos of Las Vegas — beginning in the 1980s, when the casino industry replaced mechanical slot machines and physical card tables with digital versions.
The motivation was initially economic: digital machines were cheaper to maintain. But they also allowed casinos to add features — lights, sounds, animations — and more importantly, to run continuous large-scale experiments on millions of gamblers each year. Tweak the machine, measure how long people stay on it, keep the changes that increase engagement, repeat for decades.
The result, according to anthropologist Natasha Dow Schull, who spent 15 years researching machine gambling, was the most addictive form of gambling ever created. Some users stayed at machines for 24 or 48 hours without stopping. Schull documented casino workers reporting that machines had to be cleaned nightly after users — in the grip of what she called the "machine zone" — refused to leave even to use a bathroom.
This is not a metaphor. It is a description of what happens when a carefully engineered system successfully overrides the normal regulatory mechanisms of human attention and bodily awareness. And when tech companies built social media platforms, Schull found that they had, whether by design or by parallel discovery, replicated the same four features that produce this state.
The Four Features — and What They Do to Your Brain
Science journalist Michaeleen Doucleff, in her new book Dopamine Kids, breaks down what she calls the "superglue recipe": the four features that, combined, are most effective at producing what researchers call "dark flow" — a trance-like state of absorption that is distinct from healthy engagement and consistently leaves people feeling worse rather than better after.
1. Solitude. You use the app alone. There are no other people physically present to provide social cues — cues that, in ordinary life, help regulate behavior. When we are around other people, we pick up feedback about whether what we are doing seems normal, enjoyable, or excessive. That feedback is absent when we scroll alone in a bedroom or on a couch. Research has found that children who use screens alone are more likely to remain on an app even when it interferes with sleep or homework. The same principle applies to adults: physical solitude removes one of the most natural brakes on behavior.
2. Bottomlessness. There is no end to the content. No final page, no closing credits, no natural stopping point. Infinite scroll, autoplay, and algorithmically generated feeds ensure that something new always appears the moment the previous thing ends. As Schull describes it, there is no natural stopping point — and in the absence of one, the ordinary thought "maybe I should stop now" is immediately preempted by the next item appearing before the thought can complete itself. When social media companies introduced infinite scroll in the 2010s, time-on-app increased dramatically.
3. Speed. The faster the interaction, the longer people stay engaged. The gambling industry discovered this with slot machines — at peak speed, a player can run through more than a thousand games per hour, one every three seconds. The same principle operates in scrolling: the smoother and faster the content flows, the harder it becomes to locate a natural moment of pause. Speed, Schull suggests, contributes to a blurring of the boundary between self and screen — a sense of merger that is a hallmark of the machine zone state.
4. Personalized teasing. This is perhaps the most psychologically sophisticated feature, and the one that AI has dramatically amplified. The algorithm does not give you what you want. It gives you something close to what you want — and then, a few interactions later, something a little closer. Full satisfaction would end the session. The goal is not to satisfy but to sustain the feeling of being on the verge of satisfaction.
Neuroscientist Mateusz Gola at UC San Diego explains the mechanism: when people feel they are making progress toward a goal, dopamine release intensifies their motivation to continue. The feeling of getting closer — even when the goal is receding — activates the same neurochemical drive as actual progress. The app generates the sensation of approaching satisfaction without ever delivering it. That gap is where the hours disappear.
Dark Flow Is Not the Same as Healthy Absorption
It is worth pausing on a distinction that NPR's coverage of this research highlights, because it matters for how we evaluate our own relationship with screens.
Psychologist Mihaly Csikszentmihalyi described "flow" as a highly positive state of complete absorption in a challenging, meaningful task — playing an instrument, writing, solving a difficult problem, physical exertion at the edge of one's capacity. This kind of flow is associated with a sense of aliveness and competence, and it leaves people feeling satisfied and energized afterward.
Dark flow is structurally similar but experientially and neurologically different. It is produced by easy, repetitive, low-stakes interaction with an optimized system — not by genuine engagement with something demanding. And where healthy flow tends to leave people feeling good, dark flow consistently leaves people feeling flat, lethargic, and sometimes genuinely worse than before they began.
This distinction is clinically important, because people often experience both states as a kind of absorption and may not clearly distinguish between them in the moment. The question worth asking is not "was I engaged" but "how did I feel when I stopped, and did I stop because I chose to?"
What This Has to Do With Anxiety and Mental Health
The connection between heavy phone use and mental health struggles — particularly anxiety and depression — is well-documented, though the mechanisms are still being studied and the relationship is almost certainly bidirectional.
What the four-feature framework clarifies is one important pathway: the phone is not simply a distraction from anxiety. For many people, it has become a primary avoidance mechanism. Unpleasant emotions — boredom, loneliness, low-grade stress, the approach of a difficult task — reliably trigger phone use in a way that is functionally similar to other avoidance behaviors. The phone offers immediate relief from the discomfort of unstructured attention. And like other avoidance strategies, it works in the short term while maintaining and often deepening the underlying discomfort over time.
In clinical work, I often find that the question is not just how much someone is using their phone, but what they are using it instead of. What feelings are being interrupted? What gets picked up as soon as there is a pause — a moment in an elevator, a minute between tasks, the first quiet moment before sleep? The pattern of when phone use occurs is often more clinically revealing than the raw quantity.
The solitude feature is particularly worth noting in this context. The most psychologically costly phone use — the kind most strongly associated with worse mental health outcomes — tends to happen in the absence of other people, often late at night. This is the same time when, as I have written elsewhere on this blog, the brain's capacity for rational self-regulation is at its lowest and the default mode network has the most room to amplify worry. The phone, in this context, is not just distracting — it is feeding a system that is already primed toward rumination.
What You Can Actually Do
Knowing the mechanism does not automatically change behavior, but it changes the frame — and that matters. Struggling to put down your phone is not a character flaw. It is a reasonable response to a system that has been deliberately optimized to prevent you from putting it down. Naming it as such reduces the self-criticism that tends to accompany failed attempts at self-regulation and makes it easier to approach the problem practically.
A few things that research and clinical experience suggest are genuinely useful:
Add friction. The superglue recipe works by removing every obstacle between you and the next piece of content. Reversing this means deliberately adding obstacles back in. Paying per video rather than subscribing to unlimited streaming is one version of this. Keeping your phone in a drawer when you are home, and requiring yourself to go to the drawer to use it, is another. These feel cumbersome precisely because they are working against a design that has removed all friction. That friction is the point.
Use the phone around other people. The solitude feature is real. Physical presence of other people disrupts the conditions for dark flow. This does not mean you need to socialize every time you use your phone — it means that phone use in a coffee shop or a shared room is meaningfully different, neurologically, from phone use alone at 11pm.
Identify your trigger moments. Most compulsive phone use is not random. It is reliably preceded by specific emotional states — boredom, low-grade stress, the approach of something difficult, loneliness. Noticing your own pattern is the beginning of having a choice. Therapy is well-suited for this kind of pattern-mapping, particularly when the underlying feelings that are being avoided are worth examining in their own right.
Treat this as a systemic problem, not a willpower problem. If someone consistently struggles to disengage from a system that has been engineered to resist disengagement, the solution is to change the system — the physical availability of the phone, the subscription structures, the app configuration — rather than to rely on willpower to override it.
A Note on Children
Doucleff's book, and the research it draws on, is primarily oriented toward children and adolescents. This is appropriate: the developmental implications of early, heavy exposure to dark-flow-optimized systems are significant and warrant serious attention from parents and clinicians alike.
But the same mechanisms operate in adults. Two landmark legal cases in California in 2026 found that tech companies including Meta and Google had deliberately designed their apps to be addictive for younger users. The appeal process is ongoing, but the underlying science — the four features, the dark flow state, the gambling industry origins — is not in dispute. These systems were designed for maximum engagement without reference to user wellbeing. Adults are not immune to that design.
If you are an adult who recognizes your own phone use in this description — and most people reading this will — the same principles apply. The phone has superglue on it. That is not an accident. And acknowledging that is not surrender to it. It is the beginning of a more honest, and more effective, relationship with it.
Citations
Doucleff, M. (2026). Dopamine kids. [Publisher TK].
Gola, M., Wordecha, M., Sescousse, G., Lew-Starowicz, M., Kossowski, B., Wypych, M., Makeig, S., Potenza, M. N., & Marchewka, A. (2017). Can pornography be addictive? An fMRI study of men seeking treatment for problematic pornography use. Neuropsychopharmacology, 42(10), 2021–2031. https://doi.org/10.1038/npp.2017.78
Schull, N. D. (2012). Addiction by design: Machine gambling in Las Vegas. Princeton University Press.
Twenge, J. M., Joiner, T. E., Rogers, M. L., & Martin, G. N. (2018). Increases in depressive symptoms, suicide-related outcomes, and suicide rates among U.S. adolescents after 2010 and links to increased new media screen time. Clinical Psychological Science, 6(1), 3–17. https://doi.org/10.1177/2167702617723376
Doucleff, M. (2026, June 1). How your phone keeps you scrolling — even when you want to stop. NPR Short Wave. https://www.npr.org/2026/06/01/nx-s1-5823736/phone-social-media-addiction-tech