The Brain Keeps the Score

Why cocaine makes you cocky, SSRIs make you numb, and your brain never fully forgets either

Here's something nobody in psychiatry (and recreational drug use sphere) seems to want to discuss: every time you spend extended time in a drug-induced brain state, you may be permanently voting for that state. Not through neurotoxic damage (everyone knows drugs can kill neurons) but through something subtler and arguably more interesting: the brain remembers which networks it used, and rewires accordingly.

The cocaine user gets cockier. The weed user gets slower. The SSRI patient gets flatter. The antipsychotic patient never quite comes back. These aren't anecdotes. They're probably long-term potentiation doing exactly what it's supposed to do, just that nobody consented to this long-term subtle rewiring effect.

Okay, reeling back for a second.

Almost every recreational drug taken frequently is known to affect long-term neurobiological function (usually in a negative way). For many of these drugs, we have quite a bit of preclinical toxicology data and sometimes even MRI data that they lead to visible neurotoxicity, much more than just microscopic damage.

This is known for ketamine (discussed here), MDMA (discussed here), methamphetamine, cocaine (discussed here), opioids (discussed here), THC (discussed here), and alcohol. Perhaps even therapeutic amphetamines. Exceptions are nicotine and caffeine, which are not prone to causing a high like other drugs.

However, neurotoxicity is not what this post is about. This post is about long-term brain changes. It seems that the brain can rewire depending on how it is used. In fact, drugs don’t just damage the brain. They train it. This distinction (obvious once you see it, almost completely ignored in psychiatry) is what this post is about.

Anecdotally, people who take a lot of cocaine tend to get more cocky over time, people who take a lot of psychedelics become “weird”, people who smoke a lot of weed tend to become sloooow and dumb (with a few exceptions), and people who take a lot of modafinil tend to become rational (instead of emotional) and perhaps slightly autistic. Similarly, people who take SSRIs tend to become less emotional and empathetic over time. In all cases, the changes seem to persist even after the agent is stopped.

Similarly, as people meditate, they “train” certain neural networks. As these networks are activated, they will strengthen. Networks that are activated less, will weaken (“Use it, or lose it.”). There are tons of fMRI studies confirming long-term brain changes in people who start to meditate. This is somewhat analogous to training a specific movement. If practiced enough times and with enough force, it will induce musculoskeletal adaptation.

Similarly, people jumping and scrolling on their devices for many hours a day “train” their attention pathways to adapt – presumably on a structural level. If 30min of meditation a day can bring about changes in MRI scans that can be measured macroscopically after already 6 months, what do hours and hours of daily scrolling for many years do?

In the past, I had been on low doses of semaglutide for about 4 years. I have been off the drug for about 15 months now but my appetite is nowhere near what it was before semaglutide, probably about half. In my late teens, I put myself on an aromatase inhibitor in order to keep my growth plates open and to grow taller for longer (it gave me about an inch). On it, my libido was crushed given that E2 is very important for libido. However, my libido took many years to gradually recover even after stopping the drug. Given that my brain was highly plastic then, my libido development may have actually been permanently altered. Analogous things may hold true for SSRI-induced sexual dysfunction.

In other words, it seems that the more one spends in a certain brain state, the more this state is long-term potentiated. This is probably through a combination of sustained gene expression changes (epigenetics) and perhaps more importantly by affecting the “strength” of whole brain networks.

So in a way, “the brain keeps the score” and seems to “remember” the influence of past activation patterns. From a scientific perspective, we know pretty well what a certain molecule does to an individual neuron (e.g., binding affinities to a whole range of receptors) but what this means in terms of long-term effect on brain structure and brain function seems quite unexplored.

This has huge societal implication. Psychiatrists hand out neuropharmaceuticals left and right without even asking the question whether there could be long-term structural and functional consequences beyond the “what does the drug do for as long as it is in the body” question.

We know for example that amphetamines or methylphenidate, if taken by kids with ADHD, seem to push neural activation patterns towards “non-ADHD” over time but most other neuropharmaceutical drugs seem pretty non-researched in this regard. There is ample structural MRI data to back this up.

There is more and more literature coming out, that long-term SSRI-use is associated with “amotivational syndrome”, people becoming flatter, people becoming less interesting versions of themselves. Anhedonia is just one of many consequences and sometimes the anhedonia remains even after the SSRI is stopped. People will probably remain subtly different forever as their brain has rewired to some extent. In fact, neuroplastic rewiring depening on usage is one of the diamond traits of mammalian nervous systems.

Anecdotally, I have heard from patients many times that after taking antipsychotics for some time that they were “never the same again” – usually more dumb, dull, and apathetic. Given that many antipsychotics block a whole range of receptors, first and foremost dopamine pathways, it is not too far-fetched to assume that the brain “remembers” these activation patterns through effects on neural networks (LTP/LTD). It is also conceivable that an individual neuron remembers past exposure patterns through self-sustaining loops in gene expression patterns.

So, all in all, experimenting with a specific molecule, particularly frequently and/or for a long time can have three kinds of effects on the brain:

Firstly, neurotoxicity. Unlike synaptogenesis (spinogenesis/dendritogenesis) neurotoxicity is a one-way street. While increased synaptic density can reverse over time, a dead neuron can never undie. For example, this may be the reason why the magic of the first MDMA trip can never be recaptured in subsequent trips.

Secondly, network activation changes seem to long-term potentiate and/or long-term depress if they are activated strongly and/or frequently. In other words, the brain seems to remember past brain states. This is probably what is occurring with hallucinogen persistent perception disorder (explained here). This is probably also what happened to me after having had an eating disorder in my early twenties. It took many months (perhaps even years) of normalized eating, weight gain, and leptin exposure to feel “normal” again – probably accompanied by measurable brain changes in a number of regions. There is fMRI evidence that after 1 year on metreleptin there are a number of trophic changes in brain structure.

Thirdly, an individual neuron may also remember specific activation patterns through (sustained?) changes in gene expression patterns. This may be the reason why long-term administration of amphetamines can cause anhedonia. It may also offer a competing explanation to Post-finasteride syndrome or post-SSRI sexual dysfunction.

Far-reaching implications

The uncomfortable implication of all this is that there may be no clean return to baseline. Every psychiatrist who has ever handed out an SSRI prescription, every teenager who spent two years on stimulants, every regular psychedelic user who noticed they'd become more mystical/emotional - all of them were running an experiment on a system that records its inputs and rewires accordingly, with no guarantee of reversal.

What happens to a brain that has spent considerable time in a pharmacologically enforced state (its networks quietly redrawn, its gene expression patterns subtly shifted) is largely unmeasured, poorly understood, and almost never discussed.

However, given the billions of people who take either recreational drugs regularly and/or are on psych meds, this is a highly relevant topic that very few seem to ever think about.