I'm pretty sure you're doing something wrong now. You're being very vague and not giving any examples so I can't troubleshoot anywhere near precisely, but clearly you're trying to fit a square peg into a round hole and looking for the right sledgehammer for the job. You're not Augustine of Hippo; you may endorse a set of rules as the Sacred Laws Of Rationality That You Are A Really Bad Person If You Don't Follow, but if trying to follow them causes breakdowns, you're just wrong about the rules. Taboo "rational", and ask if each rule is a maintainable habit, possible to use explicitly in extraordinary circumstances, and if you actually want to do that. ("My life is worth exactly as much as some random stranger's" sounds nice, but nobody can actually follow that long-term.)

You don't say what kind of insane you are. You mention lack of social skills and that it's a big problem for you, but that's nearly orthogonal. For the first time in history, people are publishing useful guides to social life. Of course any oddity is going to make it harder for you, but go to groups that share your interests, and you'll find more people whose personality meshes with yours and fewer who go "It's not making eye contract in the exact pattern I want! Burn the witch!". It helps to sincerely like the people you're trying to befriend, but a little dishonest manipulation can go a long way too.

If you have other insanity-related problems, I suggest you ask a psychiatrist for help with the root cause, tackle each problem directly, or start a Less Wrong mental health support group (so far the current procedure is "whine loudly enough to attract Alicorn's compassion", which might be a bit hard on Alicorn). Those bouts of self-destruction might be due to pushing too hard in bad directions, but might have other origins too.

I do not perform primate pack-bonding rituals correctly

Give three concrete examples from your life.

What few clues given above lead me to believe that if you were to fix your ability to perform "primate pack-bonding rituals" correctly, a lot of this would go away. We are social creatures, and it sounds like for all your rationalization, social interaction is a significant driver for you.

Assuming this is the case, the short answer is that those ritual skills are learnable, and there's a ton of material on how to do so out on the web. Cognitive therapy may also help; however, in the end, you're just going to have to get up, get out, experiment with rituals and fail a lot to figure out how they work. You can't just learn it reading from books and the web. You have to actually go out and go through the actions, just like learning a martial art.

That's what I did. I still view people as machinery, as you probably do. But they're at least interesting machinery, and I know the interfaces well enough to do interesting things with them, including get the positive feedback that helps me maintain my outlook.

What if the very tools that I use to make decisions are flawed?

Everybody's tools are flawed. This blog exists because it seems reasonable to believe that the tools are good enough to do bootstrapping so that the flaws are smaller.

I'm going to write about some things which have helped me which may or may not be of use to you. I've worked on compassion-- it's tempting for me to believe that I'm the only defective person in the universe, and then try to fix myself. Coming in with that attitude is part of the problem, maybe even most of it. I'm better off to the extent that I can view my current state with a neutral or (much harder) benevolent attitude and work from there.

Also, it helped to realize that my current state has seven billion years of universe behind it. I can change for the better, but whatever is wrong isn't some intrinsic personal defect, and it isn't all my mother's fault either.

Alternate explanation for "insanity": If your IQ is high enough, you're likely to have problems fitting in with others. Normally I wouldn't suggest high IQ as a reason for not fitting in since an IQ high enough to cause that problem occurs in less than 1% of the population. However, here you are posting on LessWrong, a place that is known for it's intelligent members. (See Yvain's surveys to discover that most claim a high enough IQ for the average to be in the 140's). Not only that, but if you were using Bayesian techniques as a child and experimenting with making AIs as a teen, I'd say you're very likely to be smarter than the average bear.

If you want to look into this further:

Try researching a concept called "socially optimal IQ range".

Check out this article by the Prometheus Society: The Outsiders

Consider reading this book: Misdiagnosis and Dual Diagnosis of Gifted Children and Adults

Research the term: Existential depression (common to gifted adults, and your inability to hack utility table complaint is reminiscent of this).

If you or someone reading this needs a concierge into the subject of gifted adults, I can be one. If the prospect of being flamed for claiming giftedness / looking into giftedness is a concern, use PM.

From the description you've given, which doesn't give much to go on, it sounds like you have some, but not all, of the same problems I do (in my case stress-related anxiety and clinical depression, compounding mild comorbid Asperger's and dyspraxia).

What I'd recommend in this case is that you access cognitive behavioural therapy. It's the only psychiatric intervention that has been actually shown to have any long-term effects (short of dumping people full of antipsychotic drugs, which have far too many bad side-effects for me to ever recommend them). It's also very close to applied rationality, so it might fit your worldview and be more acceptable to you than other treatments would. From my own personal experience with it, it's not a panacea, but it is useful.

If you're in the US and poor, and thus can't access medical help, I would suggest learning some of the techniques from Zen Buddhism. I don't have much experience of this myself, but several friends who I trust have told me that the meditation techniques in Zen are very similar to a less-formalised version of CBT, and in some cases have helped them more. The podcasts at zencast.org have been very helpful to several people I know with problems like that.

Also, I am NOT a doctor and this is NOT meant to be medical advice that you should take without consulting one, and I am NOT accepting liability for anything you do, but I have seen suggestions that taking large doses of niacin -- large enough to cause flushing -- can help get rid of mild paranoia, anxiety and depression. My own experiences tend to bear this out, but it could well be a placebo effect.

And finally, this is DEFINITELY NOT IN ANY WAY A RECOMMENDATION, but there are several studies that suggest that the prescription-only drug ketamine, which is not licensed for this purpose, can provide long-term relief from depression and can also aid cognitive functioning. If you have a doctor who is willing to prescribe off-label, it may be worth discussing that with her, although it is very unlikely you'd get the prescription as ketamine is widely used as a recreational drug. I have no experience of it myself, unlike the other things I've mentioned here, so can't speak directly for its efficacy.

My name is Brent, and I'm probably insane.

Chorus: Hi, Brent.

what is someone who wishes to be rational supposed to do, when the underlying hardware simply won't cooperate?

Being aware of the biases, yet unable to adapt your reasoning to compensate, seems to be contradictory. When you say "I know I only think X because of bias Y, so my actual belief should be Z", you seem to already have solved the problem in that instance, by just switching them out (in lambda calculus: E[X:=Z]).

The unknown unknowns are in my opinion the crux of the problem: those biases you did not (yet) recognize in specific situations, regardless of how well you trained yourself to reflect upon your own reasoning. Due to the nature of the problem, we wouldn't even be aware of how much progress we made in recognizing biases, and how much is left to be done. (Comparing the variance among reasoning agents would help: Based on Aumann, we can in principle eliminate - or at least notice the existence of - biases that we do not share, but two agents with a shared kind of bias would still converge on the same belief and thus be oblivious to it*).

What do? Do the best with the hand dealt to you, e.g. if it were the case (as a cosmic joke) that Occam's Razor didn't hold true for vetting ToE's after all, too bad. At least we did our very best then.

* I'm not certain this is a formal result, it should be the case for a majority of cases. Comments welcome.

This may be a stupid question: How old are you? From the pattern of your posts I seem to detect a vague hint of an American high school. Please observe: American schools are severely broken. (1) Inability to function in such a place is not necessarily a sign of insanity. Consider whether, perhaps, you may simply be surrounded by hormonal teenagers with nothing better to do than assert idiotic status hierarchies. If you're not already familiar with it, Paul Graham's essay on nerds may be relevant to your interests.

That aside: Are you sure you are really trying to be rational, as opposed to performing some ritual of cognition? It's hard to use oneself as a control group, you certainly cannot do so double-blindly, but what happens if you drop your attempts at rationality and act without thought, or with less thought than you are currently using? For all you know, that may give even worse results; but at any rate it seems to me that your problem is desperately in need of a control, or baseline. When you debug, it's no use saying that the error must be somewhere in module X. Comment out module X and see if the crash still happens.

Footnote 1: I read this on the Internet, so it must be true.