Can I just say it's fun to me that I encountered you via YouTube music, most notably the excellent Skill Issue https://www.youtube.com/watch?v=YfSGXmYJVdI before just now discovering you are connected to ACX in any way.
+1 the answer is Nicotine. A 3mg pouch is a reliable boost to focus, energy (and probably IQ) with (for me) no observable impact on sleep after an hr or so.
Nicotine's nootropic effects are well-supported. I have it on good word that Professor Henry Lester's Neuropharmacology of Addiction class at Caltech convinced two students to *start* using nicotine.
True. This was a comparison made by someone who's had both legally.
"Pseudoephedrine may also be responsible for a false-positive urine test for amphetamine and methamphetamine. The structural similarity to these drugs means that they may cross-react in a test using the immunological method"
i have quit both caffeine and nicotine. nicotine definitely has worse withdrawal symptoms in my experience but they're both pretty manageable. it's definitely not incomparable. caffeine withdrawal makes you extremely tired and sluggish for a few days. nicotine withdrawal, in my experience, was more about fighting the urge to smoke a cigarette than about any particular acute symptoms.
I have found that while the withdrawal effects for caffeine and nicotine are similar, nicotine's "mind game" is much stronger than coffee. Two weeks after quitting coffee, I no longer crave it. Even months after my last use of nicotine I still think about it ~daily, and find myself making excuses to use it.
Nicotine addiction is, I think, substantially worse. It's more addictive, aiui more harmful to be addicted to, and way more expensive to be addicted to (at least in the US/most western countries where it's taxed to the moon)
IMO I’m not sure being addicted to 3-6mg (1 or 2) pouches a day is meaningfully different than being addicted to daily caffeine. Part of what makes it addictive is that it’s *enjoyable* to drink coffee and satisfy that craving. I *enjoy* my 3-6mg of nicotine a day and don’t have a strong urge to take more. At that level, it’s mildly annoying but not disruptive to skip any given day—it’s missing out on something that feels good and perks me up. I think “addiction” is a loaded term (for obvious reasons) but functional addictions (as are commonly the case with caffeine) don’t seem particularly problematic, and can be a nice part of life (or even healthy!), in the form of coffee or tea or pouches or lozenges or gum.
I think that's true, and if you have no temptation to escalation, you are probably alright, but I found nicotine much more persuasive than caffeine: I have never, as I have with nic, found myself making excuses for why I *have* to take it today, even though I hadn't planned on it.
I had the impression that neither caffeine nor nicotine was biologically harmful in anything like a normal dose. Am I wrong?
Of course, something can be harmful to your lifestyle and through that be harmful to your body, but I don't think either of these will turn you into an up-all-night degenerate like amphetamine might.
Addicts don't take normal doses of substances. If you look through the quitting Zyn subreddit, you'll see many accounts of unpleasant side effects of nicotine addiction - anecdotal, yes, but I don't think anyone doubts that constantly taking doses of nicotine that would be lethal to someone without a tolerance is likely to cause side effects. It's much harder for a caffeine addiction to get that bad.
Similar to how if you find it too difficult to buy Sudafed at the pharmacy due to legal restrictions, you can always use readily-available methamphetamine as a precursor to make your own:
I occasionally (less than once per month) use a 1mg oral nicotine lozenge as an evening stimulant, and have never been at all tempted to escalate, for whatever that's worth.
Nicotine patches seem more dangerous, because they're typically a much larger dose released over a longer time span. Why would you use an extended-release product when the whole point was to find a shorter-acting stimulant?
Meanwhile, vaped nicotine is short-acting if you don't redose, but you *will* redose, because it hits so quickly that you get a clear "rush" and the direct stimulus-response association is really compelling.
At least for me, oral lozenges sit in the sweet spot where they kick in slowly enough as to not trigger Skinnerian conditioning, and wear off quickly enough that the system doesn't have time to form a physical dependence. YMMV.
I use 3mg nicotine patches for this purpose, similarly no temptation to escalate. Haven't used them in months actually. I take the patch off after about 30 minutes usually. I think patches are better than lozenges because less sensation/activity associated with the chemical stim. But only if you can find low-dose patches and follow strict rules about limiting dose and frequency.
While I find it helpful and safe, it probably isn't good to recommend to others because no matter how many times you reiterate that they must use low dose and sparingly, someone will ruin their life with a daily high dose and then blame you.
Ex-nicotine addict here, got addicted from nicotine gum then moved to vapes and then cigarettes. It was really tough to beat the addiction, do not recommend
I did this with nicotine lozenges and while I do take them multiple times a week, I’ve gone off them for incidental reasons (forgot to order more, vacation, etc.) without any apparent negative effects. I think they’re better than the patches since they subjectively seem hard to get addicted to.
Same here. Only with the lozenges (not gum, not cigarettes, certainly not vaping. I haven't tried patched. But with the lozenges I take them a few times a day and when I can't take them for a week or whatever it's no big deal. In fact, I sometimes just forget to take them, despite being obviously dependent on them. So it's easy to stop
Yeah I tried taking nicotine as a nootropic back in the day following Gwern's post on it. Found that I very quickly found myself making excuses to use it more and more, and reached the stage where I was anxious when I didn't have it. And quitting it was very difficult
It's... difficult to get addicted to nicotine patches on accident. Not impossible, of course not, but they're by design less reinforcing than any other ROA. I get no noticeable "rush" from a functional dose (7mg/24hr patch for me), and I can avoid tolerance by simply not taking them consecutively.
Interestingly, I did try to get addicted to Zyn (nicotine pouches) on purpose in an effort to induce a behavioral addiction to Leetcode. This backfired, as it turns out I love Leetcode and hate Zyn, and it was a while before binary trees stopped making me nauseous. But that's another story. I was put off nicotine for a long time.
I agree about patches, but after using cut up patches for a while, after having used zyns, thinking the patches were safer, when I went back to zyns, I found them much more addictive than I had the first time: never experienced any "nic sickness" which would have been, I think, a barrier.
Are you sure they got addicted after using *patches*?
Pouches and vapes are much more common but also much more addictive/habit-forming. Gwern explicitly addresses this in the infamous article, but no one seems to have actually read it.
Having experienced nicotine (pouches) addiction and withdrawal, I cannot recommend this. I ended up quitting because of exhaustion caused by the nicotine-induced highs and lows.
6 medium-strength pouches a day, roughly in intervals of 3h. Taking less would cause annoying withdrawal symptoms, taking more something discomforting, like light headaches.
This was my stable use for a couple years, with short-term withdrawals in between. I started, probably like most, with the occasional use, then daily, getting dependent and finding a balance. Interesting that you are able to use them without getting hooked. Once a month means 1 pouch per month, or one period per month where you use a bunch?
1 pouch per month on average, but it's very irregular
I also drink mate when I need a bit of late-night boost - probably twice a week on average and coffee most weekday mornings, that's the closest thing to addiction I have
In the 1996 remake of Diabolique, the character played by Kathy Bates struggles with one of those nicotine gum blister packs, “What’s in here? A piece of the True Cross?”
Nicotine is not a human carcinogen. Since cancer is very complicated, there are some contexts where it can act as a promoter for tumors and so on, but the bottom line is that you shouldn't be worried about nicotine giving you cancer.
Possible alternative lever to pull - what if we found a way to induce ADHD in people? Then caffeine might make it easier for some small proportion of those people to sleep</humor>
(though; I highly doubt the induced ADHD + caffeine sleep would be easier or better than the no-induced ADHD with no caffeine sleep)
If you want to wake up in the late afternoon, there are other methods that won't keep you up five hours later. Go running, lift weights, go swimming, take a cold shower, go to the sauna (or maybe the NYC subway platform). If the sun is still out, get some sunlight. Some of these methods can actually help you fall asleep later.
Regarding "It’s a simple molecule, so brand doesn’t matter much; this one is the cheapest I can find" - supplements often don't contains the claimed proportions or contain impurities which can be unhealthy or counterproductive. So it's worthwhile to stick to trusted supplement brands (or to Enfinity in this case since they're known for making paraxanthine).
...but I don't know how much that holds up for less-common ones, instead of the big-name nootropics/vitamins/whatever where there's enough volume to ensure decent corrective mechanisms for perfidy.
Most of those are fair, however running, swimming, and lifting, take up more time than a lot of people would like. Also, one of the benefits of stimulants is you can take them whenever, you can't easily swim, run, lift, or go to the sauna when you feel tired at work, or when you sit down to do some work in the evening and only then realize you're not mentally energized.
One of the reasons I want a stimulant to wake me up in the late afternoon is so that I have the mental and physical energy to do something like lift weights or go outside for some sunlight. Without the stimulant, I'm more likely to gradually lose energy and eventually just collapse onto my couch. Even when I was lifting weights fairly seriously 3-4x/week, it usually took a pretty significant expenditure of willpower to get me into the gym after work, and that's not something that ever really improved. If I had been able to drink a cup of coffee at 4:30 without impacting my sleep, that would have helped quite a bit.
This will be neither quick nor easy. Last time I went low caffeine I was tea-bagging in cafes. I.e. sucking on black tea bag to make sure I get the last piece of caffeine.
Easiest is to slowly go lower over about 2-3 weeks.
Never really understood this midrash. Like, shouldn’t the kapparah be waking up early to learn, not staying up all night and then sleeping all morning?
Sounds like a lot of my days as an undergrad. Sometimes I wouldn’t go to sleep at all and enjoy the added bonus of not having to deal with bed-head hair.
Coffee has little to no effect on my sleep. I often drink a coffee like an hour before I go to bed. I don't feel any different compared to the days I don't drink one so late.
Coffee definitely doesn't calm me down, in the sense that if I drink several cups I'll get a bit jittery, but it doesn't prevent (or even affect, subjectively) sleep. Going to sleep is just a matter of choosing to do so, for the most part. Like Trudel, I sometimes drink a cup an hour or two before sleep.
Yes. I mean, I like coffee but it would never have occurred to me that not being able to have (specifically) caffeine in the evening was a problem for someone.
By the end of it I figured out that the goal here was: "I am sleep deprived or otherwise need a stimulant, for a burst of work starting at 5pm, but I still want to get to sleep at 10pm".
But the intro (and the early focus on rutaecarpine) made it sound like the goal was specifically about evening coffees, so I was thinking the whole time: why not just drink decaf?
It's not about caffeine. It's aboug wanting to be productive at all hours. I often have an afternoon dip. A nap and a coffee work perfectly... but they also absolutely ruin the next night.
A short acting stimulant would be awesome for a lot of reasons.
I get that, caffeine affects people quite differently, it doesn't stimulate me enough to interrupt sleep let alone allow me to be more productive. There are a lot of short acting stimulants actually, it's just that most of them are controlled, coca tea or a small dose of methylphenidate would fit the bill perfectly.
I've missed posts like this, it's been too long. When was the last one...Lavender's Game for silexan?
All three of these "lactaid but for coffee" sound too Trivially Inconvenient for me to actually utilize on the regular, unfortunately. Definitely can't risk liver damage either, what with other meds + family history already putting me on watch for hepatitis and other failure modes. Which is too bad, cause decaf still tastes like shit in current_year, and it was an unwise decision to accept a recent gift of a giant bag of sugar-free Coffee Rio candies. Grateful for the For Science! exploration though.
Funny enough, my coworker just mentioned the silexan post to me yesterday. It was the first time somebody else has ever brought up the topic of Scott Alexander in conversation with me and caught me totally off guard.
If you’re in the US or can get it shipped - try Peet’s Decaf pre-ground House or Major Dickasons roast. Use more coffee than usual. Pretty dark and strong and fools my regular coffee-loving brain. Dramatically better than other decaf.
I do like Major Tom Dickason's roast! Wonder if the Peet's near my work carries the decaf version. That's usually where I end up wanting late-night caffeine or "caffeine", and of course working at a grocery store, we'll just use whatever one of our brand is on hand...but decaf sells so poorly that we've discontinued almost all of them. Worth looking into.
Agreed, I've always especially enjoyed posts like these (even though in this case I'm not really a caffeine guy at all) and was very happy to see it land in my inbox.
> Maximum safe dose is probably between 300 - 400 mg daily
I don't think you meant to include "daily" there, if we're using "how much of this would you be getting if you just started with caffeine instead" as a baseline. Some people routinely consume on the order of a gram of caffeine per day (I have a friend who drank 6 cups of coffee per day when she worked at SpaceX about a decade ago). Probably you meant "at once"?
(I, personally, would be climbing the walls if I took even 200mg of caffeine in one go; I'm pretty caffeine sensitive and stick to a single cup of tea that usually lasts me from 11 am to 5 pm.)
Your point makes sense, but the FDA does say 400 mg caffeine daily, and I don't know how much it's worth distinguishing daily from per-dose in a chemical with a 10 hour effective half-life.
~2018 I was having heart palpitations and went to my doctor. He asked how much coffee I was drinking daily. My answer of "ohh about 2 pots of coffee" (made extremely strong) prompted a "hey dumbass, stop doing that it'll go away" (it did). TIL that 400mg daily was seen as lot...
If you just want an instant energy boost that clears quickly and lets you sleep afterwards why not just go for cocaine? Surely there's some regulatory arbitrage here where you can give it a brand name, reduce the dose, and package it into a pill that you can sell on Amazon.
Khat is also a huge thing. Unlike Coca, it tends to lead to an entirely unsafe country (outdoors) due to traditional patterns of consumption, at least I assume there's no equivalent to Khat hour in Peru...
If your reason for drinking coffee at 5pm is to enjoy the taste or the feeling of a hot drink, sure. But if you're trying to get an energy boost for the evening without interfering with your sleep then decaf won't meet your objectives
Well the point of the whole post is to find an alternative to caffeine because the half-life of caffeine is too long. If you just switch your source of caffeine you haven't really changed that.
A cup of hot chocolate has an order of magnitude less caffeine than a cup of coffee (personally I've never experienced any boost in energy from chocolate of any sort, but I don't have a strong response to caffeine anyway). So it sounds like your suggestion is just to reduce the dose? In which case you could just have a smaller amount of coffee if you prefer it to hot chocolate (and avoid the glucose spike to boot)
I'm going to guess that's a normal cup of hot chocolate. I routinely add four times that amount (it's a very good antioxidant source -- best fruit/vegetable on the market).
Anecdote: I tried rutaercarpine a few times and it tragically did not seem to help at all with letting me drink an afternoon coffee and then get to sleep at night.
Rarebird makes paraxanthine coffee beans -- they start with decaf and then infuse them with paxanthine, iirc. I brewed this every day for a while, it was actually pretty good, and kind of felt like ordinary coffee but shorter-acting.
It doesn’t have direct antagonism, but does improve sleep on caffeine in animals via other pathways… “In addition, since L-theanine was shown to have a relaxing effect and gamma-aminobutyric acid (GABA) was an important inhibitory neurotransmitter, the mixture of L-theanine and GABA had a positive synergistic effect on sleep behavior, including sleep quality and duration in caffeine-induced awake rats, and the mechanism might be that the mixture could promote the expression of GABA receptor, which was conducive to sleep” https://www.frontiersin.org/journals/nutrition/articles/10.3389/fnut.2022.853846/full
What exactly is the point of coffee? Unlike tea, most people seem to be drinking coffee purely for the stimulant effect rather than because they actually like the taste (as seems to be acknowledged both Scott's analysis and the comments section..) and as I understand it one's metabolism adapts to regular coffee-drinking by reducing one's baseline alertness to negate its effects anyway?
So, long-term, the only effect of coffee seems to be to temporarily make you as alert as you'd be if you never drank coffee, only with more side-effects (including being *less* alert pretty much any time you're not drinking coffee..) and at noticeable expense and hassle?
I get "fresh" coffee -- yes, that's with days of resting, it's still chemically active and very potent. Not only does it taste "wild", as it's got lots of chemicals most people never see, it's also a very potent source of antioxidants. I also do a minimal-time extraction using an espresso machine, so I'm getting a very light dose of caffeine per "other psychoactive/physioactive ingredients."
I'm not sure why people drink coffee if they don't like the taste -- but everyone seems to like the smell of coffee, even kids who generally hate the taste. I was that kid, yes, I admit.
I imagine your setup is probably fairly expensive (compared to, say, instant coffee or teabags or whatever), and so it just has different tradeoffs? Scott has answered my fundamental issue downthread (viz. if you only drink 1-2 cups of coffee per week you don't actually develop a tolerance in the first place) so I couldn't say it isn't worth it, just that "ultra-fresh coffee + fancy high-pressure machine" doesn't seem to match how most people I know drink coffee.
For the taste, I'm not sure. The impression I get from this thread is that many/most people drink it for the stimulant effects (so for example the problem can't be solved by everybody just drinking decaffeinated coffee and thus getting the taste but not the stimulant effect) - but perhaps over time one comes to acclimate to and enjoy the taste too, I don't know. (I have an Italian friend who loves coffee and I strongly suspect that she would say that if the coffee-as-stimulant drinkers all switched to your ultra-fresh + machine setup they'd probably all start enjoying the taste too!)
For the smell - personally I can't blinking well stand it! I find being in a coffee shop, or even in somebody's kitchen whilst they're brewing coffee, oppressive and slightly nauseating. I think I might be in a minority here though, as I've never heard another non-coffee drinker complain about the smell.
In From Russia with Love, Fleming writes Bond as a coffee drinker.
I’m going to have to paraphrase from memory here, but he blames tea drinking culture for sapping British vitality and efficiency. In that passage of that novel at least Bond prefers the jolt of coffee.
In Thunderball Fleming has Bond taking benzedrine before his underwater assault on Largo’s yacht. There was some experimental military use of that amphetamine in WII. I think the era was still fairly fresh in Fleming’s mind when he wrote Thunderball.
I’ve read that the US military prefers things like modafinal to amphetamines for extending alert attention now. It doesn’t cause the euphoria that can cause careless mistakes.
Interesting, thanks! I imagined it was a smooth homeostatic mechanism, not a "less than this and your metabolism won't compensate" cutoff point. Does the amount of caffeine matter (ie. super-caffeinated coffee will build up tolerance quickly even at fewer cups per week) or is it literally just the number of cups-per-week that matters?
(The Wikipedia page cites some research at Johns Hopkins that suggests just 100mg per day (one small cup) is enough to develop a physiological dependence and I'm pretty sure most coffee drinkers far exceed this so I'm not sure how effective "just don't drink it often" advice would be for most people - but to misquote Churchill you've already established what the mechanism is and now we're just haggling over the right number of cups...)
I don't know. I know I drink coffee 1-2x per week and don't get tolerance, and I know from experience with patients using Adderall that if they get tolerance taking it every day, we can usually reverse/prevent it by bringing them down to 3-5 days/week.
Not sure I agree about the taste thing. I drink coffee only occasionally but I enjoy the taste if it's made in my preferred style (dark roast, no milk or sugar), and most people I know who drink coffee regularly claim to enjoy the taste as well, to the point that they buy more expensive beans and test and refine different brewing methods to optimise for flavour. There's a whole culture of fancy coffee shops that try to optimize the taste as well.
I don't know much about coffee shops other than (World Bank Economist) Tim Harford's analysis in 'The Undercover Economist', where Harford posits that what fancy coffee shops are actually optimising for is customer-signalling, not taste, and that most of their "fancy coffee" is fundamentally roughly the same. (I don't know this from personal experience, I admit, and probably a pop-Econ book isn't where one ought to look for a study of gustatory experiences...)
For the "buy more expensive beans and test and refine different brewing methods", human nature being what it is this seems like exactly what people would do if they intended to drink coffee at all, regardless of whether they were drinking for or despite the taste.
You and most people you know liking the taste, however - that's a good point and well-taken! I'm still curious about the causality, though! When you say "people I know who drink coffee regularly", do you think they came to like the taste because they drink regularly, or do they simply drink regularly because they like the taste?
It's like wine - there is a component of actual difference in taste, and some people have more skill or natural affinity for appreciating that difference. Then there is also status associated with knowing good wine and being able to buy expensive wine - and therefore lots of people who pretend to be able to know the difference, and a large psychological component to tasting something special.
Of course it is all "roughly the same"! It's all fermented grape juice, or grounded coffee beans diffused in hot water. That is not really a very interesting observation.
Your comparison to wine is interesting and useful - thanks!
For "of course it's all roughly the same", what matters is not what the available choices are made from or how they're made, but the *outcome* of choosing one over the other. I bought a motorcycle tyre today: the website offered me a choice of [checks history] twenty different tyres to choose from. Some are for road use; others for off-road. Some have a lifespan of 7,000 miles; others could last for 15,000 miles or more. Some are optimised for track use, others for urban use, others for touring. Some are rated to 500lbs and 130mph; others to only 400lbs but 170mph. Some are adapted for dry grip at the expense of wet grip, and others vice-versa, and still others for reduced grip across the board in favour of longer lifespan. There's about a 300% difference in price between the cheapest and the most expensive. You could just as easily say "they must all be the same! They're all just circles made of rubber, all exactly 17 inches in diameter, all made using the same process of extrusion and vulcanisation!" but in this case you'd be mistaken - perhaps fatally so.
Offering lots of options is inefficient. It costs more in storage space, complicates inventory management, leads to more breakage/wastage, splitting your purchase orders across 20 product lines reduces economies of scale, etc. When there are lots of options, we generally suppose there to be lots of genuine variation, reasoning that the vendor wouldn't go to the trouble of stocking all those products if they were all the same. Sometimes we're quite right to think this (as in the case of the motorcycle tyres) and other times we're wrong (as might be the case with coffee or wine). But in either case there's always some level of abstraction where you could say "They must be the same because they're all [X]!" regardless of whether the outcome of your choice would be the same of not. I share 99% of my DNA with a monkey - of course we're the same!
I'm not sure if I understand what you mean exactly. I think you are agreeing with me?
My point was that while I agree that coffee and wine is roughly the same at some level of abstractions, just as all tires are donut shaped rubber, that is not really a interesting observation in itself, and you have to go into more detail.
I don't really agree that coffees or wines are all roughly the same at all. First, there clearly exist different levels of quality that most people will agree on. For example wine that tastes like alcohol or luke warm, sour coffee will be considered inferior by most people. When it comes to the luxury coffee in coffee shops, I can agree that they mostly are somewhat similar, and serve the same purpose. But that is missing the point, because what people often want when they go to those shops, beside high qulity coffee, is novelty i.e. they want to taste some type of coffee they have not tasted before. (Ofcourse they may also want other things - like to look like successful people who goes to this kind of shops, but it is not the only reason to want luxury coffee).
I completely agreed with your first, "It's like wine" paragraph, yes (or rather - I didn't so much agree as was convinced by it!)
For your second, "of course it is all roughly the same [...] that's not an interesting observation" paragraph I don't agree: there's a level of abstraction where everything is the same ("of course it's all the same - it's all baryons!") and a level of granularity where everything is different, and it's valid and useful to be able to say that things are different when we'd expect them to be the same, or vice-versa.
Wine and coffee are perhaps a poor example to make the point, because some people believe there's a vast material difference between different options and other people believe much of the difference is in one's imagination. So: Neurofen Express Period Pain tablets and Neurofen Joint & Back Pain tablets are 100% chemically identical; the only difference is that the "Joint & Back Pain" tablets come in a grey packet with a drawing of a person's back and shoulders on it whilst the "Period Pain" tablets come in a pink packet entirely devoid of anatomical drawings.
In this example, the two different painkillers are the same product but consumers believe them to be different products based on their different name/presentation/marketing/whatever. This is an extreme example - probably most different types of coffee/wine aren't literally chemically identical - but the point is that there absolutely do exist things that are presented as different but are (at a reasonable level of abstraction..) actually the same, and some people absolutely will form preferences for one over the other. Therefore, it's perfectly valid and reasonable to say "these products are presented as being different when actually they're virtually the same".
(As for whether wines and coffees are genuinely materially different or whether the difference is mostly imaginary, I couldn't say - de gustibus non est disputandum - but I am reminded of a study (I think maybe by Prof. Richard Wiseman?) wherein experienced wine-tasters were given a selection of red wines to taste and report on, one wine was secretly a white wine dyed red, and hardly anybody noticed anything unusual about the white wine....)
To your question, I can't speak for others, but in my case I started drinking coffee many years ago while working a summer job at a Starbucks. Part of our employee training was to taste some of the drinks and get a sense for how they should taste. Prior to that I'd only rarely tried coffee and hadn't enjoyed it, but somehow while working there I pretty quickly I developed a taste for it, and that's why I started drinking it frequently. I was then a daily consumer for years, but more recently have reduced my intake since I wanted to be less stimulated.
I drink coffee every morning. The ritual of pouring a cup, slowly sipping it as I wake up, and feeling the caffeine withdrawal wear off, is very enjoyable for me - and it helps me poop as well. So even if all it is doing is fighting off withdrawal and making me feel normal, I would still accept that for the joy that it brings my mornings.
Makes total sense that the ritual of making it is a large part of the appeal! cf. the half-dozen-odd different types of tea ceremonies, cocktail mixology, flair bartending, pipe-smoking, etc.
I don't have citable evidence, but from my experience and from talking to people, I have a different impression of the effects of caffeine tolerance:
It seems that if you take a steady daily amount of caffeine at regular intervals over a long period of time, you'll converge on a state where when you're on caffeine, you're somewhat more alert and focused than if you hadn't been taking it at all. Something like this is verifiably true for most people prescribed daily amphetamines, with exceptions.
posting about buying stuff that can cause liver damage, whatever caveats you put, is still not a good idea unless you dont put links in the post and very clearly advise against all of it. that is not the tone of this post though.
Two objections: 1. Censorship of true and helpful information because you’re worried the audience is too irresponsible to use it wisely doesn’t have the best track record, I think.
2. Suppose we don’t worry about the philosophical point above. You still need to consider the counter factual. First, let’s try a simpler example. Suppose Scott had said “don’t drink and drive”. Someone objects: “how dare you say drinking is okay?!? Don’t you know that alcohol causes liver damage?”
So Scott deletes the post. But alcohol is still a thing that people will drink sometimes, and now they don’t even get a warning that they should avoid driving afterwards.
What went wrong? You assumed that by denying someone vital info, you are avoiding tempting them. This could work if you control it at least know about the rest of their infection environment, but in practice you don’t. So they end up doing something (much) worse than what you wanted to prevent.
Typing on my phone is very annoying so applying the drunk driving logic to this post shall be left as an exercise to the reader: how many different ways can you think of that someone without access to this post would do something even worse than what you’re afraid of?
Who are you to say what perfectly-legal supplement a blog may or may not bring to the attention of its audience?
> is still not a good idea
How do you know? What the expected QALY loss due to potential liver damage? What's the expected QALY gain due to enhanced functioning? Have you done that analysis?
It should be noted that virtually all of the chemicals in plants that get us high (including caffeine, nicotine, THC) are in fact toxins produced by the plant to ward off predators. So it's not surprising they can also be toxic to us.
Hot peppers and most spices also fall into this category. Cocoa continues to fester on the plant, despite the high amount of anti-oxidants... you have to do a LOT of cooking to make it non-festery.
Maybe in all the controversy about the health effects of human-made pesticides have we missed discovering that one of them has desirable psychoactive effects in humans? (Going to start drinking from a mug that say "Don't talk to me before I've had my glyphosate!" and see what sort of conversations that gets me.)
I expect this works for some people, but I doubt it has much of an advantage for the average person.
I think it has a longer half life than caffeine for most people, though possibly it has an advantage in what it metabolizes to.
Another issue is that theobromine stimulation is relatively more in the peripheral nervous system (as opposed to central) than caffeine, so if you take enough to get the same mental stimulation, you may be more jittery than with the caffeine
Fascinating post but misleading owing to huge individual differences in how rapidly caffeine is metabolized (owing in part to CYP1A2 variants) and in the intensity of the reaction to caffeine (owing in part to ADORA2A variants), both of which turn out to have independent effects on sleep. A bunch of health and lifestyle variables also contribute to the variability.
Most of the substance of this post consists of (or assumes) generalizations about the average person and is therefore misleading owing to the extent of variability around the average. Caffeine doesn't have "a" half-life, for example. That value differs greatly across individuals. Some people, like my stepfather, can have a cup of caffeinated coffee at 10 p.m. and sleep just fine.
There's also evidence that the greater the amount of time between caffeine intake and bedtime, the worse people are at gauging the effects of caffeine on objectively-measured sleep quality. This suggests a need for nuance around anecdotal evidence,
(I'm posting on the topic this evening, but if you like Scott's post I don't recommend mine, because I'll be focusing on the narrower and less inherently intriguing question of how people might manage caffeine intake so that they don't undermine their sleep. Also, my post won't be as smart.)
I took a DNA test and read the results through Promethease, and one of the things it flagged was a gene variant for high caffeine metabolism, which fits considering I can drink coffee any time of day without it affecting my sleep. I thought this was something special until I read that 45% of the population has that variant. And even more interesting, things like smoking or wood stoves can affect that same enzyme which metabolizes caffeine.
I tip my hat to pharmacologists for being able to account for such variations in genetics and lifestyle when studying chemicals.
I think most of those are bunk; some are completely false, others are real genes but explain 1% of the variance. People who know more can correct me if I'm wrong.
Way more than 1% for objective variables like metabolic speed (CYP1A2) as well as "subjective" ones like self-reported anxiety (ADORA2A). (I don't claim to know more but only to have looked, because I'm writing about this.)
I think the issue here is that promethease may not have all of the pharmacogenteically relevant polymorphisms in its database. It may not be specific enough to be super helpful.
Also With CYP1a2 i the general understanding is also that it is (relatively) more sensitive to environmental factors (e.g smoke) than genetics when compared to most CYP enzymes. So often these results dont help tell us much about who will metabolize a med fast/slow as something like smoking status would.
Totally agree with you that info on SNPs from Promethease or anyone else won't help much beyond what you notice informally about your own caffeine response (although, as I suggested in my post on this topic, one could imagine benefits for certain kinds of people. A bunch of studies show that the greater the amount of time between caffeine intake and bedtime, the less accurate people are about objectively measured quality of sleep. So, the nonsmoker who stops drinking coffee before noon and still hasn't figured out why he's not sleeping well at night might benefit from one of those imperfect genetic tests).
Scott's dismissal of genetic influences on individual differences echoes the one-size-fits-all assumptions underlying his original post. Inter- and intra-individual variability is huge. Simple example: Scott repeated the common assertion that caffeine has a half life of 4 to 6 hours. This is false. Caffeine does not have "a" half life. The range appears to be something like 1.5 to 10 or more hours, depending on the study and individual circumstances.
OP seems to describe caffeine as sometimes making him sleepy. I've been known to sleep on arabica cappachino, consumed 2 hours before bedtime. My father would have a full cup of instant coffee at midnight, and then drop straight off to sleep.
Is this non-neurotypical?
I note that different sources of caffeine also seem to induce different psychoactive effects, in terms of sleepiness, where some of the smaller sources induce larger effects. I had originally concluded these were "non caffeinated" effects, but am wondering if it's merely a buffering problem.
Do people notice a differential effect based on how much buffering they use? I'm thinking specifically in terms of milk, which is funktastic, but the same effect might be induced by other substances, like coconut milk, widely used in vietnamese coffee.
At 76% of Americans are daily coffee drinkers, finding non-addicted people wouldn't be that hard.
Hot Chocolate at night, which contains caffeine, is generally not credited with keeping people awake. Even if given at high doses (1/4 cup of pure cocoa is a high dose, right?)
I think it's generally understood there's a substantial population of caffeine non-responders, presumably for genetic reasons. I guess maybe they technically meet the definition of "non-neurotypical" in that they're less than 50% of the population and caffeine metabolism is brain-adjacent?
My understanding is that caffeine can make a subset of the population both alert when they want to be *and* sleep better, because it's just stimulating all metabolic processes, including the ones that will let you sleep.
I actually briefly had an urge to quip in this comment section that I was missing the plot twist at the end of the post saying that caffeine already lets you sleep whenever you want (assuming you don't have e.g. spiralling thoughts or other caffeine-independent sleep blockers that caffeine would make worse), but reckoned it wasn't going to be funny enough to warrant a top-level comment, and not relatable for most people. :)
But yes: Some people (myself included) can sleep fine on caffeine! I don't think this is a non-neurotypical thing specifically as much as it is just varied biochemistry in general.
The liver failure review has me curious. I've had liver problems for a couple years now (primary sclerosing cholangitis as a result of ulcerative colitis). My bilirubin has been hovering between 2-3 mg/dL since 2024, spiking a few months ago up to a high of 14.3. Only time I had a fever was about a month ago, and that was (near as we can tell) the result of a bile duct infection which had probably been hiding for some time, given how much better I felt after the first dose of Zosyn compared to how I'd been feeling up to that point.
Anyway, all this to say I know all about G.I.P. funny colors, but the dark-colored urine I've had has almost always been accompanied by fatigue and a sort of brain fog (it was also very confusing, because nobody told me about that side effect for a while, so I was slightly over-hydrating to try and bring my urine up to its usual color).
I don't really drink coffee these days, but suffice it to say I won't be taking anything with it even if I do. Although, given I'd had fewer than ten standard drinks' worth of alcohol in my life up to the point of getting diagnosed with PSC, maybe a little liver stimulation would've done me some good.
What effect do these alternative stimulants have on stomach acid production? I have to avoid caffeine altogether for that reason, but would like a safe mild stimulant without that problem.
My impression is that heartburn from coffee is mostly a pharmacologic action of caffeine, but also somewhat due to the coffee itself being acidic. Taking it as a pill (like most of these alternative stimulants are) would help with the latter, but not necessarily the former.
Paraxanthine supposedly has fewer side effects than caffeine, and this is plausible since it should be a subset of caffeine, but I don't know of any studies checking whether the stomach muscle relaxation that causes GERD is specifically improved.
I read this trying to replace “sleep” with “heart”. It would be interesting to see if rutaecarpine could treat ER patients who are presenting with ventricular issues from excessive caffeine.
Sadly none of these options seem obviously well-suited to my own use-case of "drink Snapple or Coca-Cola or other such tasty caffeinated beverages in the afternoon-to-evening without their drug-effects lingering long enough to (inconsistently but often enough to be inconvenient) make me wake up in the middle of the upcoming night's sleep". Rutaecarpine sounds too high-risk to be worth the benefit, and the others are caffeine ~emulators rather than caffeine ~mitigators.
Hopefully one day someone will develop super-Rutaecarpine, reducing all caffeine effects to zero instantaneously while causing no substantial negative effects, so I can enjoy tasty caffeinated beverages free from the mild inconvenience of actually being affected by caffeine as a drug. (And free from the weird mildly-off flavor I've found decaf coffee to have relative to regular coffee for whatever reason, the times I've tried it.) Or, you know, at least something sufficiently closer to super-Rutaecarpine than to plain Rutaecarpine as to make it worth the tradeoffs of trying out.
Caffeine free coca cola exists and tastes exactly the same. Unfortunately it's only available in big chain supermarkets and even then it's inconsistent.
perhaps the people who aren't affected by paraxanthine are simply the same people who can drink caffeine in the evening without negative effect anyway.
Rutecarpine doesn't just induce CYP1A2, though. It's a COX-2 inhibitor, like Vioxx which got taken off the market for killing people. And it's also an inhibitor of other cytochrome P450 enzymes, so maybe the stories of liver damage aren't a direct result of rutecarpine consumption, they're a result of doing things like taking Tylenol alongside rutecarpine.
I have no problem consuming caffeine and sleeping right afterward. I assume it's just because I consume over a gallon of tea a day. I have basically spent the last 20 straight years highly caffeinated at all times.
For me the tea isn't really about stimulation. It's about being able to drink an unlimited supply of something all day that isn't water and isn't loaded with sugar.
If your main goal with caffeine is to preserve its use as a stimulant, this doesn't seem like too much of a problem because I would assume the consumption of it means you *don't* want to sleep anytime in the near future.
If you just like coffee/tea, then why not just consume such an inordinate amount of it that you develop a tolerance for it?
I've never seen strong evidence that being caffeinated all the time is a major problem. The only problem it's caused me is I get headaches if I stop - but I like tea, so I have no particular reason to stop.
> For me the tea isn't really about stimulation. It's about being able to drink an unlimited supply of something all day that isn't water and isn't loaded with sugar.
How do you feel about artificial sweeteners? There are a lot of versions of artificially sweetened beverages, including highly concentrated "flavor enhancers" that come in tiny squirt bottles that you can add to water. ("Mio" is the name of one brand.)
How's verdict of artificial sweeteners so far? The more I read the more I got confused. Especially in regards of how artificial sweeteners make you crave for more sweets (calories).
Similar story here, but black coffee. I've drank 5-8 cups a day, into the evening, for the best part of two decades and can fall asleep at any point if I'm tired. Normal blood pressure, bloods etc. Assume I'm just completely tolerant.
I point out that, if you're a big-time coffee drinker, drinking decaf in the evening can work as a placebo-stimulant. Works for me.
Also consider going to half-caff, now much more popular.
I used to have energy problems. I appear to have solved them by either one of, or all of these (I don't know as I started them all about the same time):
- weightlifting 2x / week
- cordyceps powder in my morning coffee (I have a mix actually: cordyceps, creatine, L-theamine, turkey tail, cocoa, reishi).
One more clinically important CYP1A2 interaction: tobacco smoking induces the enzyme, so quitting smoking can suddenly make your usual caffeine dose much longer-lasting. In one study of heavy smokers, caffeine clearance fell 36% within a week after quitting (Faber & Fuhr, 2004).
This is an effect of smoke, not nicotine, which doesn't meaningfully induce CYP1A2. So it can happen even if you switch completely to any NRT.
Huh I didn't read that, but when I was at 'beer school' for a DWI I learned that the step response (how long does a single dose remain in your system) Is like 6 hours for alcohol and 2-3 hrs for THC and nicotine (When consumed through the lungs, not eatables) So I can have a bong at 6 and be asleep by 10. (Or a cig, but I gave up nicotine.)
Alternatively, there are Alpha-2 Adenosine Receptor Inverse Agonists like Sipagladenant that are sold on gray market sites. If you're doing a comparison of these drugs, its worth looking into the non-supplement (and thus more rigorous) side of the literature.
Pre-workouts are among the top selling supplements. Annual sales in the US total about $22 billion according to some estimates. And most are pretty much just mega doses of caffeine.
What I *really* want though is a drug that removes caffeine resistance (or at least prevents it from building up.)
As Michael Pollan notes a few times, a naive caffeine dose is the best drug in the world, but resistance builds up very quickly. An ability to maintain that initial high would be incredible.
Capitalism needs to come up with the idea of selling <200mg doses of paraxanthine. That's the smallest dose I can buy (from the UK), but I usually titrate my preferred caffeine dose from 50mg pills.
I expected the (in)famous Gwern exploration of nicotine for 5-to-10 energy to get a mention. Would love an ACX discussion of nicotine.
Anecdotally, I tried it and found it to be WAY MORE addicting than I expected, even accounting for being told that many people had read the Gwern post and found it more addicting than they expected.
I never got to daily use, but still to this day, circles with the approximate dimensions of the Zyn containers have a much higher salience in my visual field, and immediately jump out to me the way that tapeta lucida do.
There is a simple reason why paraxanthine could be ineffective that seems to be missed here. Coffee contains caffeine which is metabolized to paraxanthine and then to inert compounds by CYP1A2. Coffee also contains methylliberine which inhibits CYP1A2. So isn't it likely that the problem with taking paraxanthine alone that it is metabolized much faster than when it is taken with coffee and methylliberine? Seems like the obvious step would be to try paraxanthine + methylliberine (which is safe because coffee is safe).
I drink coffee with muuuuch milk in the morning - 2 spoons of ground coffee (or 1 normal coffee, 1 decaffeinated) in a 1 liter french press gives me like 3..4 large cups of coffee to which I add about 0.3 liter of milk. 3 or 4 Lattes is similar. Another cup of coffee after lunch. Thats working mostly for my heavy focus work in engineering, planning and fixing complex systems. if I drink more and or later, I get sleep issues.
But I've got sleep issues anyway since I'm a conscious human. So since a couple of years I mostly take 5mg Zolpidem just to have a reliable sleep. Sometimes 10mg if I'm agitated by late social interactions - or I drank too much coffee...
> So which of these options for nighttime caffeine is the best? As always, this is a complicated question, with each alternative having its own set of pros and cons and being right for dif - just kidding, it’s paraxanthine.
Thank you for having the balls to make a judgment call.
Paraxanthine is alrady in my toolkit. I'm a slow caffrine metabolizer, which is supposed to apply to paraxanthine as well, so didn't try it in the afternoon yet. But I find that normal coffee + paraxanthibe 200mg leads to a better day and better sleep than just doubling coffee intake.
I've been reflecting recently that society has come to fetishize productivity at the expense of our health. When we feel tired, it means the body wants rest. So why not rest? Instead, our mental response to fatigue is now at odds with our physical need. Instead of feeling tired and thinking "I'll take it easy," the response seems to be "I don't like this. I need to be 'productive' and get 'things' done"- so then we do the exact opposite of what's required and take drugs to mask how the body actually feels and push through it. Why do we feel compelled to do this?
Really enjoyable piece, and laying the options out by mechanism instead of just by product is what makes it click.
What falls out of your own framing is that the three approaches are all the same CYP1A2 lever pulled at different points: rutaecarpine induces the enzyme, methylliberine inhibits it, and paraxanthine skips the enzyme by being the product it makes. Which also flags the quiet problem with rutaecarpine, since revving CYP1A2 to clear caffeine faster would speed clearance of the paraxanthine doing the useful work too. Paraxanthine sidestepping the whole step looks like the cleaner bet for exactly that reason.
Paraxanthine works well for my wife.
It helps you get her out of your system quickly?
Can I just say it's fun to me that I encountered you via YouTube music, most notably the excellent Skill Issue https://www.youtube.com/watch?v=YfSGXmYJVdI before just now discovering you are connected to ACX in any way.
Aww, thank you!~
But how well does your wife work?
Ironically, I'm going to have to get some sleep before reading this properly because I can't drink coffee after midnight
I highly recommend nicotine patches as a short-term stimulant option. Great for getting work done at 9pm.
+1 the answer is Nicotine. A 3mg pouch is a reliable boost to focus, energy (and probably IQ) with (for me) no observable impact on sleep after an hr or so.
Nicotine's nootropic effects are well-supported. I have it on good word that Professor Henry Lester's Neuropharmacology of Addiction class at Caltech convinced two students to *start* using nicotine.
I heard many people who tried this option for this reason ended up getting addicted.
Next up: The Quest For Nicotine You Can Have Without Getting Addicted? https://pmc.ncbi.nlm.nih.gov/articles/PMC6493450/
A valid concern, but caffeine is addictive too.
Can most people not just taper off over 1-2 weeks?
I'm pretty sure nicotine is much more addictive and has worse withdrawal symptoms than caffeine. They're not really comparable.
How does nicotine addiction compare to liquid meth addiction? Available commercially, in pill form, as sudafed.
Sudafed does not actually contain methamphetamine.
https://pmc.ncbi.nlm.nih.gov/articles/PMC8152226/
True. This was a comparison made by someone who's had both legally.
"Pseudoephedrine may also be responsible for a false-positive urine test for amphetamine and methamphetamine. The structural similarity to these drugs means that they may cross-react in a test using the immunological method"
i have quit both caffeine and nicotine. nicotine definitely has worse withdrawal symptoms in my experience but they're both pretty manageable. it's definitely not incomparable. caffeine withdrawal makes you extremely tired and sluggish for a few days. nicotine withdrawal, in my experience, was more about fighting the urge to smoke a cigarette than about any particular acute symptoms.
I have found that while the withdrawal effects for caffeine and nicotine are similar, nicotine's "mind game" is much stronger than coffee. Two weeks after quitting coffee, I no longer crave it. Even months after my last use of nicotine I still think about it ~daily, and find myself making excuses to use it.
Nicotine addiction is, I think, substantially worse. It's more addictive, aiui more harmful to be addicted to, and way more expensive to be addicted to (at least in the US/most western countries where it's taxed to the moon)
Take care not to conflate nicotine with tobacco.
Everything I said I meant about nicotine, not tobacco
I am seeing some places don't tax the patches though, which I did not know, so I stand corrected on this
+1, never used tobacco, and I agree with what M M said, minus that it is more harmful: that's not as clear to me though it is probably true.
IMO I’m not sure being addicted to 3-6mg (1 or 2) pouches a day is meaningfully different than being addicted to daily caffeine. Part of what makes it addictive is that it’s *enjoyable* to drink coffee and satisfy that craving. I *enjoy* my 3-6mg of nicotine a day and don’t have a strong urge to take more. At that level, it’s mildly annoying but not disruptive to skip any given day—it’s missing out on something that feels good and perks me up. I think “addiction” is a loaded term (for obvious reasons) but functional addictions (as are commonly the case with caffeine) don’t seem particularly problematic, and can be a nice part of life (or even healthy!), in the form of coffee or tea or pouches or lozenges or gum.
I think that's true, and if you have no temptation to escalation, you are probably alright, but I found nicotine much more persuasive than caffeine: I have never, as I have with nic, found myself making excuses for why I *have* to take it today, even though I hadn't planned on it.
I had the impression that neither caffeine nor nicotine was biologically harmful in anything like a normal dose. Am I wrong?
Of course, something can be harmful to your lifestyle and through that be harmful to your body, but I don't think either of these will turn you into an up-all-night degenerate like amphetamine might.
Addicts don't take normal doses of substances. If you look through the quitting Zyn subreddit, you'll see many accounts of unpleasant side effects of nicotine addiction - anecdotal, yes, but I don't think anyone doubts that constantly taking doses of nicotine that would be lethal to someone without a tolerance is likely to cause side effects. It's much harder for a caffeine addiction to get that bad.
Nicotine is substantially more addictive.
I'm not sure what this comment is meant to be saying that M M didn't say already.
Perhaps people addicted to nicotine patches could try to wean themselves off by taking small doses of cigarettes.
Similar to how if you find it too difficult to buy Sudafed at the pharmacy due to legal restrictions, you can always use readily-available methamphetamine as a precursor to make your own:
https://www.semanticscholar.org/paper/A-Simple-and-Convenient-Synthesis-of-From-Hai-Hakkenshit/06b66e418bbd93184228fbc5c0395eb1bf15a5d3
Remind me not to go to you for advice if I ever accidentally swallow a fly.
I occasionally (less than once per month) use a 1mg oral nicotine lozenge as an evening stimulant, and have never been at all tempted to escalate, for whatever that's worth.
Nicotine patches seem more dangerous, because they're typically a much larger dose released over a longer time span. Why would you use an extended-release product when the whole point was to find a shorter-acting stimulant?
Meanwhile, vaped nicotine is short-acting if you don't redose, but you *will* redose, because it hits so quickly that you get a clear "rush" and the direct stimulus-response association is really compelling.
At least for me, oral lozenges sit in the sweet spot where they kick in slowly enough as to not trigger Skinnerian conditioning, and wear off quickly enough that the system doesn't have time to form a physical dependence. YMMV.
I use 3mg nicotine patches for this purpose, similarly no temptation to escalate. Haven't used them in months actually. I take the patch off after about 30 minutes usually. I think patches are better than lozenges because less sensation/activity associated with the chemical stim. But only if you can find low-dose patches and follow strict rules about limiting dose and frequency.
While I find it helpful and safe, it probably isn't good to recommend to others because no matter how many times you reiterate that they must use low dose and sparingly, someone will ruin their life with a daily high dose and then blame you.
You can just cut the patches. They tell you not to, but it's fake news.
Ex-nicotine addict here, got addicted from nicotine gum then moved to vapes and then cigarettes. It was really tough to beat the addiction, do not recommend
I did this with nicotine lozenges and while I do take them multiple times a week, I’ve gone off them for incidental reasons (forgot to order more, vacation, etc.) without any apparent negative effects. I think they’re better than the patches since they subjectively seem hard to get addicted to.
Same here. Only with the lozenges (not gum, not cigarettes, certainly not vaping. I haven't tried patched. But with the lozenges I take them a few times a day and when I can't take them for a week or whatever it's no big deal. In fact, I sometimes just forget to take them, despite being obviously dependent on them. So it's easy to stop
Yeah I tried taking nicotine as a nootropic back in the day following Gwern's post on it. Found that I very quickly found myself making excuses to use it more and more, and reached the stage where I was anxious when I didn't have it. And quitting it was very difficult
I had the same experience!
I tried it for focus and quickly became reliant. Getting off was not very pleasant, but hey, I give no weight to my anecdote.
It's... difficult to get addicted to nicotine patches on accident. Not impossible, of course not, but they're by design less reinforcing than any other ROA. I get no noticeable "rush" from a functional dose (7mg/24hr patch for me), and I can avoid tolerance by simply not taking them consecutively.
Interestingly, I did try to get addicted to Zyn (nicotine pouches) on purpose in an effort to induce a behavioral addiction to Leetcode. This backfired, as it turns out I love Leetcode and hate Zyn, and it was a while before binary trees stopped making me nauseous. But that's another story. I was put off nicotine for a long time.
I agree about patches, but after using cut up patches for a while, after having used zyns, thinking the patches were safer, when I went back to zyns, I found them much more addictive than I had the first time: never experienced any "nic sickness" which would have been, I think, a barrier.
Yeah, I wouldn't recommend Zyns, they're definitely designed to be addictive.
Are you sure they got addicted after using *patches*?
Pouches and vapes are much more common but also much more addictive/habit-forming. Gwern explicitly addresses this in the infamous article, but no one seems to have actually read it.
Patch or pouch?
Having experienced nicotine (pouches) addiction and withdrawal, I cannot recommend this. I ended up quitting because of exhaustion caused by the nicotine-induced highs and lows.
How often did you use it?
6 medium-strength pouches a day, roughly in intervals of 3h. Taking less would cause annoying withdrawal symptoms, taking more something discomforting, like light headaches.
Was it the peak use or what you started with?
I'm using them about once a month when I need a boost
This was my stable use for a couple years, with short-term withdrawals in between. I started, probably like most, with the occasional use, then daily, getting dependent and finding a balance. Interesting that you are able to use them without getting hooked. Once a month means 1 pouch per month, or one period per month where you use a bunch?
1 pouch per month on average, but it's very irregular
I also drink mate when I need a bit of late-night boost - probably twice a week on average and coffee most weekday mornings, that's the closest thing to addiction I have
Same. Use the gum. The blister pack is annoying enough not to get addicted.
In the 1996 remake of Diabolique, the character played by Kathy Bates struggles with one of those nicotine gum blister packs, “What’s in here? A piece of the True Cross?”
What's the cancer risk?
Nicotine is not a human carcinogen. Since cancer is very complicated, there are some contexts where it can act as a promoter for tumors and so on, but the bottom line is that you shouldn't be worried about nicotine giving you cancer.
Has anyone been able to find a source for 50mg paraxanthine? I can only find 100mg and 200mg on Amazon.
One can buy gadgets to break pills neatly in half, or even a third or a quarter, like this bad boy:
https://www.amazon.com/Splitter-Quarters-Accurate-Stainless-Transparent/dp/B0GTQ25KC7/
But obviously they won't work very well on liquid-filled capsules if that is how paraxanthine is sold
Although liquid capsules are easy to split, by mixing into a small quantity of warm water; I have 15 mL test tubes that are great for this.
Dissolve the capsule in a small measured jar of decaf, then stir and measure out by the tablespoon.
I hadn't even considered mixing into decaf, my main objection is that I *enjoy* drinking coffee.
...DM me if you're interested in starting a business together.
Possible alternative lever to pull - what if we found a way to induce ADHD in people? Then caffeine might make it easier for some small proportion of those people to sleep</humor>
(though; I highly doubt the induced ADHD + caffeine sleep would be easier or better than the no-induced ADHD with no caffeine sleep)
If you want to wake up in the late afternoon, there are other methods that won't keep you up five hours later. Go running, lift weights, go swimming, take a cold shower, go to the sauna (or maybe the NYC subway platform). If the sun is still out, get some sunlight. Some of these methods can actually help you fall asleep later.
Regarding "It’s a simple molecule, so brand doesn’t matter much; this one is the cheapest I can find" - supplements often don't contains the claimed proportions or contain impurities which can be unhealthy or counterproductive. So it's worthwhile to stick to trusted supplement brands (or to Enfinity in this case since they're known for making paraxanthine).
Glad to know that somebody else had the same reaction reading this post (to the first part of your comment).
Im surprised he doesnt mention this. Maybe he assumes everyone who reads his blog knows this about supplements?
Scott's moderately sanguine on general Truth In Supplements: https://astralcodexten.substack.com/p/how-trustworthy-are-supplements
...but I don't know how much that holds up for less-common ones, instead of the big-name nootropics/vitamins/whatever where there's enough volume to ensure decent corrective mechanisms for perfidy.
Why would you want to wake up in the late afternoon… the goal is to have a boost of energy at night without the delayed sleep
He means an energy boost in the late afternoon, I was also confused.
Thank you for this reminder!
Most of those are fair, however running, swimming, and lifting, take up more time than a lot of people would like. Also, one of the benefits of stimulants is you can take them whenever, you can't easily swim, run, lift, or go to the sauna when you feel tired at work, or when you sit down to do some work in the evening and only then realize you're not mentally energized.
One of the reasons I want a stimulant to wake me up in the late afternoon is so that I have the mental and physical energy to do something like lift weights or go outside for some sunlight. Without the stimulant, I'm more likely to gradually lose energy and eventually just collapse onto my couch. Even when I was lifting weights fairly seriously 3-4x/week, it usually took a pretty significant expenditure of willpower to get me into the gym after work, and that's not something that ever really improved. If I had been able to drink a cup of coffee at 4:30 without impacting my sleep, that would have helped quite a bit.
What I really need is a way to quickly reset my caffeine tolerance…
This will be neither quick nor easy. Last time I went low caffeine I was tea-bagging in cafes. I.e. sucking on black tea bag to make sure I get the last piece of caffeine.
Easiest is to slowly go lower over about 2-3 weeks.
You can have regular caffeine at night too. Just stay up.
Never really understood this midrash. Like, shouldn’t the kapparah be waking up early to learn, not staying up all night and then sleeping all morning?
House generally cooler at night, particularly in summer. Hence more efficient thinking.
The more I abuse my body with coffee, the easier it is to collapse into sleep from burnout at 5am.
Sounds like a lot of my days as an undergrad. Sometimes I wouldn’t go to sleep at all and enjoy the added bonus of not having to deal with bed-head hair.
Coffee has little to no effect on my sleep. I often drink a coffee like an hour before I go to bed. I don't feel any different compared to the days I don't drink one so late.
This is often the case for people with ADHD. Coffee calms us down.
Coffee definitely doesn't calm me down, in the sense that if I drink several cups I'll get a bit jittery, but it doesn't prevent (or even affect, subjectively) sleep. Going to sleep is just a matter of choosing to do so, for the most part. Like Trudel, I sometimes drink a cup an hour or two before sleep.
Typo:
> liberine. If There
The ‘if’ shouldn’t be there
Jesus Christ. I can't imagine liking caffeine enough to care about this but I guess I'm just jealous.
Yes. I mean, I like coffee but it would never have occurred to me that not being able to have (specifically) caffeine in the evening was a problem for someone.
The framing of this post was confusing.
By the end of it I figured out that the goal here was: "I am sleep deprived or otherwise need a stimulant, for a burst of work starting at 5pm, but I still want to get to sleep at 10pm".
But the intro (and the early focus on rutaecarpine) made it sound like the goal was specifically about evening coffees, so I was thinking the whole time: why not just drink decaf?
It's not about caffeine. It's aboug wanting to be productive at all hours. I often have an afternoon dip. A nap and a coffee work perfectly... but they also absolutely ruin the next night.
A short acting stimulant would be awesome for a lot of reasons.
I get that, caffeine affects people quite differently, it doesn't stimulate me enough to interrupt sleep let alone allow me to be more productive. There are a lot of short acting stimulants actually, it's just that most of them are controlled, coca tea or a small dose of methylphenidate would fit the bill perfectly.
Not my quest.
I've missed posts like this, it's been too long. When was the last one...Lavender's Game for silexan?
All three of these "lactaid but for coffee" sound too Trivially Inconvenient for me to actually utilize on the regular, unfortunately. Definitely can't risk liver damage either, what with other meds + family history already putting me on watch for hepatitis and other failure modes. Which is too bad, cause decaf still tastes like shit in current_year, and it was an unwise decision to accept a recent gift of a giant bag of sugar-free Coffee Rio candies. Grateful for the For Science! exploration though.
Funny enough, my coworker just mentioned the silexan post to me yesterday. It was the first time somebody else has ever brought up the topic of Scott Alexander in conversation with me and caught me totally off guard.
If you’re in the US or can get it shipped - try Peet’s Decaf pre-ground House or Major Dickasons roast. Use more coffee than usual. Pretty dark and strong and fools my regular coffee-loving brain. Dramatically better than other decaf.
I do like Major Tom Dickason's roast! Wonder if the Peet's near my work carries the decaf version. That's usually where I end up wanting late-night caffeine or "caffeine", and of course working at a grocery store, we'll just use whatever one of our brand is on hand...but decaf sells so poorly that we've discontinued almost all of them. Worth looking into.
They have it on Amazon:
Major Dickason's - https://a.co/d/03sP5zAK
House Blend - https://a.co/d/06gXVlRF
On the tea side, Ito-en's decaf genmai-cha is the only decaf tea I've wanted more of: https://yamamotoyama.com/products/organic-decaf-genmaicha-green-tea-bag
Agreed, I've always especially enjoyed posts like these (even though in this case I'm not really a caffeine guy at all) and was very happy to see it land in my inbox.
> Maximum safe dose is probably between 300 - 400 mg daily
I don't think you meant to include "daily" there, if we're using "how much of this would you be getting if you just started with caffeine instead" as a baseline. Some people routinely consume on the order of a gram of caffeine per day (I have a friend who drank 6 cups of coffee per day when she worked at SpaceX about a decade ago). Probably you meant "at once"?
(I, personally, would be climbing the walls if I took even 200mg of caffeine in one go; I'm pretty caffeine sensitive and stick to a single cup of tea that usually lasts me from 11 am to 5 pm.)
Your point makes sense, but the FDA does say 400 mg caffeine daily, and I don't know how much it's worth distinguishing daily from per-dose in a chemical with a 10 hour effective half-life.
~2018 I was having heart palpitations and went to my doctor. He asked how much coffee I was drinking daily. My answer of "ohh about 2 pots of coffee" (made extremely strong) prompted a "hey dumbass, stop doing that it'll go away" (it did). TIL that 400mg daily was seen as lot...
If you just want an instant energy boost that clears quickly and lets you sleep afterwards why not just go for cocaine? Surely there's some regulatory arbitrage here where you can give it a brand name, reduce the dose, and package it into a pill that you can sell on Amazon.
I thought the same thing. How much we’ve lost to paternalism!
Coca tea is a huge thing (and totally legal) in Peru.
I've wondered for a while if the addictive power of cocaine is actually much higher than caffeine, or if this is just a case of https://www.astralcodexten.com/p/drug-users-use-a-lot-of-drugs
Coca tea tastes worse than coffee, makes your mouth numb like you've been to the dentist and keeps you awake like coffee.
Khat is also a huge thing. Unlike Coca, it tends to lead to an entirely unsafe country (outdoors) due to traditional patterns of consumption, at least I assume there's no equivalent to Khat hour in Peru...
Sudafed is about as awake-making as liquid methamphetamine, being not very chemically different. Really interferes with sleep though.
Methamphetamine has a very long metabolic half life so that's not surprising.
You all know that decaf is an option, right?
If your reason for drinking coffee at 5pm is to enjoy the taste or the feeling of a hot drink, sure. But if you're trying to get an energy boost for the evening without interfering with your sleep then decaf won't meet your objectives
Why not hot chocolate? That'll give you the caffeine, and vasodilators too.
Well the point of the whole post is to find an alternative to caffeine because the half-life of caffeine is too long. If you just switch your source of caffeine you haven't really changed that.
A cup of hot chocolate has an order of magnitude less caffeine than a cup of coffee (personally I've never experienced any boost in energy from chocolate of any sort, but I don't have a strong response to caffeine anyway). So it sounds like your suggestion is just to reduce the dose? In which case you could just have a smaller amount of coffee if you prefer it to hot chocolate (and avoid the glucose spike to boot)
I'm going to guess that's a normal cup of hot chocolate. I routinely add four times that amount (it's a very good antioxidant source -- best fruit/vegetable on the market).
That’s what I have in the evening. (I gotta be careful with that, too, tho... too much theobromine will give me a headache)
Anecdote: I tried rutaercarpine a few times and it tragically did not seem to help at all with letting me drink an afternoon coffee and then get to sleep at night.
Rarebird makes paraxanthine coffee beans -- they start with decaf and then infuse them with paxanthine, iirc. I brewed this every day for a while, it was actually pretty good, and kind of felt like ordinary coffee but shorter-acting.
Expensive, but I liked the taste and effect. I micro-dose and time stimulants such as coffee, and this product and strategy suited me very well.
Why not use l-Theanine which is what makes some people able to sleep with green tea?
L-theanine removes "jitters" from caffeine but doesn't remove the actual adenosine antagonism that prevents sleep.
It doesn’t have direct antagonism, but does improve sleep on caffeine in animals via other pathways… “In addition, since L-theanine was shown to have a relaxing effect and gamma-aminobutyric acid (GABA) was an important inhibitory neurotransmitter, the mixture of L-theanine and GABA had a positive synergistic effect on sleep behavior, including sleep quality and duration in caffeine-induced awake rats, and the mechanism might be that the mixture could promote the expression of GABA receptor, which was conducive to sleep” https://www.frontiersin.org/journals/nutrition/articles/10.3389/fnut.2022.853846/full
This was fascinating, I don't drink coffee in the first place though
What exactly is the point of coffee? Unlike tea, most people seem to be drinking coffee purely for the stimulant effect rather than because they actually like the taste (as seems to be acknowledged both Scott's analysis and the comments section..) and as I understand it one's metabolism adapts to regular coffee-drinking by reducing one's baseline alertness to negate its effects anyway?
So, long-term, the only effect of coffee seems to be to temporarily make you as alert as you'd be if you never drank coffee, only with more side-effects (including being *less* alert pretty much any time you're not drinking coffee..) and at noticeable expense and hassle?
I get "fresh" coffee -- yes, that's with days of resting, it's still chemically active and very potent. Not only does it taste "wild", as it's got lots of chemicals most people never see, it's also a very potent source of antioxidants. I also do a minimal-time extraction using an espresso machine, so I'm getting a very light dose of caffeine per "other psychoactive/physioactive ingredients."
I'm not sure why people drink coffee if they don't like the taste -- but everyone seems to like the smell of coffee, even kids who generally hate the taste. I was that kid, yes, I admit.
I imagine your setup is probably fairly expensive (compared to, say, instant coffee or teabags or whatever), and so it just has different tradeoffs? Scott has answered my fundamental issue downthread (viz. if you only drink 1-2 cups of coffee per week you don't actually develop a tolerance in the first place) so I couldn't say it isn't worth it, just that "ultra-fresh coffee + fancy high-pressure machine" doesn't seem to match how most people I know drink coffee.
For the taste, I'm not sure. The impression I get from this thread is that many/most people drink it for the stimulant effects (so for example the problem can't be solved by everybody just drinking decaffeinated coffee and thus getting the taste but not the stimulant effect) - but perhaps over time one comes to acclimate to and enjoy the taste too, I don't know. (I have an Italian friend who loves coffee and I strongly suspect that she would say that if the coffee-as-stimulant drinkers all switched to your ultra-fresh + machine setup they'd probably all start enjoying the taste too!)
For the smell - personally I can't blinking well stand it! I find being in a coffee shop, or even in somebody's kitchen whilst they're brewing coffee, oppressive and slightly nauseating. I think I might be in a minority here though, as I've never heard another non-coffee drinker complain about the smell.
In From Russia with Love, Fleming writes Bond as a coffee drinker.
I’m going to have to paraphrase from memory here, but he blames tea drinking culture for sapping British vitality and efficiency. In that passage of that novel at least Bond prefers the jolt of coffee.
Typical insipid British half-measures. If Bond had started the day with Red Bull or Monster Energy he could have foiled the Soviets by about page 27.
In Thunderball Fleming has Bond taking benzedrine before his underwater assault on Largo’s yacht. There was some experimental military use of that amphetamine in WII. I think the era was still fairly fresh in Fleming’s mind when he wrote Thunderball.
I’ve read that the US military prefers things like modafinal to amphetamines for extending alert attention now. It doesn’t cause the euphoria that can cause careless mistakes.
Just don't drink it so often that you get tolerance. A few days a week is fine imo.
Interesting, thanks! I imagined it was a smooth homeostatic mechanism, not a "less than this and your metabolism won't compensate" cutoff point. Does the amount of caffeine matter (ie. super-caffeinated coffee will build up tolerance quickly even at fewer cups per week) or is it literally just the number of cups-per-week that matters?
(The Wikipedia page cites some research at Johns Hopkins that suggests just 100mg per day (one small cup) is enough to develop a physiological dependence and I'm pretty sure most coffee drinkers far exceed this so I'm not sure how effective "just don't drink it often" advice would be for most people - but to misquote Churchill you've already established what the mechanism is and now we're just haggling over the right number of cups...)
https://en.wikipedia.org/wiki/Caffeine_dependence#Definition
I don't know. I know I drink coffee 1-2x per week and don't get tolerance, and I know from experience with patients using Adderall that if they get tolerance taking it every day, we can usually reverse/prevent it by bringing them down to 3-5 days/week.
I wish that was the case for me. I have adhd and get tolerance to stimulants insanely fast.
3-5 days per week is far more often than I can take them. Heck, 3 -5 times per month is sometimes pushing it.
Not sure I agree about the taste thing. I drink coffee only occasionally but I enjoy the taste if it's made in my preferred style (dark roast, no milk or sugar), and most people I know who drink coffee regularly claim to enjoy the taste as well, to the point that they buy more expensive beans and test and refine different brewing methods to optimise for flavour. There's a whole culture of fancy coffee shops that try to optimize the taste as well.
I don't know much about coffee shops other than (World Bank Economist) Tim Harford's analysis in 'The Undercover Economist', where Harford posits that what fancy coffee shops are actually optimising for is customer-signalling, not taste, and that most of their "fancy coffee" is fundamentally roughly the same. (I don't know this from personal experience, I admit, and probably a pop-Econ book isn't where one ought to look for a study of gustatory experiences...)
For the "buy more expensive beans and test and refine different brewing methods", human nature being what it is this seems like exactly what people would do if they intended to drink coffee at all, regardless of whether they were drinking for or despite the taste.
You and most people you know liking the taste, however - that's a good point and well-taken! I'm still curious about the causality, though! When you say "people I know who drink coffee regularly", do you think they came to like the taste because they drink regularly, or do they simply drink regularly because they like the taste?
It's like wine - there is a component of actual difference in taste, and some people have more skill or natural affinity for appreciating that difference. Then there is also status associated with knowing good wine and being able to buy expensive wine - and therefore lots of people who pretend to be able to know the difference, and a large psychological component to tasting something special.
Of course it is all "roughly the same"! It's all fermented grape juice, or grounded coffee beans diffused in hot water. That is not really a very interesting observation.
Your comparison to wine is interesting and useful - thanks!
For "of course it's all roughly the same", what matters is not what the available choices are made from or how they're made, but the *outcome* of choosing one over the other. I bought a motorcycle tyre today: the website offered me a choice of [checks history] twenty different tyres to choose from. Some are for road use; others for off-road. Some have a lifespan of 7,000 miles; others could last for 15,000 miles or more. Some are optimised for track use, others for urban use, others for touring. Some are rated to 500lbs and 130mph; others to only 400lbs but 170mph. Some are adapted for dry grip at the expense of wet grip, and others vice-versa, and still others for reduced grip across the board in favour of longer lifespan. There's about a 300% difference in price between the cheapest and the most expensive. You could just as easily say "they must all be the same! They're all just circles made of rubber, all exactly 17 inches in diameter, all made using the same process of extrusion and vulcanisation!" but in this case you'd be mistaken - perhaps fatally so.
Offering lots of options is inefficient. It costs more in storage space, complicates inventory management, leads to more breakage/wastage, splitting your purchase orders across 20 product lines reduces economies of scale, etc. When there are lots of options, we generally suppose there to be lots of genuine variation, reasoning that the vendor wouldn't go to the trouble of stocking all those products if they were all the same. Sometimes we're quite right to think this (as in the case of the motorcycle tyres) and other times we're wrong (as might be the case with coffee or wine). But in either case there's always some level of abstraction where you could say "They must be the same because they're all [X]!" regardless of whether the outcome of your choice would be the same of not. I share 99% of my DNA with a monkey - of course we're the same!
I'm not sure if I understand what you mean exactly. I think you are agreeing with me?
My point was that while I agree that coffee and wine is roughly the same at some level of abstractions, just as all tires are donut shaped rubber, that is not really a interesting observation in itself, and you have to go into more detail.
I don't really agree that coffees or wines are all roughly the same at all. First, there clearly exist different levels of quality that most people will agree on. For example wine that tastes like alcohol or luke warm, sour coffee will be considered inferior by most people. When it comes to the luxury coffee in coffee shops, I can agree that they mostly are somewhat similar, and serve the same purpose. But that is missing the point, because what people often want when they go to those shops, beside high qulity coffee, is novelty i.e. they want to taste some type of coffee they have not tasted before. (Ofcourse they may also want other things - like to look like successful people who goes to this kind of shops, but it is not the only reason to want luxury coffee).
I completely agreed with your first, "It's like wine" paragraph, yes (or rather - I didn't so much agree as was convinced by it!)
For your second, "of course it is all roughly the same [...] that's not an interesting observation" paragraph I don't agree: there's a level of abstraction where everything is the same ("of course it's all the same - it's all baryons!") and a level of granularity where everything is different, and it's valid and useful to be able to say that things are different when we'd expect them to be the same, or vice-versa.
Wine and coffee are perhaps a poor example to make the point, because some people believe there's a vast material difference between different options and other people believe much of the difference is in one's imagination. So: Neurofen Express Period Pain tablets and Neurofen Joint & Back Pain tablets are 100% chemically identical; the only difference is that the "Joint & Back Pain" tablets come in a grey packet with a drawing of a person's back and shoulders on it whilst the "Period Pain" tablets come in a pink packet entirely devoid of anatomical drawings.
In this example, the two different painkillers are the same product but consumers believe them to be different products based on their different name/presentation/marketing/whatever. This is an extreme example - probably most different types of coffee/wine aren't literally chemically identical - but the point is that there absolutely do exist things that are presented as different but are (at a reasonable level of abstraction..) actually the same, and some people absolutely will form preferences for one over the other. Therefore, it's perfectly valid and reasonable to say "these products are presented as being different when actually they're virtually the same".
(As for whether wines and coffees are genuinely materially different or whether the difference is mostly imaginary, I couldn't say - de gustibus non est disputandum - but I am reminded of a study (I think maybe by Prof. Richard Wiseman?) wherein experienced wine-tasters were given a selection of red wines to taste and report on, one wine was secretly a white wine dyed red, and hardly anybody noticed anything unusual about the white wine....)
To your question, I can't speak for others, but in my case I started drinking coffee many years ago while working a summer job at a Starbucks. Part of our employee training was to taste some of the drinks and get a sense for how they should taste. Prior to that I'd only rarely tried coffee and hadn't enjoyed it, but somehow while working there I pretty quickly I developed a taste for it, and that's why I started drinking it frequently. I was then a daily consumer for years, but more recently have reduced my intake since I wanted to be less stimulated.
I drink coffee every morning. The ritual of pouring a cup, slowly sipping it as I wake up, and feeling the caffeine withdrawal wear off, is very enjoyable for me - and it helps me poop as well. So even if all it is doing is fighting off withdrawal and making me feel normal, I would still accept that for the joy that it brings my mornings.
Makes total sense that the ritual of making it is a large part of the appeal! cf. the half-dozen-odd different types of tea ceremonies, cocktail mixology, flair bartending, pipe-smoking, etc.
I don't have citable evidence, but from my experience and from talking to people, I have a different impression of the effects of caffeine tolerance:
It seems that if you take a steady daily amount of caffeine at regular intervals over a long period of time, you'll converge on a state where when you're on caffeine, you're somewhat more alert and focused than if you hadn't been taking it at all. Something like this is verifiably true for most people prescribed daily amphetamines, with exceptions.
This is an irresponsible post. Delete it.
How so?
posting about buying stuff that can cause liver damage, whatever caveats you put, is still not a good idea unless you dont put links in the post and very clearly advise against all of it. that is not the tone of this post though.
Two objections: 1. Censorship of true and helpful information because you’re worried the audience is too irresponsible to use it wisely doesn’t have the best track record, I think.
2. Suppose we don’t worry about the philosophical point above. You still need to consider the counter factual. First, let’s try a simpler example. Suppose Scott had said “don’t drink and drive”. Someone objects: “how dare you say drinking is okay?!? Don’t you know that alcohol causes liver damage?”
So Scott deletes the post. But alcohol is still a thing that people will drink sometimes, and now they don’t even get a warning that they should avoid driving afterwards.
What went wrong? You assumed that by denying someone vital info, you are avoiding tempting them. This could work if you control it at least know about the rest of their infection environment, but in practice you don’t. So they end up doing something (much) worse than what you wanted to prevent.
Typing on my phone is very annoying so applying the drunk driving logic to this post shall be left as an exercise to the reader: how many different ways can you think of that someone without access to this post would do something even worse than what you’re afraid of?
Who are you to say what perfectly-legal supplement a blog may or may not bring to the attention of its audience?
> is still not a good idea
How do you know? What the expected QALY loss due to potential liver damage? What's the expected QALY gain due to enhanced functioning? Have you done that analysis?
This is an offensive comment. Delete it or I'll report you.
As they say, safety third!
It should be noted that virtually all of the chemicals in plants that get us high (including caffeine, nicotine, THC) are in fact toxins produced by the plant to ward off predators. So it's not surprising they can also be toxic to us.
Hot peppers and most spices also fall into this category. Cocoa continues to fester on the plant, despite the high amount of anti-oxidants... you have to do a LOT of cooking to make it non-festery.
Maybe in all the controversy about the health effects of human-made pesticides have we missed discovering that one of them has desirable psychoactive effects in humans? (Going to start drinking from a mug that say "Don't talk to me before I've had my glyphosate!" and see what sort of conversations that gets me.)
What about chocolate (i.e., theobromine) as an alternative?
I expect this works for some people, but I doubt it has much of an advantage for the average person.
I think it has a longer half life than caffeine for most people, though possibly it has an advantage in what it metabolizes to.
Another issue is that theobromine stimulation is relatively more in the peripheral nervous system (as opposed to central) than caffeine, so if you take enough to get the same mental stimulation, you may be more jittery than with the caffeine
Fascinating post but misleading owing to huge individual differences in how rapidly caffeine is metabolized (owing in part to CYP1A2 variants) and in the intensity of the reaction to caffeine (owing in part to ADORA2A variants), both of which turn out to have independent effects on sleep. A bunch of health and lifestyle variables also contribute to the variability.
Most of the substance of this post consists of (or assumes) generalizations about the average person and is therefore misleading owing to the extent of variability around the average. Caffeine doesn't have "a" half-life, for example. That value differs greatly across individuals. Some people, like my stepfather, can have a cup of caffeinated coffee at 10 p.m. and sleep just fine.
There's also evidence that the greater the amount of time between caffeine intake and bedtime, the worse people are at gauging the effects of caffeine on objectively-measured sleep quality. This suggests a need for nuance around anecdotal evidence,
(I'm posting on the topic this evening, but if you like Scott's post I don't recommend mine, because I'll be focusing on the narrower and less inherently intriguing question of how people might manage caffeine intake so that they don't undermine their sleep. Also, my post won't be as smart.)
I wonder if roon knows about Roon.
That was my first thought as well.
I took a DNA test and read the results through Promethease, and one of the things it flagged was a gene variant for high caffeine metabolism, which fits considering I can drink coffee any time of day without it affecting my sleep. I thought this was something special until I read that 45% of the population has that variant. And even more interesting, things like smoking or wood stoves can affect that same enzyme which metabolizes caffeine.
I tip my hat to pharmacologists for being able to account for such variations in genetics and lifestyle when studying chemicals.
I think most of those are bunk; some are completely false, others are real genes but explain 1% of the variance. People who know more can correct me if I'm wrong.
Way more than 1% for objective variables like metabolic speed (CYP1A2) as well as "subjective" ones like self-reported anxiety (ADORA2A). (I don't claim to know more but only to have looked, because I'm writing about this.)
I think the issue here is that promethease may not have all of the pharmacogenteically relevant polymorphisms in its database. It may not be specific enough to be super helpful.
Also With CYP1a2 i the general understanding is also that it is (relatively) more sensitive to environmental factors (e.g smoke) than genetics when compared to most CYP enzymes. So often these results dont help tell us much about who will metabolize a med fast/slow as something like smoking status would.
Totally agree with you that info on SNPs from Promethease or anyone else won't help much beyond what you notice informally about your own caffeine response (although, as I suggested in my post on this topic, one could imagine benefits for certain kinds of people. A bunch of studies show that the greater the amount of time between caffeine intake and bedtime, the less accurate people are about objectively measured quality of sleep. So, the nonsmoker who stops drinking coffee before noon and still hasn't figured out why he's not sleeping well at night might benefit from one of those imperfect genetic tests).
Scott's dismissal of genetic influences on individual differences echoes the one-size-fits-all assumptions underlying his original post. Inter- and intra-individual variability is huge. Simple example: Scott repeated the common assertion that caffeine has a half life of 4 to 6 hours. This is false. Caffeine does not have "a" half life. The range appears to be something like 1.5 to 10 or more hours, depending on the study and individual circumstances.
OP seems to describe caffeine as sometimes making him sleepy. I've been known to sleep on arabica cappachino, consumed 2 hours before bedtime. My father would have a full cup of instant coffee at midnight, and then drop straight off to sleep.
Is this non-neurotypical?
I note that different sources of caffeine also seem to induce different psychoactive effects, in terms of sleepiness, where some of the smaller sources induce larger effects. I had originally concluded these were "non caffeinated" effects, but am wondering if it's merely a buffering problem.
Do people notice a differential effect based on how much buffering they use? I'm thinking specifically in terms of milk, which is funktastic, but the same effect might be induced by other substances, like coconut milk, widely used in vietnamese coffee.
https://www.usatoday.com/story/shopping/kitchen/coffee-tea/2026/05/28/coffee-jitters-everyday-dose-2026/90295234007/
At 76% of Americans are daily coffee drinkers, finding non-addicted people wouldn't be that hard.
Hot Chocolate at night, which contains caffeine, is generally not credited with keeping people awake. Even if given at high doses (1/4 cup of pure cocoa is a high dose, right?)
I think it's generally understood there's a substantial population of caffeine non-responders, presumably for genetic reasons. I guess maybe they technically meet the definition of "non-neurotypical" in that they're less than 50% of the population and caffeine metabolism is brain-adjacent?
It might be because a liver fluke tried to turn your nth great grandmother into a zombie: https://slatestarcodex.com/2019/08/19/maybe-your-zoloft-stopped-working-because-a-liver-fluke-tried-to-turn-your-nth-great-grandmother-into-a-zombie/
My understanding is that caffeine can make a subset of the population both alert when they want to be *and* sleep better, because it's just stimulating all metabolic processes, including the ones that will let you sleep.
I actually briefly had an urge to quip in this comment section that I was missing the plot twist at the end of the post saying that caffeine already lets you sleep whenever you want (assuming you don't have e.g. spiralling thoughts or other caffeine-independent sleep blockers that caffeine would make worse), but reckoned it wasn't going to be funny enough to warrant a top-level comment, and not relatable for most people. :)
But yes: Some people (myself included) can sleep fine on caffeine! I don't think this is a non-neurotypical thing specifically as much as it is just varied biochemistry in general.
The liver failure review has me curious. I've had liver problems for a couple years now (primary sclerosing cholangitis as a result of ulcerative colitis). My bilirubin has been hovering between 2-3 mg/dL since 2024, spiking a few months ago up to a high of 14.3. Only time I had a fever was about a month ago, and that was (near as we can tell) the result of a bile duct infection which had probably been hiding for some time, given how much better I felt after the first dose of Zosyn compared to how I'd been feeling up to that point.
Anyway, all this to say I know all about G.I.P. funny colors, but the dark-colored urine I've had has almost always been accompanied by fatigue and a sort of brain fog (it was also very confusing, because nobody told me about that side effect for a while, so I was slightly over-hydrating to try and bring my urine up to its usual color).
I don't really drink coffee these days, but suffice it to say I won't be taking anything with it even if I do. Although, given I'd had fewer than ten standard drinks' worth of alcohol in my life up to the point of getting diagnosed with PSC, maybe a little liver stimulation would've done me some good.
I'm sure glad I never got into coffee or caffeine.
What effect do these alternative stimulants have on stomach acid production? I have to avoid caffeine altogether for that reason, but would like a safe mild stimulant without that problem.
Unclear.
My impression is that heartburn from coffee is mostly a pharmacologic action of caffeine, but also somewhat due to the coffee itself being acidic. Taking it as a pill (like most of these alternative stimulants are) would help with the latter, but not necessarily the former.
Paraxanthine supposedly has fewer side effects than caffeine, and this is plausible since it should be a subset of caffeine, but I don't know of any studies checking whether the stomach muscle relaxation that causes GERD is specifically improved.
I read this trying to replace “sleep” with “heart”. It would be interesting to see if rutaecarpine could treat ER patients who are presenting with ventricular issues from excessive caffeine.
Plausibly, but it's still going to take a couple of hours to get the caffeine out of one's system, I don't know if that's fast enough for ER purposes.
Hugely appreciate this type of post!
Sadly none of these options seem obviously well-suited to my own use-case of "drink Snapple or Coca-Cola or other such tasty caffeinated beverages in the afternoon-to-evening without their drug-effects lingering long enough to (inconsistently but often enough to be inconvenient) make me wake up in the middle of the upcoming night's sleep". Rutaecarpine sounds too high-risk to be worth the benefit, and the others are caffeine ~emulators rather than caffeine ~mitigators.
Hopefully one day someone will develop super-Rutaecarpine, reducing all caffeine effects to zero instantaneously while causing no substantial negative effects, so I can enjoy tasty caffeinated beverages free from the mild inconvenience of actually being affected by caffeine as a drug. (And free from the weird mildly-off flavor I've found decaf coffee to have relative to regular coffee for whatever reason, the times I've tried it.) Or, you know, at least something sufficiently closer to super-Rutaecarpine than to plain Rutaecarpine as to make it worth the tradeoffs of trying out.
Caffeine free coca cola exists and tastes exactly the same. Unfortunately it's only available in big chain supermarkets and even then it's inconsistent.
perhaps the people who aren't affected by paraxanthine are simply the same people who can drink caffeine in the evening without negative effect anyway.
Rutecarpine doesn't just induce CYP1A2, though. It's a COX-2 inhibitor, like Vioxx which got taken off the market for killing people. And it's also an inhibitor of other cytochrome P450 enzymes, so maybe the stories of liver damage aren't a direct result of rutecarpine consumption, they're a result of doing things like taking Tylenol alongside rutecarpine.
I have no problem consuming caffeine and sleeping right afterward. I assume it's just because I consume over a gallon of tea a day. I have basically spent the last 20 straight years highly caffeinated at all times.
For me the tea isn't really about stimulation. It's about being able to drink an unlimited supply of something all day that isn't water and isn't loaded with sugar.
If your main goal with caffeine is to preserve its use as a stimulant, this doesn't seem like too much of a problem because I would assume the consumption of it means you *don't* want to sleep anytime in the near future.
If you just like coffee/tea, then why not just consume such an inordinate amount of it that you develop a tolerance for it?
I've never seen strong evidence that being caffeinated all the time is a major problem. The only problem it's caused me is I get headaches if I stop - but I like tea, so I have no particular reason to stop.
> For me the tea isn't really about stimulation. It's about being able to drink an unlimited supply of something all day that isn't water and isn't loaded with sugar.
How do you feel about artificial sweeteners? There are a lot of versions of artificially sweetened beverages, including highly concentrated "flavor enhancers" that come in tiny squirt bottles that you can add to water. ("Mio" is the name of one brand.)
How's verdict of artificial sweeteners so far? The more I read the more I got confused. Especially in regards of how artificial sweeteners make you crave for more sweets (calories).
Similar story here, but black coffee. I've drank 5-8 cups a day, into the evening, for the best part of two decades and can fall asleep at any point if I'm tired. Normal blood pressure, bloods etc. Assume I'm just completely tolerant.
I point out that, if you're a big-time coffee drinker, drinking decaf in the evening can work as a placebo-stimulant. Works for me.
Also consider going to half-caff, now much more popular.
I used to have energy problems. I appear to have solved them by either one of, or all of these (I don't know as I started them all about the same time):
- weightlifting 2x / week
- cordyceps powder in my morning coffee (I have a mix actually: cordyceps, creatine, L-theamine, turkey tail, cocoa, reishi).
- methylene blue
One more clinically important CYP1A2 interaction: tobacco smoking induces the enzyme, so quitting smoking can suddenly make your usual caffeine dose much longer-lasting. In one study of heavy smokers, caffeine clearance fell 36% within a week after quitting (Faber & Fuhr, 2004).
This is an effect of smoke, not nicotine, which doesn't meaningfully induce CYP1A2. So it can happen even if you switch completely to any NRT.
What I really want is to be able to use caffeine without exacerbating acid reflux.
Huh I didn't read that, but when I was at 'beer school' for a DWI I learned that the step response (how long does a single dose remain in your system) Is like 6 hours for alcohol and 2-3 hrs for THC and nicotine (When consumed through the lungs, not eatables) So I can have a bong at 6 and be asleep by 10. (Or a cig, but I gave up nicotine.)
Alternatively, there are Alpha-2 Adenosine Receptor Inverse Agonists like Sipagladenant that are sold on gray market sites. If you're doing a comparison of these drugs, its worth looking into the non-supplement (and thus more rigorous) side of the literature.
Pre-workouts are among the top selling supplements. Annual sales in the US total about $22 billion according to some estimates. And most are pretty much just mega doses of caffeine.
Semi related but this conversation makes me wonder if there could be a use case for a 1a2 inducing agent for mgmt of clozapine toxicity.
I bought some rutaecarpine and paraxanthine, is there a place I should post my results after I try them for awhile?
Useful and interesting, thank you!
What I *really* want though is a drug that removes caffeine resistance (or at least prevents it from building up.)
As Michael Pollan notes a few times, a naive caffeine dose is the best drug in the world, but resistance builds up very quickly. An ability to maintain that initial high would be incredible.
Capitalism needs to come up with the idea of selling <200mg doses of paraxanthine. That's the smallest dose I can buy (from the UK), but I usually titrate my preferred caffeine dose from 50mg pills.
It would also be great if they stopped packaging it with stuff like '400mg PEAK ATM (R)', which I don't want.
I expected the (in)famous Gwern exploration of nicotine for 5-to-10 energy to get a mention. Would love an ACX discussion of nicotine.
Anecdotally, I tried it and found it to be WAY MORE addicting than I expected, even accounting for being told that many people had read the Gwern post and found it more addicting than they expected.
I never got to daily use, but still to this day, circles with the approximate dimensions of the Zyn containers have a much higher salience in my visual field, and immediately jump out to me the way that tapeta lucida do.
My tolerance is so high that I doubt any of these substances would do anything for me.
There is a simple reason why paraxanthine could be ineffective that seems to be missed here. Coffee contains caffeine which is metabolized to paraxanthine and then to inert compounds by CYP1A2. Coffee also contains methylliberine which inhibits CYP1A2. So isn't it likely that the problem with taking paraxanthine alone that it is metabolized much faster than when it is taken with coffee and methylliberine? Seems like the obvious step would be to try paraxanthine + methylliberine (which is safe because coffee is safe).
“Your ancestors” “since YOU were all sea slugs” are you telling us something Scott??
This is my personal strategy:
I drink coffee with muuuuch milk in the morning - 2 spoons of ground coffee (or 1 normal coffee, 1 decaffeinated) in a 1 liter french press gives me like 3..4 large cups of coffee to which I add about 0.3 liter of milk. 3 or 4 Lattes is similar. Another cup of coffee after lunch. Thats working mostly for my heavy focus work in engineering, planning and fixing complex systems. if I drink more and or later, I get sleep issues.
But I've got sleep issues anyway since I'm a conscious human. So since a couple of years I mostly take 5mg Zolpidem just to have a reliable sleep. Sometimes 10mg if I'm agitated by late social interactions - or I drank too much coffee...
I can't express how refreshing this joke was:
> So which of these options for nighttime caffeine is the best? As always, this is a complicated question, with each alternative having its own set of pros and cons and being right for dif - just kidding, it’s paraxanthine.
Thank you for having the balls to make a judgment call.
Paraxanthine is alrady in my toolkit. I'm a slow caffrine metabolizer, which is supposed to apply to paraxanthine as well, so didn't try it in the afternoon yet. But I find that normal coffee + paraxanthibe 200mg leads to a better day and better sleep than just doubling coffee intake.
Nobody tried yet to only sell the s isomer of modafinil? Esmodafinil. That would be perfect...
When I need to do something late in the evening, I just eat a chocolate, sugar keeps me awake for one or two hours and after that I sleep like a baby
I've been reflecting recently that society has come to fetishize productivity at the expense of our health. When we feel tired, it means the body wants rest. So why not rest? Instead, our mental response to fatigue is now at odds with our physical need. Instead of feeling tired and thinking "I'll take it easy," the response seems to be "I don't like this. I need to be 'productive' and get 'things' done"- so then we do the exact opposite of what's required and take drugs to mask how the body actually feels and push through it. Why do we feel compelled to do this?
Intereating post, what am I missing with the approach "just take less amount of caffeine"?
Really enjoyable piece, and laying the options out by mechanism instead of just by product is what makes it click.
What falls out of your own framing is that the three approaches are all the same CYP1A2 lever pulled at different points: rutaecarpine induces the enzyme, methylliberine inhibits it, and paraxanthine skips the enzyme by being the product it makes. Which also flags the quiet problem with rutaecarpine, since revving CYP1A2 to clear caffeine faster would speed clearance of the paraxanthine doing the useful work too. Paraxanthine sidestepping the whole step looks like the cleaner bet for exactly that reason.
Loved this essay, I recently wrote about what happens when our urge to optimize and exert control over our lives runs up against sleep. https://substack.com/@newyorkjournalofphilosophy/note/p-213017393?r=4qfv2t&utm_medium=ios&utm_source=notes-share-action