Six claims from quitting forums, checked against the studies

Six claims from the forums

Quitting forums and recovery apps pass around a few numbers that sound like science. Ninety days to rewire the brain. A testosterone peak on day 7. A withdrawal week you have to push through. Porn as the reason young men lose erections, and tastes that keep sliding toward the extreme. We took each one back to the studies it comes from, or to the closest studies that exist. Some of these claims fall apart on contact. Some partly hold, and the forums deserve credit for that. And one finding from a brain scanner holds up better than the slogans that tend to get built on it.

The 90 day clock

Several recovery apps sell a 90 or 100 day plan to rewire the brain. Of three we looked at in 2026, two sell 90 days and one sells 100. We found no trial that timed recovery from pornography, so the number has nothing under it. The nearest real data comes from a 2010 study of habits, not of porn and not of brains¹.

Lally and colleagues asked 96 adults, most of them postgraduate students, to pick one healthy eating, drinking or exercise habit, like a bottle of water with lunch, and do it every day for 12 weeks¹. Each day they rated how automatic it felt. For the 39 people whose data the model fitted well, that rating took a median of 66 days to level off¹. The fastest got there in 18 days¹. The slowest was projected at 254, which is 170 days after the study ended¹.

Even that range is generous. The curve fitted well for only 39 of the 82 people who gave enough data, and the authors say about half did not do the habit consistently enough for it to become one¹. Exercise habits took 91 days at the median, against 59 for drinking, a gap the authors found not significant in groups of 13 and 15¹. So the one study people point to gives a spread of months, not a date. A 90 day plan can still be a useful stretch of time to commit to. It is not a measured clock, and a man still struggling on day 91 has not failed a test that science set.

The day 7 testosterone peak

The testosterone claim has a clear paper trail. It goes back to a 2003 study by Jiang and colleagues, in which 28 men gave blood at the same time every day after ejaculating². NoFap grew out of a June 2011 Reddit post about that study, Business Insider reported in 2013³.

The daily averages do rise to a peak on day 7, at 524 ng/dL against 384 on day 1². Then the peak is gone. On day 8 the average was back to 417, and the 8 men who kept abstaining stayed between 370 and 389 through day 16². Nothing in the table supports the idea that testosterone keeps climbing the longer you abstain. The paper's headline figure of 145.7% of baseline takes each man's highest reading over his own lowest, which makes the rise look bigger than the daily averages do².

In December 2021 the journal retracted the paper, because it overlapped one the lead author had already published in Chinese⁴. The notice does not question the data. Still, a retracted paper cannot carry a claim, and we found no study that repeated the day 7 result. A 2001 study by Exton and colleagues followed 10 men through 3 weeks of abstinence but tested blood only on day 0 and day 21, so it cannot show a day 7 peak either way⁵. Its day 21 average, read off the paper's chart, was about 450 ng/dL against 390 on day 0⁵. Even that gap opened while the men waited for an erotic film to start, and the authors suggest anticipation caused it⁵.

So the forum number has a real source, and the source did show a bump. It lasted one day in a small study that has since been retracted².

A withdrawal week

If quitting porn works like quitting a drug, the first week should hurt. A trial by Fernandez and colleagues, run in 2021 and published in 2023, tested that week directly⁶. 176 psychology students at one Malaysian university, all watching porn at least three times a week, were randomly told to try to stop for 7 days or to carry on as usual⁶. Each night they rated craving, mood and 14 withdrawal symptoms⁶.

The two groups did not differ⁶. On day 7 the average symptom score was 1.70 out of 4 for those asked to stop and 1.71 for those who carried on⁶. Craving and low mood did not differ either⁶. The group asked to stop did cut back hard, to 4 minutes a day against 17, and 47 of the 86 watched none at all⁶.

This is the claim where we would hold back hardest from a verdict. The sample was 64% women and 35% men, and most used porn three or four times a week⁶. Only 27 of the 176 scored as problem users, and 9 in 10 had no wish to quit⁶. Seven days may be too short, and the symptom scales came from smoking and alcohol research⁶. In an exploratory look, people with high problem scores who had been watching daily or more did crave more when they stopped⁶. The authors call that a lead to test, since the study had little power to detect it, and suggest sexual desire as one explanation. A forum full of heavy daily users who want to quit is the group this trial says least about.

Erection problems

The erection claim is where the forums are partly right, though not in the way they usually put it. A web survey by Jacobs and colleagues, published in 2021 but run from May 2019 to March 2020 and spread mostly in Belgium and Denmark, kept 3,419 men aged 18 to 35⁷. Among those who had tried penetrative sex in the past 4 weeks, 21.5% had some degree of erection problems on a standard 5-question scale⁷. The paper's text counts these as 444 of 2,067 men, while its own table lists 404, with 189 who left the questions blank⁷.

Split into quarters, how much porn the men watched each week showed no significant link to erection problems, though men with erection problems did watch a median of 39.8 minutes a week against 31.5 for the rest⁷. How often they masturbated showed no significant link either⁷. What did matter was how problematic their porn use scored. In the quarter whose porn use scored most problematic, 34.5% reported erection problems, against 12.9% in the quarter that scored least⁷. Above a score of 28 the share reached 49.6%⁷. Men who found porn more arousing than sex with a partner had 2.34 times the odds of erection problems of men who found both equally arousing⁷.

Two earlier studies point the same way. In 2019 Grubbs and Gola surveyed three US groups of sexually active men who used porn, 147 undergraduates, 297 men matched to national norms and 433 men followed for a year⁸. In all three, men who felt their use was a problem reported worse erections, though in the student group the link fell short of significance⁸. Plain use went with worse erections in only one group, the one followed for a year⁸. Over that year erections barely changed for anyone, so no porn measure could predict a change, and the authors say their links cannot show which came first⁸. A 2015 analysis of 3,948 sexually active heterosexual men aged 18 to 40, in Croatia, Norway and Portugal, found a single link, in the 2011 Croatian sample and among men in the middle third for porn use, not the heaviest third⁹. Its text says those men had higher odds of erection problems, while its own table gives an odds ratio of 0.53, which would mean lower odds⁹. A second Croatian survey in 2014, with a stronger erection questionnaire, found no link at all⁹. The 2021 survey is a snapshot. Its authors say it is possible that erection problems lead to more porn use, though their model treated erection problems as the outcome⁷. They add that men with sexual health problems may have been more prone to take part, and that the problem use score came from a scale they had not first meant to use for that⁷. Across the 2019 and 2021 studies, problem use scores tracked erection problems more closely than hours watched did⁷⁸.

Escalation

Forums also warn that porn use escalates, that what worked last year stops working and the content gets more extreme. A 2021 national survey in Poland by Lewczuk and colleagues asked people to rate almost that exact statement, that they now watch more extreme and varied porn than in the past because it is more stimulating¹⁰. Across 1,277 adults who had sex, masturbated or watched porn at least monthly, 15.8% agreed¹⁰.

The split followed the same line as the erection data. Among people who screened at risk of problem porn use, 34.5% said yes, against 9.6% of everyone else¹⁰. Among those who screened for compulsive sexual behaviour, it was 40.9% against 14.5%¹⁰. A survey of 4,260 Polish university students who used porn, run in 2017 and 2018, found 46.0% had switched to a new genre, and 32.0% reported a need for more extreme or violent material¹¹.

So a sizeable minority report escalation, and it is far more common among people who already struggle. It is not what happens to everyone who watches. The Polish survey was a snapshot, so it cannot say which way the link runs¹⁰. Its screener flagged 22.8% of everyone as at risk, which the authors say probably overcounts problem use, and the survey ran during the COVID-19 pandemic¹⁰.

Wanting without liking

Under several of these claims sits a story about the brain. It is the one that holds up best, and it helps explain why problem scores keep mattering more than hours. In a 2014 brain-scan study, Voon and colleagues compared 19 heterosexual men with compulsive use of explicit material against 19 men without¹². The compulsive group watched 13.21 hours a week on average, against 1.75, and had first seen it at 13.89 years old, against 17.15¹².

In the scanner, explicit videos drew more activity in reward and motivation areas in the compulsive group, and the authors compare it to the network seen in studies of drug cues¹². The men rated the videos too. For the explicit videos, desire was higher in the compulsive group and liking was not significantly higher¹². For the erotic but not explicit videos it ran the other way, with liking higher and desire not¹². In the compulsive group, desire, not liking, went with stronger links between those brain areas¹². A sister study by Mechelmans and colleagues, with 22 men with compulsive use and 44 without, found the compulsive group's attention pulled toward explicit images when tested 250 to 350 milliseconds after they appeared¹³.

That gap between wanting and liking fits what the surveys found. Erection problems tracked problem use scores more than hours watched. The pull toward more extreme content was strongest in the people who struggled most. A man can want something more and more without liking it significantly more than others do, and the cost lands elsewhere in his life. 16 of the 19 men said their use had damaged their relationships or social life, 11 had less desire or weaker erections with a partner, and 2 had lost a job over use at work¹².

The limits are the same ones that run through this whole page. The groups were small, and no formal diagnosis for the condition existed at the time¹². One scan at one point in time cannot show whether porn use caused the brain differences or the other way round, and the differences held only in areas chosen in advance¹². The authors say the results may not apply to people who use explicit material without harm¹².

How they found it

• 96 adults, most of them postgraduate students, each tried to do one healthy eating, drinking or exercise habit every day for 12 weeks and rated how automatic it felt. 82 gave enough data to analyse.
• 28 men aged 21 to 45 gave blood at the same time every day after ejaculating. In a second phase, 8 men ejaculated again and peaked again on day 7. The 8 who kept abstaining stayed between 370 and 389 through day 16.
• Psychology students at one Malaysian university, all using porn at least three times a week, were randomly assigned to try to stop for 7 days or carry on as usual. Each night they reported craving, mood and 14 withdrawal symptoms.
• 5,770 men answered the erection survey. The analysis kept 3,419 aged 18 to 35, and the erection questions counted only those who had tried penetrative sex in the past 4 weeks.
• The Polish survey of 1,541 adults asked people to rate statements, among them "I watch more extreme and diverse types of pornography than in the past because they are more stimulating."
• In the scanner, explicit videos drew more activity in reward and motivation areas in the compulsive group. The authors compare it to the network seen in studies of drug cues.

What it cannot tell you

• The habit curve fitted well for only 39 of the 82 people, and the study stopped at day 84. Anything later, including the 254, is projected. These were new healthy habits, not quitting pornography.
• The testosterone study was retracted in December 2021 because it overlapped a paper the lead author had already published in Chinese. It was small, from one site, with no comparison group in the first phase.
• The withdrawal trial was mostly women, most participants used porn three or four times a week, and seven days may be too short to show withdrawal. 39 of the 86 asked to stop watched at least once during the week.
• The erection and escalation surveys are snapshots. They cannot tell whether porn use led to the change or the change led to more porn use.
• The 2019 and 2015 erection papers used self-reports from volunteers, and neither checked for medical causes of erection problems.
• The brain-scan groups were small and all heterosexual men, and the brain differences held only in areas chosen in advance, not across the whole brain.

In P-Minus

P-Minus names no finish date. Each step asks for its own work, and a slip resets nothing. The gaze counter has two wheels you move yourself, one for each time you lowered your gaze and one for each time you did not.

Sources

1. Lally P, van Jaarsveld CHM, Potts HWW, Wardle J. How are habits formed: modelling habit formation in the real world. Eur J Soc Psychol. 2010;40(6):998-1009.
2. Jiang M, Jiang X, Zou Q, Shen JW. A research on the relationship between ejaculation and serum testosterone level in men. J Zhejiang Univ Sci. 2003;4(2):236-40. Retracted 2021.
3. Love D. Inside NoFap, the Reddit community for people who want to be 'masters of their domain'. Business Insider. 28 November 2013.
4. Retraction note to: A research on the relationship between ejaculation and serum testosterone level in men. J Zhejiang Univ Sci. 2021.
5. Exton MS, Krüger TH, Bursch N, et al. Endocrine response to masturbation-induced orgasm in healthy men following a 3-week sexual abstinence. World J Urol. 2001;19(5):377-82.
6. Fernandez DP, Kuss DJ, Justice LV, Fernandez EF, Griffiths MD. Effects of a 7-day pornography abstinence period on withdrawal-related symptoms in regular pornography users: a randomized controlled study. Arch Sex Behav. 2023;52(4):1819-40.
7. Jacobs T, Geysemans B, Van Hal G, et al. Associations between online pornography consumption and sexual dysfunction in young men: multivariate analysis based on an international web-based survey. JMIR Public Health Surveill. 2021;7(10):e32542.
8. Grubbs JB, Gola M. Is pornography use related to erectile functioning? Results from cross-sectional and latent growth curve analyses. J Sex Med. 2019;16(1):111-25.
9. Landripet I, Štulhofer A. Is pornography use associated with sexual difficulties and dysfunctions among younger heterosexual men? J Sex Med. 2015;12(5):1136-9.
10. Lewczuk K, Wizła M, Glica A, Potenza MN, Lew-Starowicz M, Kraus SW. Withdrawal and tolerance as related to compulsive sexual behavior disorder and problematic pornography use: preregistered study based on a nationally representative sample in Poland. J Behav Addict. 2022;11(4):979-93.
11. Dwulit AD, Rzymski P. Prevalence, patterns and self-perceived effects of pornography consumption in Polish university students: a cross-sectional study. Int J Environ Res Public Health. 2019;16(10):1861.
12. Voon V, Mole TB, Banca P, Porter L, Morris L, Mitchell S, et al. Neural correlates of sexual cue reactivity in individuals with and without compulsive sexual behaviours. PLoS One. 2014;9(7):e102419.
13. Mechelmans DJ, Irvine M, Banca P, Porter L, Mitchell S, Mole TB, et al. Enhanced attentional bias towards sexually explicit cues in individuals with and without compulsive sexual behaviours. PLoS One. 2014;9(8):e105476.

Sources checked October 2026.