What a Two-Week Phone-Blocking Study Found About Attention

Two weeks without mobile internet nudged measured sustained attention up, by a small to modest amount. That is the short version of a 2025 randomized trial by Castelo, Kushlev, Ward, Esterman and Reiner in PNAS Nexus, which followed 467 iPhone users. The attention gain was real but modest, and most participants didn't fully stick to the block. Below: how the trial worked, the numbers, the caveats, and what it means for your own focus routine.

How the phone-blocking trial worked

The study was a preregistered randomized trial with a crossover waitlist design. Participants were recruited through Prolific, had a mean age of 32, and 63% were female. 242 were assigned to the intervention group and 225 to a delayed-intervention group, which blocked internet two weeks later. Surveys came at three points, two weeks apart.

The block itself was blunt. The Freedom app shut off all mobile internet on the phone, both Wi-Fi and mobile data, for two weeks. Calls and texts still worked, and participants could use the internet on other devices such as a laptop. So it wasn't a ban on phones. It was a ban on the phone as an internet device.

Attention was measured with the gradCPT, a five-minute go/no-go task. City scenes appear on 90% of trials and you press for each one; mountain scenes appear on 10% and you hold back. The primary score is d-prime, which captures how well you tell the two apart. It's an objective task, not a questionnaire, and that's why the result carries weight.

What the attention result does and doesn't show

In the intervention group, d-prime rose from 2.75 to 2.91 after two weeks of blocking. In the delayed group it rose from 2.81 to 3.00 during their own block. In effect-size terms (a within-person measure called dz), those are 0.24 and 0.28, usually read as small. Pooled across everyone assigned to a block, the effect was 0.23; among only the people classed as compliant, it was 0.26. Self-reported attention lapses also improved in both groups.

Attention wasn't the biggest winner. Mental health improved with an effect of 0.57 and subjective well-being with 0.46. The authors found that time offline, less media use, social connectedness, self-control and sleep helped explain the mental health and well-being gains. None of those explained the attention gain, so the study can't say why attention improved.

After the block ended, well-being and mental health stayed above their starting levels, while attention stayed at a similar level. Screen time tells the same story of drift: in the intervention group, self-uploaded iPhone data showed 314 minutes a day before, 161 during the block, and 265 afterward.

Caveats worth knowing

Adherence was low. Of the 467 people who committed, 119 (25.5%) were classed as compliant, and the paper's results and methods sections define compliance slightly differently. The authors note that this likely means the intention-to-treat effects are underestimated, but also that the compliant-only figures are less certain.

The sample was also motivated: 83% rated their wish to cut back at 5 or higher on a 7-point scale. Participants knew the study was about smartphones, so expectancy effects are possible. Only iPhone users took part, and people who finished all three surveys (67%) had better baseline mental health and attention than those who dropped out. The mediation analyses can't establish causal pathways.

You may have seen headlines claiming the result reversed a decade of attention decline or beat antidepressants. Those are press framings, not the paper's conclusions, and we wouldn't lean on them.

What to take from it

First, the effect on attention is real enough to take seriously and small enough to stay honest about. A d-prime shift of 0.16 to 0.19 on a five-minute task won't transform a workday on its own. It's a nudge.

Second, a total block is hard to keep. With about a quarter of people classed as compliant, the practical lesson is to pick something you'll actually stick to. The authors themselves suggest narrower blocks, such as social media only or limited hours, might be easier to sustain. If you want to try one, our comparison of which blockers people actually stick with is a place to start, and grayscale mode is a lighter experiment. The cost of constant switching is covered in our piece on attention residue.

Third, measure instead of guessing. Run a two-week trial of one narrow block, and compare how many focused hours you log before and after.

How Pomlo fits in

Pomlo is a simple time tracker for iOS, Android and the web, and it gives you the before-and-after numbers a block experiment needs. Focus sessions track focused work rather than just total time, so you can see whether your focused hours actually go up during a two-week trial. Reports show where your week went, and one-tap start and stop keeps the overhead low. Your data isn't sold or used to train models.

Try it on the App Store or Google Play.

Frequently Asked Questions

Did the study ban phones completely?

No. The Freedom app blocked mobile internet (Wi-Fi and mobile data) on the phone for two weeks. Calls and texts still worked, and participants could use the internet on other devices.

How big was the attention improvement?

Small to modest. On the gradCPT task, d-prime rose from 2.75 to 2.91 in one group (effect size 0.24) and from 2.81 to 3.00 in the other (0.28).

Did most participants stick to the block?

No. 119 people, 25.5% of those who committed, were classed as compliant. The authors say this likely means the overall effects are underestimated.

Do I need to block everything to get a benefit?

The study can't say. It tested one blunt intervention. The authors suggest narrower blocks, such as social media only or limited hours, might be easier to keep up.