We tend to talk about attention as though it exists entirely inside the person. If you can concentrate, you have focus; if you become distracted, you lack discipline. And when your phone keeps pulling you away from something you intended to do, the obvious explanation is that you failed to exercise enough self-control. There is some truth in that account, but it leaves out an important part of the problem: attention never operates in isolation. It operates inside an environment.
Those environments are not neutral. They determine what is visible, what announces itself, what requires effort, and what continues indefinitely unless we deliberately stop it. That does not remove personal agency, but it does mean focus cannot be understood entirely as a personality trait or an act of will. Smartphones make the point clearly: the same architecture that removes friction from things we intend to do also removes friction from things we did not particularly intend to do.
The environment is already part of the decision
Consider a sequence you may recognize. You finish writing an email and have not yet decided what to do next. Your phone is beside you, so you pick it up. It unlocks almost instantly. You notice a notification, open one application, then another, and several minutes later realize you never intended to be there at all. We often describe that sequence afterward as though it contained a clear decision: I decided to check my phone instead of working. Sometimes it does. But often the more interesting question is where the deliberate decision actually occurred.
Research on habitual behavior helps explain why that can be difficult to identify. Behaviors repeated in stable contexts become increasingly associated with the cues surrounding them, so familiar circumstances can trigger familiar responses without the conscious deliberation that was present when the behavior was new. Wendy Wood and her colleagues demonstrated this pattern in everyday behavior, and later work has emphasized how recurring environmental cues activate well-learned responses.
Recent smartphone-specific evidence points the same way. In a 2025 field study combining mobile sensing, app logs, and map-based questionnaires, smartphone habits were stronger in the spaces people travelled and visited most habitually. The study does not prove that location causes smartphone use, but it does show that smartphone habits and recurring spatial contexts interact in everyday life.
This does not mean people lose control of their behavior. It means intention is only one of the forces acting on it. Context, convenience, and cues all matter, and it is when the three of them line up — the phone within reach, already unlocked, already holding something new — that the next action takes almost no motivation at all. Notifications report a change, badges report something waiting, frequently used applications stay logged in, and personalized feeds remove the work of deciding what to look at next. Each feature is small, but together they create an environment in which many behaviors have almost no entry cost.
Friction is not always something to eliminate
Good design usually tries to remove friction, and for good reason. We want the map to open quickly when we are lost, and the banking application to remember enough about us that checking a balance does not become an administrative project.
The behavioral consequence of convenience, however, is that easier actions require less motivation to perform. B.J. Fogg’s 2009 behavior model made this explicit by treating motivation, ability, and a trigger as conditions that converge when a behavior occurs. When an action becomes easier, the threshold for performing it becomes lower, and that holds regardless of whether the behavior is one we later judge to be useful. Opening a social application, switching from email to video to a news feed, finding something novel: each costs almost nothing, so an impulse that might have faded if it required a little effort becomes behavior almost immediately.
This suggests that friction is not inherently bad. Sometimes it is exactly what we need. Moving an application off the home screen, logging out, or leaving the phone in another room during concentrated work does not make the behavior impossible. It simply inserts another step between impulse and action, and that step creates enough space for a question that might otherwise go unasked: Did I actually intend to do this?
Good environmental design, then, is not merely the art of making desirable actions easy. It also involves making unwanted automatic actions just difficult enough that they become decisions again.
Attention also needs endings
Ease helps explain why behaviors begin, but not why certain digital activities are difficult to leave. For that, we need stopping points. Older media imposed visible boundaries. A newspaper article ended, a television episode reached the credits, a page of search results required another click. None of those forced anyone to stop, but each created a moment in which continuation required a new action. Infinite scroll removes that moment, because there is no page two and no final item. Research on infinite scrolling has found that users describe becoming caught in loops that extend sessions beyond what they intended, with the reason for stopping usually arriving from outside the application rather than from any boundary within it.
A stopping cue does not need to make us stop in order to be useful. Its function is to return continuation to the realm of choice. Autoplay reverses that relationship: instead of choosing whether to begin the next episode, the user chooses whether to stop it from beginning. The content is the same, but the default has changed.
Boundaries also help people judge what counts as a complete unit of activity, an effect documented in classic “unit bias” research on portion size — though that work concerns food, not feeds. When digital systems remove those boundaries, users may need to restore them deliberately: turning off autoplay, setting a defined period, or deciding what “finished” means before opening an endless feed.
The interruption can happen before you respond
Notifications reveal something counterintuitive: successfully resisting an interruption does not mean the interruption was free. In one controlled experiment, participants performed an attention-demanding task while phone notifications arrived. They did not interact with the device, yet performance still declined when notifications occurred. The significance of that finding is not that every vibration causes major cognitive impairment. It is that the attentional event can begin before the behavioral decision to pick up the phone.
If the only problem were whether we responded, becoming better at ignoring alerts would solve most of it. But if the signal itself recruits attention, the more useful question is whether it needed to arrive at that moment at all.
A later field experiment makes the practical implications more nuanced. Researchers altered notification delivery over a two-week period. People who received notifications in three predictable batches per day reported improvements in attentiveness, perceived control, and stress compared with people receiving them normally. Yet eliminating notifications entirely did not produce an even better outcome; participants in that condition experienced more anxiety and fear of missing out.
The lesson is not that everyone should receive exactly three notification batches per day. It is that notification architecture matters. There is a meaningful difference between information being available when we decide to check it and information repeatedly deciding when we should attend to it. That suggests a standard: immediate access to our attention should be treated as a privilege rather than a default application setting. A call from a family member, a security alert, or a time-sensitive work message may deserve it. A retailer announcing a sale probably does not.
Screen time hides the question that matters
Once attention is understood this way, the familiar focus on total screen time looks incomplete. A screen-time report can say how long a device was in use. It cannot say whether those hours were spent talking to family, working, watching something deliberately chosen, or moving between applications without any clear purpose. A 2020 critical analysis of the screen-time literature pointed to inconsistent definitions, inconsistent measurement, and mixed findings, and research examining individual smartphone sessions suggests that purpose matters more than duration: habitual use to pass the time, along with some entertainment and passive social-media use, was associated with a lower sense of meaningfulness. The more useful distinction is between intentional and automatic use, which is the subject of the companion note, Why Screen Time Metrics Are Not Adequate Measures of Attention.
Design for the kind of attention the work requires
The larger principle extends beyond technology. Different activities place different demands on attention, so there is no single ideal attentional environment. Routine administrative work may tolerate interruptions reasonably well; writing a difficult argument may not. Learning unfamiliar material often requires enough continuity to encounter confusion and work through it.
If a task requires sustained attention, protecting that attention should be part of preparing to do the work rather than something we hope to accomplish through force of will once the work has begun.
That might mean:
- Closing unrelated applications
- Placing the phone elsewhere for forty-five minutes
- Limiting which people or services can interrupt a focus period
- Or creating a clear stopping rule before entering an open-ended digital environment
None of those changes is especially dramatic, and that is partly the point. They reduce the number of moments in which self-control has to solve a problem that could have been prevented by design. We often respond to distraction by asking how to become more disciplined. A better question comes first: have we created conditions in which discipline is being asked to do unnecessary work?
Attention is never perfectly stable. Minds wander, people get tired, and sometimes changing tasks is the right decision. The goal is not to become impossible to distract. The goal is to preserve enough agency over when attention moves, because learning difficult things and producing work of substance both require periods in which attention stays available long enough for something to develop.
Increasingly, those conditions cannot simply be assumed. They have to be designed.
References
Wood, W., Quinn, J. M., & Kashy, D. A. (2002). Habits in everyday life: Thought, emotion, and action. Journal of Personality and Social Psychology, 83(6), 1281–1297. https://doi.org/10.1037/0022-3514.83.6.1281
Wood, W., & Neal, D. T. (2007). A new look at habits and the habit–goal interface. Psychological Review, 114(4), 843–863. https://doi.org/10.1037/0033-295X.114.4.843
Ross, M. Q., Rhee, L., Le, H., Mount, J., Chang, Y.-J., & Bayer, J. B. (2025). Smartphone habits are stronger in spaces chosen out of habit. Scientific Reports, 15, 41252. https://doi.org/10.1038/s41598-025-25174-2
Fogg, B. J. (2009). A behavior model for persuasive design. Proceedings of the 4th International Conference on Persuasive Technology, Article 40, 1–7. https://doi.org/10.1145/1541948.1541999
Rixen, J. O., Meinhardt, L.-M., Glöckler, M., Ziegenbein, M.-L., Schlothauer, A., Colley, M., Rukzio, E., & Gugenheimer, J. (2023). The loop and reasons to break it: Investigating infinite scrolling behaviour in social media applications and reasons to stop. Proceedings of the ACM on Human-Computer Interaction, 7(MHCI), Article 228, 1–22. https://doi.org/10.1145/3604275
Geier, A. B., Rozin, P., & Doros, G. (2006). Unit bias: A new heuristic that helps explain the effect of portion size on food intake. Psychological Science, 17(6), 521–525. https://doi.org/10.1111/j.1467-9280.2006.01738.x
Stothart, C., Mitchum, A., & Yehnert, C. (2015). The attentional cost of receiving a cell phone notification. Journal of Experimental Psychology: Human Perception and Performance, 41(4), 893–897. https://doi.org/10.1037/xhp0000100
Fitz, N., Kushlev, K., Jagannathan, R., Lewis, T., Paliwal, D., & Ariely, D. (2019). Batching smartphone notifications can improve well-being. Computers in Human Behavior, 101, 84–94. https://doi.org/10.1016/j.chb.2019.07.016
Kaye, L. K., Orben, A., Ellis, D. A., Hunter, S. C., & Houghton, S. (2020). The conceptual and methodological mayhem of “screen time.” International Journal of Environmental Research and Public Health, 17(10), 3661. https://doi.org/10.3390/ijerph17103661
Lukoff, K., Yu, C., Kientz, J., & Hiniker, A. (2018). What makes smartphone use meaningful or meaningless? Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 2(1), Article 22, 1–26. https://doi.org/10.1145/3191754
