Neuroscience

Does porn shrink your brain? What the grey matter study actually found

Heard that porn cuts grey matter by 20%? Here is what the 2014 brain study really showed, what it cannot show, and what the science does support.

By the WINNR team · 17 October 2026 · 7 min read

You read it somewhere, probably in a post with a brain scan graphic: porn use causes a 20% reduction in grey matter and a 10% drop in brain connectivity. Now you are lying awake wondering whether you have already done permanent damage.

The short answer is that the brain-damage version of this story is not supported by the research it claims to rest on. The longer answer matters more, because there are real, better-evidenced reasons some people struggle with their porn use, and they have nothing to do with a shrunken brain.

Where the "20% less grey matter" claim comes from

Almost every version of this claim traces back to one study: Kühn and Gallinat (2014), published in JAMA Psychiatry as "Brain structure and functional connectivity associated with pornography consumption: The brain on porn".

Researchers at the Max Planck Institute in Berlin recruited 64 healthy men aged 21 to 45, asked them to self-report their weekly hours of porn, and scanned their brains with MRI. Three results came out of it:

• Men who reported more weekly hours had, on average, slightly less grey matter volume in one region, the right caudate, part of the striatum.
• More weekly hours went with lower activation in the left putamen while viewing sexual images.
• More weekly hours went with weaker functional connectivity between the right caudate and the left dorsolateral prefrontal cortex.

Notice what is missing: any percentage. The paper reports statistical associations, not "20% shrinkage" or "10% less connectivity". Those figures appear in neither the published abstract, nor the Max Planck Institute's own press release, nor the contemporaneous science reporting on the study (Ghose, 2014). Three independent accounts of the same research give the same sample of 64 men and no percentages at all. The numbers appear to have entered somewhere in the retelling, and been repeated until they read as fact.

There is one small irony worth noting. The only figure close to 20 in the original coverage is the viewing time: the men averaged around four hours a week, and the heaviest users reported close to 20 hours (Ghose, 2014). Whether that is where the "20%" came from is guesswork, but it is a reminder of how easily a number migrates from one claim to another once it leaves the paper.

What that study cannot tell you

Correlation is not cause

The study was cross-sectional: every participant was scanned once. That design can show that two things go together; it cannot show that one caused the other. The authors said so themselves. In the paper they raise the alternative explanation that a smaller striatum could be a pre-existing trait that makes porn more rewarding to some men, rather than a consequence of watching it, and they called for longitudinal studies to settle it.

In the press release, co-author Jürgen Gallinat went further and said the team "assume that frequent pornography use leads to these changes". That is an assumption, stated as one. More than a decade later, we could not locate a longitudinal study that has tested it.

A small, single sample with real limitations

Sixty-four men is a small sample for structural imaging. Porn use was self-reported, on a topic where under-reporting is likely. Nobody in the sample had been assessed for problematic use, so the findings say nothing about people who feel out of control. The authors also noted that the grey matter cluster extended into neighbouring white matter and that a technical artefact could not be fully ruled out (Kühn & Gallinat, 2014).

None of this makes it a bad study. It was a reasonable first study. The problem is what happened to it afterwards.

The scientific dispute over "porn addiction" neuroscience

This is not one study versus the internet. Researchers disagree with each other, in print, about what brain data on porn use mean.

In 2014, Ley, Prause and Finn published a review in Current Sexual Health Reports arguing that the "pornography addiction" model lacked empirical support. Citing a review by Mudry and colleagues, they note that only 27% (13 of 49) of articles on high-frequency sexual behaviours contained actual data, and they conclude that long-term neural changes have yet to be demonstrated in any study of "porn addiction" (Ley, Prause & Finn, 2014).

A year later, Prause and colleagues (2015) ran an EEG study with 122 adults, 55 of whom reported problems regulating their viewing of sexual images. Models borrowed from substance use predict that such users should show a heightened brain response to sexual cues. The study found the opposite: among those problem users who also reported higher sexual desire, a smaller late positive potential to sexual images than in controls (Prause et al., 2015). That the finding is conditional on sexual desire is the detail critics of the study press on, and it is worth knowing before anyone leans on it.

On the other side, Voon and colleagues (2014) scanned 19 men with compulsive sexual behaviour and 19 controls and found greater activity in the ventral striatum, dorsal anterior cingulate and amygdala when the compulsive group viewed explicit video, with more "wanting" but similar "liking", a pattern they read as consistent with incentive-motivation theories. Potenza and colleagues (2017) later argued in The Lancet Psychiatry that excessive sexual behaviour may fit an addictive-disorder framework.

A 2025 preregistered systematic review of striatal responses to sexual cues concluded that some studies find differences between problematic and non-problematic users, others find none, and the evidence is "still limited" (Lichte et al., 2025). Nineteen per group, 64 men, 122 adults: these are the sample sizes the whole debate rests on.

When the World Health Organization added compulsive sexual behaviour disorder to the ICD-11, it placed it among impulse control disorders, not addictive disorders (Kraus et al., 2018). The diagnostic text also states that distress "entirely related to moral judgements" about one's own sexual behaviour does not meet the requirement for the diagnosis.

What the research does reasonably support

Rejecting the brain-damage story is not the same as saying porn has no effects. Three findings are better evidenced.

Loss of control is real, and it is defined by behaviour, not scans

The ICD-11 criteria are behavioural: repeated failed attempts to cut back, sexual activity crowding out other responsibilities, continuing despite harm or despite little satisfaction. These are things you can notice in your own life, and they are what clinicians look for.

How much you watch predicts problems less than you would think

A meta-analysis by Grubbs and colleagues (2019) pooled studies on self-reported porn problems. Amount of use was related to feeling you have a problem (r = .27, 10 studies), but moral incongruence, the gap between what you believe about porn and what you do, was related far more strongly (r = .58, 7 studies). A lot of the "I'm addicted" feeling comes from the conflict between values and behaviour, not from the hours alone.

Porn use seems to nudge short-term reward-seeking, and that can shift back

Negash and colleagues (2016) ran two small studies. In the first, people with higher porn use showed steeper delay discounting (preferring smaller rewards now over larger ones later) four weeks on, after controlling for where they started. In the second, participants randomly assigned to abstain from porn for three weeks showed lower delay discounting than those assigned to give up a favourite food. The samples were small, but random assignment is a stronger basis for a causal claim than a one-off scan. It also points away from "permanent damage": the effect moved within three weeks.

A more useful way to think about your own brain

The Dual Control Model of sexual response, developed by John Bancroft and Erick Janssen, proposes that sexual response is the balance of two systems: sexual excitation (the accelerator) and sexual inhibition (the brake). People vary on both, roughly the way they vary on any trait. Someone with a sensitive accelerator and a weak brake will respond sexually even when there are good reasons not to, and that combination is associated with higher-risk sexual behaviour (Bancroft & Janssen, 2000).

That frame explains why two people with the same screen time can have very different experiences, and why "my reward system is broken" is the wrong starting point. Your pattern of excitation and inhibition is something you can map, understand and work with. A damaged brain is not.

What to do with this

If you are worried, drop the brain-damage question and ask the behavioural ones. Have you tried to cut back and not managed? Is porn crowding out sleep, work, sex with a partner, or things you used to enjoy? Do you keep going after it has stopped being satisfying? If the answer is yes and it is causing you real distress, talk to a GP, a psychologist or a sexologist. That is not a sign something is wrong with your brain; it is the sensible response to a habit that has got ahead of you.

If you want structured help getting back in control, WINNR is a science-based app built for exactly this, for €6 a month or €49 a year at winnr.now.

Want a plan for your own pattern? Take the private 4-minute check. Fighting an urge right now? Open the urge page.