What the data actually shows
The clearest finding concerns interruption, not anxiety directly. Sophie Leroy's research on "attention residue" showed that when people switch from one task to another, part of their attention stays stuck on the previous task, degrading performance on the new one. A notification is an interruption that triggers exactly this: even a glance leaves residue, so the cost of an alert is not just the seconds spent reading it but the lingering drag on whatever you return to.
There is also the phenomenon of anticipatory or habitual checking — reaching for the phone in expectation of an alert, sometimes before any has arrived. A related, well-documented experience is the "phantom vibration," the sensation that your phone buzzed when it didn't; surveys of phone users have found a large majority report having felt it at least occasionally. It is generally read as a sign of how primed people become to detect and respond to alerts, rather than as a disorder in itself.
On the anxiety link specifically, the evidence is more mixed. Some experimental work suggests that batching notifications or going without them can reduce stress and improve mood and attention for some people, while other studies find effects that are small or depend heavily on the person and the type of alert. The honest reading is that constant interruption and constant availability are plausibly arousing and stressful, but the research base is largely correlational and the effect sizes are modest, so notifications are better described as one contributor to feeling on edge than as a proven cause of anxiety.
Why this feels different from how it actually is
It feels like notifications barely cost anything because each one is tiny — a glance, a few seconds. What's hidden is the residue: the part of your attention that stays snagged on the interruption after you've put the phone down. The felt cost is one alert; the real cost is the fragmented attention that follows, which is much harder to notice from the inside.
Constant availability also feels normal because it arrived gradually, but it represents a genuine shift — for most of human history, you were not reachable by everyone at all times. Being persistently reachable keeps a low-level expectation running in the background that something may need your response, and that mild, sustained readiness is exactly the kind of arousal people often experience as a vague unease rather than as a clear cause they can name.
And anticipatory checking feels like your own free choice each time, which obscures how habitual it has become. The phantom vibration is the clearest tell that the habit has its own momentum: the impulse to check can fire even with no actual signal, because you've become finely tuned to detect one.
The felt cost is one alert; the real cost is the fragmented attention that follows — much harder to notice from the inside.
What the research says to do about it
The intervention with the most consistent support is reducing interruptions rather than reducing the phone entirely: turning off non-essential notifications, batching the rest to set times, and using do-not-disturb during focused work. Because the documented cost is interruption and attention residue, fewer interruptions is the lever that most directly targets the mechanism.
Creating predictable check-in windows can reduce anticipatory checking, because much of the pull comes from not knowing whether something is waiting. When checking happens on a schedule you set rather than on every buzz, the background "is there something I need to see" tension has fewer openings to fire.
For the arousal side, ordinary nervous-system basics apply — and this page is educational only, not medical advice. If anxiety is persistent and affecting your daily life, the right step is a qualified clinician, not a notification setting. Managing alerts may make the day feel calmer, but it is not a treatment for an anxiety disorder, and treating it as one can delay help that works.
What the research says does not help
Simply trying to ignore notifications while leaving them switched on tends not to work, because the cost is paid the moment your attention is pulled — even an alert you choose not to act on has already interrupted you and left residue. The decision to ignore comes after the interruption, not instead of it.
Assuming notifications are the sole cause of feeling anxious is also a mistake, and a potentially harmful one. Anxiety has many sources, and pinning it entirely on the phone can mean overlooking sleep, stress, health, or circumstances that matter more — and can delay seeking help that actually addresses it.
Going fully offline in dramatic, short bursts — a one-day cleanse followed by a return to the same settings — rarely produces lasting change, because it doesn't alter the steady stream of interruptions you come back to. The research favours changing the default settings over occasional total abstinence.
Notifications are better described as one contributor to feeling on edge than as a proven cause of anxiety.
What this looks like in real life
The alert you chose to ignore still cost you
A buzz arrives mid-task; you glance and decide not to act on it. It feels free. But the cost was already paid the moment your attention was pulled — attention residue means part of your focus stays snagged on the interruption after you put the phone down. The decision to ignore comes after the interruption, not instead of it, which is why simply ignoring notifications while leaving them switched on tends not to work.
The buzz that never happened
You feel your phone vibrate, reach for it, and there's nothing there — a phantom vibration, which a large majority of regular phone users report having felt. It's generally read not as a disorder but as a sign of how primed people become to detect and respond to alerts. The impulse to check can fire even with no actual signal, because you've become finely tuned to detect one.
Real numbers in context
There is no clean "notifications cause X% more anxiety" figure, and you should be wary of anyone who offers one. What is reasonably established is the interruption cost: switching tasks leaves attention residue, so the price of an alert extends well beyond the moment of reading it. Surveys on phantom vibrations have repeatedly found that a large majority of regular phone users — often reported in the range of two-thirds or more — have experienced the sensation of a buzz that never happened, which illustrates how primed people become to detect alerts.
On the wellbeing side, experimental studies that reduce or batch notifications have reported improvements in attention, productivity, and mood for some participants, but the effects are generally modest and vary by person. The fair conclusion is that interruption and constant availability plausibly raise low-grade arousal and stress, while the direct causal link to clinical anxiety remains uncertain — context that argues for managing notifications without over-claiming what doing so will fix.