Daniel sent us this one — and it's a really good one. He had a single panic attack about fifteen years ago, after a night of conference drinking followed by too much coffee and too little sleep. What stuck with him was how overwhelmingly physical it was — blurry vision, trouble peeing, the sense that something was medically wrong. And he got curious about why that combination of alcohol withdrawal and caffeine is such a perfect recipe for panic. He learned it had a lot to do with glutamate — how alcohol suppresses it, and then it rebounds higher than baseline the next day. Which led him to a bigger question. If glutamate is so central to this, why aren't there drugs that directly regulate it in widespread use for anxiety? Instead, you'd probably get prescribed an SSRI. So does serotonin somehow stabilize the glutamate system? And if so, does that have a role in treating alcoholism secondary to anxiety disorders? That's a lot to unpack.
It is, and the physicality Daniel described — that's not incidental. That's the whole story. The reason a panic attack feels like a medical event is that, at the neurochemical level, it kind of is one. Your brain is flooding itself with excitatory signals, and your body is along for the ride. The blurry vision, the urinary hesitation — those are autonomic nervous system effects. Your body has diverted resources to fight or flight, and things like fine visual accommodation and smooth muscle relaxation in the bladder get deprioritized. It's not in your head. It's in your entire nervous system.
The bladder thing is such a specific detail. You're standing there thinking, I've done this thousands of times, why can't I do it now?
And it's because the sympathetic nervous system has clamped down on the internal urethral sphincter. The same mechanism that dilates your pupils and speeds your heart is telling that sphincter to stay tight. It's all one coordinated survival response — except there's no predator. Just your own neurochemistry turning on you.
Walk me through the glutamate part. Alcohol suppresses it, then it rebounds. What's actually happening?
Glutamate is your brain's primary excitatory neurotransmitter. It's the gas pedal. GABA is the brake. Alcohol is a GABAergic drug — it enhances the effect of GABA, which is why you feel relaxed, disinhibited, sedated. But your brain is a homeostatic machine. It doesn't like being pushed off balance. So when alcohol is flooding the system and enhancing GABA, the brain compensates by downregulating GABA receptors and upregulating glutamate receptors. It's trying to stay functional. This is tolerance.
The brain is essentially turning up the volume on excitation to counteract the sedation.
And then the alcohol wears off. Now you've got fewer GABA receptors doing their braking job, and more glutamate receptors primed and ready to fire. The gas pedal is bigger and the brake is smaller. This is why the morning after heavy drinking, you're jittery, anxious, sensitive to stimuli. It's not a hangover in the headache-and-nausea sense. It's a state of mild hyperexcitability.
Then you add caffeine.
Which is an adenosine antagonist. Adenosine builds up during the day and makes you feel sleepy. Caffeine blocks those receptors, so you don't feel the sleep pressure. But adenosine also modulates glutamate release. When you block adenosine, you disinhibit glutamate further. So now you've got a brain already primed for excitation from the alcohol rebound, and you've just removed another layer of braking. Plus caffeine directly stimulates the sympathetic nervous system. It's not additive. It's multiplicative.
The conference scenario Daniel described — drinking at a social event, minimal sleep, then coffee to power through the next day — that's basically a neurochemical trap. You've set up a glutamate surge, you've slept too little to let the system reset, and then you've poured caffeine on it.
Sleep deprivation itself increases extracellular glutamate. The glymphatic system, which clears waste products from the brain including excess glutamate, is most active during deep sleep. If you've had four hours of restless sleep after drinking — which isn't real sleep, by the way, alcohol fragments sleep architecture — your brain hasn't cleared anything. You're starting the day with elevated glutamate, upregulated receptors, and then you add caffeine. A panic attack in that scenario is almost predictable.
But Daniel had one, and then never again in fifteen years. So there's something about individual vulnerability too.
Some people's glutamate systems are more reactive. There's a genetic component to NMDA receptor subunit composition that affects how readily the system kindles. The term "kindling" is important here. It comes from epilepsy research. Each seizure makes the next one more likely, because the neural pathways involved become progressively more sensitized. The same phenomenon appears in mood and anxiety disorders. Each episode of severe dysregulation lowers the threshold for the next one.
Which is what makes the alcohol-anxiety relationship so insidious. You drink to calm down, and it works in the moment, but the rebound makes you more anxious than you were before. So you drink again.
The kindling effect means that over time, the same amount of alcohol produces a worse rebound. This is why some people with anxiety disorders reach a point where they simply can't drink anymore. Their system has been sensitized to the point where even one or two drinks produces a disproportionate glutamate surge the next day. They'll describe it as "hangxiety" — a portmanteau that's actually pretty neurologically accurate.
The glockenspiel of millennial neologisms.
It is, but it captures something real. There's a subset of people for whom the emotional hangover is far worse than the physical one. And it tends to get worse with age, which makes sense because the brain's homeostatic mechanisms become less efficient over time. You don't bounce back as fast.
Let's get to the core of the question. If glutamate dysregulation is central to this — and to anxiety more broadly, and to alcohol withdrawal — why aren't there glutamate-modulating drugs everywhere? Why is the standard of care still SSRIs?
This is where the story gets complicated and frankly a little frustrating. There are glutamate-modulating drugs. Some of them have been around for decades. The problem is that direct glutamate modulation is pharmacologically messy in ways that serotonin modulation isn't.
Glutamate is everywhere in the brain. It's the workhorse. It's involved in basically every neural process — learning, memory, sensory processing, motor control, everything. If you broadly suppress glutamate, you don't just reduce anxiety. You impair cognition. You cause sedation. At high enough levels, you cause anesthesia or coma. The therapeutic window is narrow. Serotonin, by contrast, is more of a modulatory system. It tunes things rather than driving them directly. You can tweak serotonin and affect mood and anxiety without knocking out basic brain function.
It's the difference between adjusting the thermostat and taking a sledgehammer to the furnace.
There's more. Glutamate acts through multiple receptor subtypes — NMDA, AMPA, kainate, and metabotropic receptors — and they're distributed differently across brain regions. An ideal anxiety drug would modulate glutamate in the amygdala and the bed nucleus of the stria terminalis, which are fear circuitry hubs, without touching glutamate in the hippocampus, which you need for memory, or the prefrontal cortex, which you need for executive function. We don't have that kind of precision yet.
We have some drugs that try.
Ketamine is the most famous example. It's an NMDA receptor antagonist, and it produces rapid antidepressant and anxiolytic effects at sub-anesthetic doses. The effects can appear within hours, compared to weeks for SSRIs. That's a huge deal. But ketamine is also a dissociative anesthetic that can cause psychosis-like effects, has abuse potential, and requires medical supervision. Esketamine, the nasal spray version, got FDA approval back in twenty nineteen for treatment-resistant depression, but it's tightly controlled.
There's the other direction — drugs that enhance glutamate clearance rather than blocking receptors.
N-acetylcysteine, or NAC, is an over-the-counter supplement that modulates the glutamate system indirectly. It's a precursor to glutathione, but it also acts on the cystine-glutamate antiporter, which regulates extracellular glutamate levels. There's decent evidence it reduces craving in substance use disorders and may help with compulsive behaviors. But it's modest in effect. It's not a panic attack abortive.
What about the drugs used for alcohol withdrawal? If the seizure risk comes from glutamate rebound, surely they're targeting that system.
This is an important point. Benzodiazepines are the first-line treatment for alcohol withdrawal, and they work by enhancing GABA, not by directly touching glutamate. They're essentially substituting for alcohol at the GABA receptor, stabilizing the system while the brain readjusts. But you're right that the seizure risk is glutamate-driven. In severe withdrawal, the glutamate rebound can be so extreme that it produces excitotoxicity — neurons literally firing themselves to death. That's what a withdrawal seizure is. The benzodiazepines calm things down indirectly by boosting the brake, but they don't directly address the excess glutamate.
Even in the condition where glutamate dysregulation is most dangerous, our primary intervention works on the other side of the balance.
Because it's safer and more controllable. There are drugs that directly target glutamate in this context — topiramate, for example, an anticonvulsant that reduces glutamate release and enhances GABA. It's actually FDA-approved for alcohol use disorder, but it's not widely prescribed. The side effect profile is rougher than naltrexone or acamprosate, and many physicians aren't comfortable with it.
Acamprosate — that's the one specifically for alcohol craving, right? How does it work?
Acamprosate is interesting because its mechanism was debated for years. The current consensus is that it modulates the NMDA receptor, normalizing the glutamate system after chronic alcohol use has dysregulated it. It doesn't directly agonize or antagonize. It's more of a stabilizer. Which is conceptually elegant — instead of pushing the system in one direction, you're helping it regain homeostasis. But in practice, it's modestly effective. It reduces relapse risk by maybe ten to fifteen percent over placebo. Not nothing, but not transformative.
Ten to fifteen percent. That's a margin that makes you grateful it exists and also frustrated that it's the best we've got.
That's basically the story of glutamate-modulating drugs in psychiatry. They work for some people. But they're not the clean, well-tolerated, broadly effective medications that SSRIs turned out to be. And that brings us to Daniel's core question. Does serotonin somehow stabilize the glutamate system?
Because if SSRIs work for panic disorder — and they do, quite well for many people — and panic is at least partly a glutamate phenomenon, there must be a connection.
There is, and it's one of the more elegant findings in psychopharmacology. Serotonin doesn't just do one thing. Different serotonin receptor subtypes are expressed in different brain regions, and some of them directly modulate glutamate release. The five-HT one A receptor, which is heavily targeted by SSRIs, is expressed on pyramidal neurons in the prefrontal cortex and hippocampus. When you activate it chronically — which is what SSRIs do after the several-week lag — you reduce glutamate release in those circuits.
The SSRI is indirectly turning down the gas pedal.
Through a completely different route than a direct glutamate drug would. It's upstream. Serotonin neurons project to glutamatergic neurons and tell them to quiet down. But this takes time because the receptors have to desensitize and the system has to recalibrate. That's why SSRIs take weeks to work for anxiety, even though serotonin levels rise within hours of the first dose.
That lag has always been one of the big mysteries of SSRIs. If serotonin goes up immediately, why doesn't mood go up immediately?
Because it's not about the serotonin level. It's about the downstream adaptations. The five-HT one A autoreceptors — serotonin receptors on the serotonin neurons themselves — act as a brake. When you first take an SSRI, serotonin rises, these autoreceptors detect it, and they reduce firing. So you're actually getting less serotonin signaling in the projection areas initially. Over weeks, those autoreceptors desensitize, the brake comes off, and serotonin signaling in the prefrontal cortex and amygdala actually increases. That's when the therapeutic effects appear.
The whole system has to retune itself. It's not a volume knob. It's more like an orchestra that needs to rehearse.
During that rehearsal period, some people feel worse. Increased anxiety is a known side effect in the first week or two of SSRI treatment. Which makes sense — you're perturbing a system that's already dysregulated, and the initial compensation can be in the wrong direction before it corrects.
Which must be a nightmare for someone with panic disorder. You're already terrified of your own body's responses, and now the medication that's supposed to help is making them more intense for a while.
It's a real clinical challenge. Good psychiatrists will start at very low doses for panic disorder patients and titrate up slowly. Sometimes they'll bridge with a benzodiazepine for the first few weeks. But it requires trust and communication, and not every patient gets that level of care.
Let's trace the full pathway. Alcohol suppresses glutamate acutely, the brain compensates by upregulating glutamate receptors, alcohol wears off, you get a glutamate surge, and if you're prone to panic, that surge can trigger an attack. SSRIs, over time, reduce glutamate release in fear circuits by modulating serotonin receptors. So they're addressing the same final common pathway from a different angle.
And it explains why SSRIs can help with both anxiety disorders and alcohol use disorder when they co-occur. There's evidence that treating the underlying anxiety with SSRIs reduces drinking in people who are using alcohol to self-medicate. The logic is straightforward — if the anxiety is driving the drinking, and you reduce the anxiety, the drinking becomes less reinforcing.
That only works if the anxiety came first. What about people whose anxiety is largely caused by the alcohol use itself?
That's the diagnostic challenge. Alcohol-induced anxiety disorder is a recognized diagnosis, distinct from primary anxiety disorders. In practice, it's often hard to disentangle because the two conditions feed each other. Someone starts drinking to manage social anxiety, the drinking dysregulates their glutamate system, the rebound anxiety gets worse, they drink more. After a while, you can't tell which came first. But the treatment implications are different. If it's primarily alcohol-induced, the anxiety should resolve with sustained abstinence. If it's a primary anxiety disorder, it won't.
Daniel mentioned people with alcoholism being kept on a hamster wheel of suppressing glutamate, which in severe cases can lead to life-threatening seizures. That's the kindling effect in action, right?
Yes, and it's worth being precise about this because it's one of the most dangerous aspects of severe alcohol use disorder. Kindling was first described in the nineteen sixties by Graham Goddard, who found that if you apply a sub-threshold electrical stimulus to the amygdala repeatedly, eventually it triggers a full seizure. The brain has learned to seize. The same thing happens with alcohol withdrawal. Each withdrawal episode is more severe than the last, and the risk of seizures and delirium tremens increases.
It's not just that withdrawal is unpleasant. Each cycle is literally making the brain more fragile.
More likely to progress to life-threatening withdrawal. This is why benzodiazepines are so important in medically supervised detox. They prevent the kindling process by keeping the GABA system stable while the glutamate system gradually normalizes. Without them, the mortality rate from severe alcohol withdrawal would be much higher. Delirium tremens has a mortality rate of up to thirty seven percent untreated.
Thirty seven percent. That's not a psychiatric condition at that point. That's a medical emergency.
It always was. The distinction between psychiatric and medical is artificial when you're talking about severe alcohol withdrawal. You have autonomic instability — blood pressure, heart rate, temperature regulation all going haywire. You have electrolyte imbalances from vomiting and sweating. You can have cardiac arrhythmias. It's a systemic crisis driven by a brain that's lost its ability to regulate its own excitation.
This is the same glutamate system we're talking about for a panic attack after a night of conference drinking. Just at a completely different order of magnitude.
That's what I find so compelling about this whole picture. It's a continuum. At the mild end, you have the hangover anxiety that millions of people experience. At the severe end, you have seizures and delirium tremens. The same mechanism, just pushed to different extremes by dose, duration, and individual vulnerability.
Why haven't we developed a drug that directly targets this mechanism for anxiety? You mentioned the messiness, the ubiquity of glutamate, the narrow therapeutic window. But we've developed drugs for other messy systems.
We have, and there's active research. The holy grail is subtype-selective glutamate modulators. If you could target metabotropic glutamate receptors — mGluR two and three, specifically — you might get anxiolytic effects without the cognitive impairment. There were some promising mGluR two agonists in development about a decade ago. They worked beautifully in animal models. Then they failed in phase three human trials.
Placebo effect was stronger than expected, and the drug effect wasn't robust enough to justify approval. This happens a lot in psychiatry drug development. Animal models of anxiety are limited. A mouse in an elevated plus maze is not the same thing as a human with generalized anxiety disorder. The circuitry is conserved, but the complexity isn't.
The elevated plus maze. That's the one where they see if the mouse is willing to walk on an exposed platform?
Mice don't like open, elevated spaces. An anxious mouse stays in the enclosed arms. An anxiolytic drug makes them more willing to explore the open arms. It's a decent screen, but it doesn't capture the cognitive and emotional complexity of human anxiety. A mouse isn't worrying about its mortgage or ruminating about something embarrassing it said three years ago.
Or standing at a conference unable to pee.
The physical symptoms we can model. The cognitive layer is much harder.
What about the drugs that did make it? You mentioned ketamine. What about memantine?
Memantine is an NMDA receptor antagonist approved for Alzheimer's disease. It's better tolerated than ketamine because it's a low-affinity antagonist with fast off-rate kinetics — it blocks the receptor when glutamate is high but gets out of the way when glutamate is at normal levels. It's been studied off-label for anxiety disorders, and there's some signal there, but not strong enough to drive widespread adoption. The effect sizes are modest.
We have drugs that kind of work, but not well enough to displace SSRIs as first-line treatment.
That's the key point. SSRIs aren't perfect. They have side effects, they take weeks to work, they don't work for everyone. But they're oral, they're safe in overdose, they're well-tolerated by most people long-term, and they have robust effect sizes across multiple anxiety disorders. Any glutamate-modulating drug that wants to replace them has to beat that profile. So far, none have.
What about the supplement angle? NAC you mentioned. What about magnesium?
Magnesium is interesting because it's a natural NMDA receptor antagonist. The receptor has a magnesium binding site that acts as a voltage-dependent block. Supplemental magnesium, particularly magnesium threonate which crosses the blood-brain barrier more effectively, has some evidence for anxiety reduction. But again, modest effects. Not for someone with panic disorder. The kind of glutamate storm Daniel described — that's not a magnesium deficiency.
Let me ask you about the second part of Daniel's question more directly. If SSRIs stabilize the glutamate system indirectly, do they have a specific role in treating alcoholism when it's secondary to an anxiety disorder?
The evidence says yes, with important caveats. Several studies have looked at SSRIs for alcohol use disorder, and the results were initially disappointing. In unselected populations, SSRIs don't reliably reduce drinking. But when you stratify by subtype, a different picture emerges. There's a distinction between early-onset and late-onset alcoholism. Early-onset, often called type B alcoholism, is characterized by impulsivity, novelty-seeking, and antisocial traits. It starts in adolescence or early adulthood. Late-onset, or type A alcoholism, typically develops in the context of anxiety or depression and has a more episodic course.
One is more about disinhibition, the other is more about self-medication.
And SSRIs seem to work for the late-onset, anxiety-driven subtype but not for the early-onset subtype. In some studies, SSRIs actually made outcomes worse for the early-onset group. The hypothesis is that serotonin has complex effects on impulse control, and in people whose drinking is driven by impulsivity rather than anxiety, boosting serotonin might reduce inhibition further.
Which is a good reminder that "serotonin" isn't a synonym for "good mental health." It's a modulator that does different things in different circuits.
That's the danger of the popular "chemical imbalance" narrative. It's not that serotonin is low and you just top it up. It's that the system is dysregulated in complex ways, and SSRIs shift the regulatory set points over time. For some people, in some circuits, that shift is helpful. For others, it's not.
If you're a clinician and someone comes in with both anxiety and alcohol problems, you have to figure out which subtype you're dealing with. And the treatment implications are opposite.
This is why good psychiatric assessment matters. You can't just throw an SSRI at anyone who drinks too much and seems anxious. You need to understand the temporal relationship, the personality profile, the family history. Is the anxiety driving the drinking, or is the drinking driving the anxiety? Is this someone who's always been anxious and discovered alcohol as a coping mechanism? Or someone who was never particularly anxious until years of heavy drinking dysregulated their system?
In practice, a lot of people are probably somewhere in between.
Most people are. The clean subtypes are useful for research but messier in the clinic. Still, the distinction matters because it predicts treatment response. If you've got a late-onset, anxiety-driven pattern, an SSRI might reduce both the anxiety and the drinking. If you've got an early-onset, impulsive pattern, you might be better served by naltrexone, which targets the opioid system and reduces the rewarding effects of alcohol, or topiramate, or acamprosate.
Naltrexone is interesting because it's not touching glutamate or serotonin at all. Alcohol releases endorphins, which reinforce drinking through the opioid system. Naltrexone blocks that reinforcement.
It's one of the most effective medications we have for alcohol use disorder, with a number needed to treat of about nine to prevent one return to heavy drinking. Not spectacular, but real. The point is that alcohol use disorder isn't one thing. It's multiple different neurochemical pathways that converge on the same behavior, and different people need different interventions.
To answer Daniel's question directly — yes, serotonin does stabilize the glutamate system, indirectly and over time. And yes, SSRIs can have a role in treating alcoholism secondary to anxiety disorders, but it depends on the subtype. If the anxiety came first and the drinking is self-medication, treating the anxiety with an SSRI can reduce the drinking. If the anxiety is a consequence of the drinking, the SSRI might not help, and you'd want to look at medications that more directly address alcohol use.
That's a good summary. And I'd add that the reason we don't have direct glutamate-modulating drugs as first-line treatments for anxiety isn't that the mechanism is wrong. It's that we haven't figured out how to target it precisely enough. The glutamate system is the most fundamental excitatory system in the brain. You can't just turn it down globally without consequences. The future is probably in receptor subtype selectivity — drugs that hit specific glutamate receptors in specific circuits. But we're not there yet.
It's the same problem across a lot of neuroscience. The brain uses the same basic chemical tools for everything. Glutamate is the excitatory workhorse everywhere. Dopamine does reward prediction in one circuit and motor control in another. Serotonin modulates mood, appetite, and gut motility. You can't tweak one without affecting the others.
Which is why the history of psychopharmacology is largely a history of serendipity. The first antidepressants and antipsychotics were discovered by accident. Researchers noticed that drugs developed for other purposes had mood effects, and they followed the thread. We're only now, decades later, starting to understand the downstream mechanisms that made those accidental discoveries work.
The SSRI story is a perfect example. Fluoxetine wasn't designed to reduce glutamate in the amygdala. It was designed to block serotonin reuptake because someone had hypothesized that serotonin was involved in depression. The glutamate connection was worked out much later.
It's still not fully understood. We know that chronic SSRI treatment reduces glutamate release in certain circuits. We know it increases neuroplasticity — BDNF levels go up, neurogenesis in the hippocampus increases. But the exact cascade from serotonin reuptake inhibition to reduced panic attacks involves dozens of intermediate steps that are still being mapped.
BDNF — brain-derived neurotrophic factor. That's the one that's basically fertilizer for neurons.
And it's decreased in chronic stress and depression. SSRIs increase it, but so does exercise, so does enriched environment, so does electroconvulsive therapy. It seems to be a common pathway for multiple interventions that improve mood and anxiety. Glutamate modulation is part of that story, but not all of it.
Daniel mentioned that some people with anxiety eventually stop being able to enjoy alcohol at all. Is that the kindling effect making the glutamate rebound so aversive that the cost-benefit calculation flips?
That's a big part of it. As the system becomes sensitized, the same amount of alcohol produces a worse neurochemical hangover. People will say things like, "I used to be able to have three drinks and feel fine the next day, now one drink and I'm anxious for forty eight hours." That's not psychological. That's their glutamate system telling them something has changed.
It's almost adaptive in a way. The brain is making the behavior so unpleasant that you stop doing it.
If only it worked that way for everyone. For some people, the solution to that rebound anxiety is more alcohol, which works temporarily, which deepens the dysregulation, which makes the rebound worse, and so on. That's the hamster wheel Daniel referred to. The brain's homeostatic mechanisms, which evolved to keep us stable, can trap us in escalating cycles when a psychoactive substance is introduced.
The homeostatic mechanisms are doing their job. They're just doing it in a context they never evolved for. The brain didn't evolve to handle regular doses of ethanol.
Ethanol is a remarkably promiscuous molecule. It interacts with GABA, glutamate, dopamine, opioids, endocannabinoids, serotonin. It's not a targeted drug. It's a chemical sledgehammer that happens to produce pleasurable effects at moderate doses. The brain's attempt to compensate for that sledgehammer is what produces both tolerance and withdrawal.
The compensation is specific to the individual. Some people's brains compensate more aggressively on the glutamate side, which makes them vulnerable to anxiety and seizures. Some people compensate more on the dopamine side, which makes them vulnerable to craving and compulsive use. The same drug, different vulnerabilities.
That's the future of addiction medicine, and psychiatry more broadly. Moving from diagnosis based on symptoms to diagnosis based on underlying neurobiological subtypes. Two people can have the same symptoms — panic attacks, excessive drinking — but completely different circuit-level dysfunctions. Treating them the same way is going to fail for some percentage of them.
We're a long way from that though.
We're making progress. Neuroimaging is getting better. Genetic studies are identifying risk variants. But we're not at the point where you can walk into a clinic, get a scan, and have a computer tell the doctor which medication will work for you. We're still largely in the trial-and-error era.
Which is frustrating for patients. Try this for six weeks, see if it helps, if not try something else. Meanwhile you're suffering.
It's deeply frustrating. And it's one reason a lot of people with anxiety disorders turn to alcohol in the first place. Alcohol works immediately. You don't have to wait six weeks. The tragedy is that it works by digging the hole deeper.
Immediate relief, long-term cost. The classic trap.
The trap is biochemical, not moral. That's the piece I think is most important to communicate. People with anxiety disorders who develop alcohol problems aren't weak or lacking willpower. They found something that worked, temporarily, and their brains adapted in ways that made it harder to stop. The neurochemistry explains the trap. It also points to the way out — medications that stabilize the underlying dysregulation so that alcohol isn't doing a job that needs doing anymore.
Which brings us back to SSRIs. For the right patient, they can do exactly that. Stabilize the system so the self-medication isn't necessary. But for the wrong patient, they might do nothing or make things worse.
That's why Daniel's question about the connection between serotonin and glutamate is so practically important. Understanding that connection helps clinicians make better decisions about who should get an SSRI and who should get something else. It's not just academic neurochemistry. It's directly clinically relevant.
Although I suspect Daniel asked partly because he's genuinely curious about the mechanism. He's that kind of guy.
And it's a great question. The fact that serotonin, a modulatory neurotransmitter, can indirectly stabilize the brain's primary excitatory system is one of those things that makes you appreciate how interconnected everything is. There are no isolated systems in the brain. Everything talks to everything else.
The brain as a gossip network.
That's not a bad analogy. Every neuron is listening to multiple conversations and adjusting its behavior accordingly. When you introduce a drug — whether it's alcohol, caffeine, or an SSRI — you're not just affecting one conversation. You're changing the signal-to-noise ratio across the whole network. The art of psychopharmacology is figuring out which conversations to target and how.
And now: Hilbert's daily fun fact.
Hilbert: In nineteen thirty three, an Inuit elder near Pond Inlet, Nunavut, recorded a detailed oral account of the moon appearing to rock back and forth unusually over several nights — a vivid description of lunar libration at its maximum seven degrees latitude swing, passed down as a single surviving artefact of traditional celestial observation from that region.
The moon, rocking back and forth. Good to know it's been keeping itself busy.
Seven degrees, apparently.
To wrap this up — the answer to Daniel's question is yes, serotonin does stabilize the glutamate system, indirectly and over time, through receptor-mediated modulation of glutamate release in fear circuits. That's why SSRIs work for panic disorder and why they can help with alcohol use disorder when it's driven by anxiety. But the connection also explains why they don't work for everyone and why direct glutamate-modulating drugs haven't replaced them — glutamate is too fundamental, too ubiquitous, and we haven't yet figured out how to target it with the precision that serotonin modulation allows.
The experience Daniel described — that single panic attack after a night of drinking and too much coffee — that's a window into a neurochemical process that, at its extreme, can become life-threatening. The same glutamate rebound that made him feel like he was having a medical emergency is what causes withdrawal seizures in severe alcohol use disorder. The difference is one of degree, not kind.
Which is both sobering and, in a weird way, reassuring. The physicality of the experience wasn't a sign that something else was wrong. It was a sign that the system was working exactly as it does — just pushed past its normal limits by a combination of circumstances that happened to converge on a conference afternoon fifteen years ago.
It never happened again, which suggests his system is fairly resilient. Not everyone's is. For those whose systems are more vulnerable, the medications we have can help — but we still have a long way to go in understanding how to target these mechanisms precisely.
Thanks to Hilbert Flumingtop for producing. This has been My Weird Prompts. You can find every episode at myweirdprompts dot com.
If you're enjoying the show, leave us a review wherever you listen.
See you next time.
