[greenpeas]

> hallucinogens sometimes cause minor hallucinations indefinitely

The author definitely did not think those were minor side-effects.

> What's missing, crucially, is the only useful information- how many of each there are. There is nothing else that is relevant!

the evidence that a bad trip can potentially cause permanent damage is useful and relevant.

> If "some guy had a bad time" was a robust argument for "never ever do this thing", it would be a bad idea to do anything ever.

"some guy got permanently fucked up doing something" is a strong argument for me to very carefully consider the potential benefits of "doing something" and whether they are worth it; in the case of LSD, for me personally, the benefits don't outweigh the potential downsides, and anecdata is more than enough for me to make this decision.

[ravi-delia]

> the evidence that a bad trip can potentially cause permanent damage is useful and relevant.

I would argue that this is not true in isolation! I guess it's not technically wrong, in that a new article is not 0 information, but without an attached probability there isn't much to learn. Frankly, some of the symptoms the author experienced are typical of HPPD, and some are not. Correlation isn't causation, and one datapoint absolutely does not show so much as correlation.

You almost certainly partake in activities with a chance to go far more wrong every day! In fact, if the mere possibility of something going wrong is enough to dissuade you, then you shouldn't have needed the article in the first place- there's always a chance things could go horribly wrong!

To be clear, I'm not saying LSD has no risks. The author got very unlucky, but not as unlucky as others who have been documented in the literature. My pitch is not even that those risks are mostly worth it- for many they probably aren't. What I'm saying is that the author's experience is every bit as useless as the people who constantly hype LSD up. The conclusion is "Fuck you, Jobs"- can you say with a straight face that the author wouldn't have concluded the same way even if no one else in the world had ever had a bad experience on LSD?

[greenpeas]

> but without an attached probability there isn't much to learn.

But there is an attached probability implicitly. I can make some common sense assumptions about the number of people that have ever done LSD and get a lower bound on the risk (with a huge confidence interval, but still). Of course a chance of say 1 in ten millions is really small, but if the perceived benefits of doing LSD are also small, then that might be enough for me to make up my mind.

> Frankly, some of the symptoms the author experienced are typical of HPPD, and some are not. Correlation isn't causation, and one datapoint absolutely does not show so much as correlation.

This is a good point, I have to decide for myself whether that person is trustworthy or not, and whether their experience is relevant to my case. But if I decide that I can trust their evidence, then I'm going to factor it into my decision, even if it's only one datapoint. If in the future a controlled study on the risks of doing LSD gets published, I can reevaluate my position then.

> You almost certainly partake in activities with a chance to go far more wrong every day! In fact, if the mere possibility of something going wrong is enough to dissuade you, then you shouldn't have needed the article in the first place- there's always a chance things could go horribly wrong!

Yes, I partake in many activities that have a nonzero chance of harming me temporarily or permanently. But with each of those activities, I've decided that the potential benefits outweigh the risks. And mind you, I did not need to read extensive studies with attached probabilities do decide that crossing the road or swimming in the sea is worth the risks.

This touches on another disagreement I have with your original response. We don't live our lives constantly evaluating the risks based on peer reviewed studies with attached probabilities. We rely on common sense and our lived experiences as well as other people's experience that we learn about. The plural of anecdote is not data, but together with a healthy dose of common sense, it can provide acceptable heuristics for evaluating risks and making decisions.

[ravi-delia]

I think I'm making two assumptions here, and they explain the bulk of our disagreement.

First my intuition here is that in most situations, we should expect any specific example brought to our attention to be actively misleading. Out of thousands of stories about LSD, this one happens to be on the HN frontpage- should we not expect it to be unusual in some way?

Second, I think I assume that for any given activity, there exists at least one story about it that is very bad. So finding such a story shouldn't swing your judgement too much. It isn't new information!

To put it another way, we can't possibly be using the pair (average case gain, example very bad outcome) to judge if things are worth it. When we decide to cross the road, we are drawing on our extensive record of safely crossing roads, as well as our highly accurate model of the road (I would feel much less safe crossing a highway at rush hour than a suburban road at 1am) to judge the situation. If we just drew on benefit (get to the other side of the road) and potential cost (get hit by car), there would be no difference between crossing at an intersection and jaywalking. Clearly there's a probability component which is not drawn from the ether, but instead from our internal model of how the world works. I think that in your argument, this would be the "healthy dose of common sense".

In the case of drug effects, or in medical interventions overall, we don't have that benefit. Here anecdotes and common sense commonly steer us wrong- that is the undeniable conclusion of decades of disappointing research.

My specific disagreement with the way the author framed their point, which prompted my response, was how they felt that people who have had good experiences and talked about them were "naive", because the author themselves had a bad experience. I take the perspective that those people are no more wrong than the author- both are generalizing entirely from their own experiences. But your perspective would seem to argue that those people talking about how LSD is great are correct! After all, many psychonauts have tried LSD many times and come out fine, and have seen others do the same. They are reasoning from anecdotes and a healthy dose of common sense.

It may well be the case that for most peoples' risk budgets, LSD isn't worth it. The author obviously things so. But at no point do they make that argument! The sum total of the post is "Some people anecdotally say they get great benefit from LSD", "I had extremely negative effects from LSD (as do some other people)", followed by "therefore LSD is definitely not worth it for anyone". Huh? It's totally understandable why someone who went through what the author did would feel this way. It would be entirely unreasonable to expect someone bitten by a shark to couch their very real and very bad lived experience with the statistics on shark attacks. Hell, there's a legitimate point to be made about how the spiritualism and hype around psychedelics mean that positive anecdotes are treated as data while negative anecdotes are brushed under the rug. But the author, however understandably, did not make that point.