You are describing something researchers have recorded, in their participants' own words. In an interview study of 67 people who identified a problem with their use, one man's account was that his tastes had changed and that he had looked at things he would once have found disgusting. It is documented and it is not rare. That is worth having before anything else here.
What the research actually measures is one question. Almost every published escalation finding rests on a single retrospective item on a questionnaire — I gradually watched more ‘extreme’ pornography, because the porn I watched before was less satisfying — answered by people already unhappy about their use. That item sits inside the tolerance subscale of the standard instrument, built on a six-component model of behavioural addiction, and only one of its three items concerns content at all. In a nationally representative Polish sample, self-reported tolerance did track severity, and that is a real association. It is also a memory, reported once, by a group selected for distress. Nobody can tell you how common this is.
It is not a ratchet. The only study that followed people forward rather than asking them to look back — 248 male adolescents across five waves over two years — found that preference for violent and coercive content fell over time, and that its change was unrelated to how much their overall use grew. One country, one age group, one kind of content. It is still the only forward-looking test of the escalation idea there is, and it did not find escalation.
Habituation is ordinary, and in the laboratory it recovers. Arousal to one repeated sexual stimulus falls within a session. A novel stimulus restores it. And — this is the half that gets left out — re-presenting the original stimulus restores it too. Across sessions the evidence is thinner and disagrees with itself: a five-man study found habituation carrying over week to week, while a 24-man study found none, and found no change in participants' sexual behaviour outside the laboratory either. Dulling to a specific thing you have seen a great deal of is not the same as a permanent change in what you can be aroused by.
The dopamine part that is established is narrow, and it fits here: dopamine is more closely associated with the anticipation and pursuit of a reward than with the pleasure of getting it. That is why an endless search can be more compelling than anything it turns up. It is not a mechanism for permanently rewritten taste, and nobody has demonstrated one.
So will they go back?
Unknown. Not "probably", not "in time" — unknown. No study has measured what a person is aroused by before, during and after a period of abstinence. The closest thing in print is a single patient inside a four-case series, reported at second hand, and even that is a report about restored sexual function rather than about what he wanted. Anyone handing you a timeline for this is inventing it, which is worth remembering about the rest of what they say too.
What is supported is the part you are actually carrying. Shame is a concealment emotion — it points at withdrawal and hiding — and it predicts poorer behaviour-change outcomes than guilt, which stays on the behaviour and points at action. If this question has been sitting in you unspoken for a long time, that is the part with evidence behind it, and it is the part that responds to being worked on.
Two things this page will not do. It will not tell you what your tastes mean about you — your character, your identity, or anything you might do — because nothing in this literature supports that and it is not an app's business. And it will not reassure you that they will go back, because nobody has measured whether they do.
What is left is the workable part: how much distress this is causing, how much of it you are concealing, and how much of your life it is taking. Those are measurable, and unlike the question in the title, they move.
Where this comes from
Read the full lesson: Recalibrating arousal →
Sources and how strong they are (5)
- Mixed evidenceSome people's use shifts toward content they would not have chosen at the start. Among people who already describe their use as a problem this is commonly reported, and in a nationally representative Polish sample self-reported tolerance tracked severity (β = 0.38 for compulsive sexual behaviour, β = 0.27 for problematic pornography use). What is measured, almost everywhere in this literature, is a single retrospective question — the Problematic Pornography Consumption Scale's "I gradually watched more 'extreme' pornography, because the porn I watched before was less satisfying" — answered by people who are already unhappy about their use. Escalation has not been shown to be universal or progressive. The one study that followed people forward rather than asking them to look back — 248 male adolescents across five waves over 24 months — found preference for violent and coercive content fell over time, and that its change was unrelated to how much their overall use grew.Lewczuk K, Wizła M, Glica A, Potenza MN, Lew-Starowicz M, Kraus SW 2022, J Behav Addict 11(4):979-993, doi:10.1556/2006.2022.00076, PMID 36269607
- AnecdotalWhether tastes return to where they started after someone stops is unknown. No study has measured what a person is aroused by before, during and after a period of abstinence. What the laboratory shows is narrower and cuts both ways: arousal to one repeated sexual stimulus falls within a session, a novel stimulus restores it, and — the half usually left out of the popular version — re-presenting the original stimulus restores it too. Across sessions the evidence is weaker still. A five-man study found habituation carrying over from week to week; a 24-man study found no habituation across sessions and no change in participants' sexual behaviour outside the laboratory.Koukounas E & Over R 1993, Behaviour Research and Therapy 31(6):575-585, doi:10.1016/0005-7967(93)90109-8, PMID 8347116
- Well establishedDopamine is more closely associated with the anticipation and pursuit of a reward - "wanting" - than with the pleasure of consuming it - "liking".Berridge & Robinson 1998, Brain Research Reviews 28(3):309-369, doi:10.1016/S0165-0173(98)00019-8, PMID 9858756, and Berridge 2007, Psychopharmacology 191(3):391-431, doi:10.1007/s00213-006-0578-x…
- Well establishedThe Problematic Pornography Consumption Scale measures problematic use through the six components of Griffiths' (2005) model - salience, mood modification, tolerance, withdrawal, conflict and relapse. Not every instrument in the field is built this way: the Brief Pornography Screen, which produces the highest prevalence figures quoted at C05 and C06, is a short screen rather than a six-component measure. The tolerance subscale is three items, only one of which concerns content — "I gradually watched more 'extreme' pornography, because the porn I watched before was less satisfying" — so most published escalation findings rest on one retrospective self-report question (see C55).Bothe et al., The Development of the Problematic Pornography Consumption Scale (PPCS), J Sex Res 2018;55(3):395-406, doi:10.1080/00224499.2017.1291798
- Well establishedShame is a concealment emotion associated with withdrawal and hiding, and predicts poorer behaviour-change outcomes than guilt, which is behaviour-focused and action-oriented.Tangney, Stuewig & Mashek 2007, Annual Review of Psychology 58:345-372, doi:10.1146/annurev.psych.56.091103.070145, PMID 16953797, PMC3083636; Tangney & Dearing 2002
More on this in the app.
The full 41-lesson course is in the app, every claim graded the same way, and an anonymous community of people doing the same thing. Free.
Coming to iOS and Android.