Notes

Sweet.  Now that I’m public about being manic depressive, I get to write about fun analogies with various aspects of mathematics and computing.  Today’s analogy: depression and mania are like least and greatest fixpoints in lattices.

To the depressive mind, the world is as bad as it could possibly be given some of the available evidence.  (I say “some” because a depressed person tends to ignore a good deal of positive evidence).  Negative facts are probably true unless obviously false.  Random ideas are almost certainly worthless, since few random ideas are backed by hard evidence.

Mania is the opposite: every hunch not obviously wrong is probably correct, every idea potentially brilliant.  Logic is no barrier to this excessive optimism; when manic I was constantly reassuring myself that I didn’t “know” various crazy ideas for certain, that I was only “exploring” the possibilities.

In other words, “manic depression” is just “guess and check”, but with the guessing and the checking partitioned too far apart in time.  Sane thought requires both forces working in concert; enough fluidity to keep new thoughts and associations flowing, and enough doubt and reiteration to prune away the garbage.

To make the analogy specific: the lattice is a space of sets ideas we believe in, and we’re taking the fixpoint of the “basic thought” operator which takes a applies basic reasoning to support or refute existing ideas and propose new ones.  If the lattice is the set of positive statements, mania will pick out the greatest fixpoint, the everything we could possibly believe in that isn’t obviously ridiculous, and depression will choose the opposite.  The situation is flipped if we consider the dual lattice of negative statements.

This analogy is why the underlying fundamentals of manic depression can be simple, and hopefully simply curable, even while its symptoms and consequences are not.  We take a simple heuristic tweak, run it through all the vast complexity of the human brain, and repeat until something snaps.

Incidentally, once we figure out how to correct the tweak, deciding what to do with this ability (or how to control it) will become important.  There’s a TED talk by Joshua Walters on being manic depressive where he notes, “There’s no amount of drugs you can take that will get you as high as if you think you’re Jesus Christ.”  Very true (I know the feeling if not the particular scenario), and perhaps a bit dangerous if we’re eventually able to turn mania and depression on and off at will.

I believe life is more joyful and free when one is open and honest about as many things as possible.  Therefore, I’ve decided to be open about having manic depressive illness, in the belief that being public about it will lift a bit of mental weight from my shoulders, and act as a personal vote in favor of openness and lack of stigma about mental illness.

So far I’ve had two manic periods.  I recovered from the first on my own with the help of friends and family, with the significant downside of not being diagnosed.  The second one happened five years later, and was much worse: I was hospitalized twice (yes, that means they made a mistake letting me go the first time), and lost both a job and a girlfriend.  I’m taking appropriate medication, now, so with luck there won’t be a third by the time the real cures come along (see below).

The worst part of mania, for me, is that it adds a small stain to normal happiness and interests.  One is always left wondering: is this happiness related to mental illness? This phenomenon is especially annoying because all of my manic delusions are on topics that I naturally find fascinating, such as programming languages, simulation, complexity theory, physics, etc.  Rationally there’s not much to be worried about–I’m fairly confident I can notice mania now that I know to look for it–but it remains a mental drag.

I won’t go into details about my manic delusions, but here’s a taste: At one point I wrote P < BPP = PSPACE < EXPTIME on someone’s whiteboard, convinced that I had plausible physical arguments for each part (involving IP and the holographic principle).  Needless to say, this is insane: the sane version of me believes P = BPP < NP < PSPACE < EXPTIME.

A few good things: optimism and nonviolence are baked deeply enough into my personality that mania doesn’t seem to touch them, so I haven’t been a danger to myself or others.  Not everyone is as lucky, so this is something to be thankful for.  I also have no issues taking medication, since (1) I believe completely in my diagnosis, and (2) I don’t seem to have any major side effects (other than possible weight gain).

In the long run, I’m fairly confident that a couple decades from now we’ll have a much better understanding of both the causes of mood disorders and of how exactly the various drugs act.  Part of this is general optimism about medical technological advance, and part of it is an intuitive sense that mania and depression are simple enough that they can be mostly understand in terms of a small constant number of variables (activity levels of a few parts of the brain, key neurotransmitters, etc.).  I think the fact that something as simple as lithium works so well is evidence of this simplicity.  If this turns out to be true, we should be able to make a gadget that monitors the appropriate variables and detects mania and depression essentially perfectly (no more subjective mood charts).

Actually curing the source of manic depression might take a lot longer, since it seems to be intricately mixed with personality, creativity, etc., but an automatic mood detection gadget is probably enough to render the chance of relapses negligible, in which case it’d be an effective cure.  Something to look forward to.  Since manic depression and mood disorders in general are largely genetic, this will hopefully happen within 15 or 20 years, rendering any possible children of mine safe.

That’s enough for now.  There’s a fairly good chance that only a small number of my good friends will read this, but it’s still nice to be effectively public.  Especially since this website is still the top Google hit for “Geoffrey Irving”.