Filters did not lower teenagers' exposure and ID checks cut adults' porn time by about a tenth, while small changes to the phone moved habits a little

Walls and friction

There are two ways to put something between a person and porn. One is a wall, like a filter on the home router or a law that makes a site ask for ID. The other is friction, a small change that makes the phone a little duller or a little further away at night. Walls get most of the attention from parents and from lawmakers. The studies we found give them less credit than they get, and give small frictions a little more.

Filters at home

In 2018 Przybylski and Nash surveyed 1,004 pairs of 14 and 15 year olds and their parents in England, Scotland and Wales¹. They registered their analysis plan before collecting the data. About half the parents used content filters from their broadband or mobile provider¹.

Among the 940 families in the main comparison, 39.1% of teenagers in homes without a filter had seen sexual material online in the past six months¹. In homes with a filter it was 39.4%¹. The gap is three tenths of a point, and it runs the wrong way.

Their exploratory look at 2010 data from 25 European countries had found a small gap, 23% against 16%¹. Even that gap was thin. Fifteen homes had to filter to spare one child¹.

One number in the 2018 survey has no easy reading¹. Teenagers in homes with filters were more likely to report seeing violent porn, 14% against 7%¹. A survey cannot say why. It is a snapshot, not a trial, and both filter use and exposure were self-reported¹. Whatever the reason, in these homes the filter did not change how many teenagers saw sexual material.

Making the site ask for ID

A filter is a choice one family makes. Several US states tried a wall on a bigger scale, with laws that made porn sites check the ID of visitors. In most of those states Pornhub answered by shutting off access².

Brown, Davis and Pope used Comscore data on the browsing of US adults who agreed to be tracked, from 2022 to 2024². The paper's text calls it a panel of about 550,000 users, and its tables count about 550,000 devices². They compared states where Pornhub shut off access with states where it did not, before and after each law².

Pornhub lost a lot. Time on the site itself fell 54%². Time on porn as a whole fell far less. Across the top 25 porn sites, adults cut their time by only about 10%².

The chart shows where the rest went. Half of it, 49%, was already on sites that ignored the law and never asked for ID². Another 31% stayed on the blocked sites, which the authors put down to VPNs, and 10% moved to sites that did not block².

Google searches show the same thing from a different angle. A peer-reviewed 2026 study by Lang and colleagues found that searches for Pornhub fell 51% in states with these laws³. Searches for a site that ignored the law rose 48%, and searches for VPNs rose 23.6%³.

Some caveats matter here. The browsing panel has no minors, and the laws were written to protect minors, so the study shows what adults did². Panel members volunteered, and 87% of their devices were desktops². The browsing study is also a working paper that has not been peer-reviewed yet². Still, the pattern is hard to miss. Shutting one site cut the time spent on that site by about half, and the time spent on porn by about a tenth².

What a wall cannot reach

The filter and the ID check share a design. Each sits outside the person and tries to stop the content arriving. In the British homes and in the blocked states, it arrived anyway, or people went and got it¹². Neither study looked at what the person wanted in that moment, or at the habits around the device in their hand.

That raises a different question. If a wall mostly moves people somewhere else, does a smaller change to the phone itself move them at all? Two kinds of study give a partial answer. We should say plainly that none of them measured porn. They measured screen time and sleep.

A duller screen

In a study published in 2024, Dekker and Baumgartner asked 84 Android users, recruited through the University of Amsterdam and the first author's own network, to run a research app that logged their screen time⁴. For 6 days they used the phone as normal⁴. For the next 6 they set the screen to grayscale, and they said they kept it on about 90% of the time⁴.

Screen time fell from 287 minutes a day to 268, about 20 minutes⁴.

The part we find most interesting is what stayed the same. People unlocked the phone just as often⁴. They picked it up as many times as before and spent less time on it once it was in their hand. Grayscale did not stop the reach for the phone. It made the screen less worth staying on.

A randomised trial of 112 students pointed the same way⁵. Zimmermann and Sobolev found that students told to use grayscale averaged 233 minutes a day, against 283 for students who only tracked their time⁵. Those figures come from the authors' preprint⁵.

The Amsterdam study is weak in ways that matter. It had no control group and only one week of each condition⁴. Its volunteers knew they were testing a phone intervention, and 85% of them already wanted to cut down⁴. And screen time is not porn use. We found no study that measured porn directly.

The phone away at night

The second friction is time and distance. In a 2019 pilot trial at a university in Shanghai, He and colleagues took 38 students who slept badly and used their phone in bed for more than 30 minutes a night, and split them at random into two groups of 19⁶. One group stopped using the phone 30 minutes before their usual bedtime⁶. A bedtime mode locked 17 of their 19 phones, and the other 2 students were told to switch theirs off⁶. Researchers called after the turn-off time to check⁶. The other group got no instructions⁶.

In the phone-away group, the time it took to fall asleep dropped from 31 minutes to 18⁶. In the other group it went from 29 to 27⁶. The gap between the groups was significant, but only just, at p = 0.049⁶.

Sleep quality moved too. On the Pittsburgh index, where lower is better, the phone-away group went from 10.05 to 5.63, and the other group from 9.63 to 8.58⁶.

This is a small trial with soft edges. It had 19 people in each group and no follow-up once it ended⁶. Sleep came from the students' own diaries, and they already believed the phone hurt their sleep⁶. The paper even disagrees with itself on how long the restriction ran, four weeks in its discussion and one week in its methods⁶.

A much larger survey shows how common the habit is. Of 45,654 Norwegian students surveyed in 2022, 87% used screens after going to bed⁷. Each extra hour of screen time in bed went with 59% higher odds of insomnia symptoms and 24 minutes less sleep⁷. A survey like that is a snapshot, so it cannot show the screens caused the poor sleep, and only 35.1% of the students invited took part⁷.

How they found it

• In 2018 researchers surveyed 1,004 pairs of British 14 and 15 year olds and their parents, following an analysis plan they registered in advance. About half the parents used content filters from their broadband or mobile provider.
• An exploratory look at 2010 data from 25 European countries, in the same paper, found a small gap, 23% against 16%. Fifteen homes had to filter to spare one child.
• Comscore tracked the browsing of about 550,000 devices belonging to US adults who agreed to be monitored, from 2022 to 2024. The researchers compared states where Pornhub shut off access with states where it did not, before and after each law.
• A peer-reviewed 2026 study of Google searches in states with these laws found the same pattern. Searches for Pornhub fell 51%, while searches for a site that ignored the law rose 48% and for VPNs 23.6%.
• In the grayscale study a research app logged screen time automatically for 6 normal days, then 6 days with the screen set to grayscale. People said they kept grayscale on about 90% of the time.
• In the bedtime trial, students with poor sleep who used their phone in bed for more than 30 minutes a night were split at random into two groups of 19. One group stopped using the phone 30 minutes before bedtime, with most phones locked by a bedtime mode. The other got no instructions.

What it cannot tell you

• The filter study is a survey from one point in time, not a trial, and filter use and exposure were both self-reported. Teenagers in homes with filters were more likely to report seeing violent porn, 14% against 7%.
• The browsing panel has no minors, and the laws were written to protect minors. Panel members volunteered, and 87% of the devices were desktops.
• The panel does not record VPN use directly, so the VPN share is the authors' reading of the time left on blocked sites.
• The browsing study is a working paper and has not been peer-reviewed yet.
• The grayscale study had no control group, only one week of each, and volunteers, 85% of whom already wanted to cut down.
• The bedtime trial was a pilot with 19 people in each group, self-reported sleep diaries and no follow-up. The paper says the restriction ran for four weeks in its discussion and one week in its methods.
• Screen time and sleep are not porn use. We found no study of these frictions that measured porn directly.

In P-Minus

P-Minus has no content blocker. Its tasks work on your triggers and the reasons behind use. It asks for standing changes to your phone and room, and one of the four is charging the phone outside the bedroom.

Sources

1. Przybylski AK, Nash V. Internet filtering and adolescent exposure to online sexual material. Cyberpsychol Behav Soc Netw. 2018;21(7):405-10.
2. Brown M, Davis EJ, Pope DG. Can online activity be regulated? Evidence from adult websites. NBER Working Paper 35322. 2026.
3. Lang D, Listyg B, Ross BV, Musquera AV, Sanderson Z. Age verification and public adaptation: a pre-registered synthetic control multiverse. J Law Empir Anal. 2026;3(1):23-50.
4. Dekker CA, Baumgartner SE. Is life brighter when your phone is not? The efficacy of a grayscale smartphone intervention addressing digital well-being. Mob Media Commun. 2024;12(3):688-708.
5. Zimmermann L, Sobolev M. Digital strategies for screen time reduction: a randomized field experiment. Cyberpsychol Behav Soc Netw. 2023;26(1):42-9.
6. He JW, Tu ZH, Xiao L, Su T, Tang YX. Effect of restricting bedtime mobile phone use on sleep, arousal, mood, and working memory: a randomized pilot trial. PLoS One. 2020;15(2):e0228756.
7. Hjetland GJ, Skogen JC, Hysing M, Gradisar M, Sivertsen B. How and when screens are used: comparing different screen activities and sleep in Norwegian university students. Front Psychiatry. 2025;16:1548273.

Sources checked October 2026.