Jun 17, 2009
Related to: Akrasia, Hyperbolic Discounting, and Picoeconomics, Fix It And Tell Us What You Did
A while back, ciphergoth posted an article on "picoeconomics", the theory that akrasia could be partially modeled as bargaining between present and future selves. I think the model is incomplete, because it doesn't explain how the analogy is instantiated in the real world, and I'd like to investigate that further sometime1 - but it's a good first-order approximation.
For those of you too lazy to read the article (come on! It has pictures of naked people! Well, one naked person. Suspended from a graph of a hyperbolic curve) Ainslie argues that "intertemporal bargaining" is one way to overcome preference reversal. For example, an alcoholic has two conflicting preferences: right now, he would rather drink than not drink, but next year he would rather be the sort of person who never drinks than remain an alcoholic. But because his brain uses hyperbolic discounting, a process that pays more attention to his current utility than his future utility, he's going to hit the whiskey.
This sticks him in a sorites paradox. Honestly, it's not going to make much of a difference if he has one more drink, so why not hit the whiskey? Ainslie's answer is that he should set a hard-and-fast rule: "I will never drink alcohol". Following this rule will cure his alcoholism and help him achieve his dreams. He now has a very high preference for following the rule; a preference hopefully stronger than his current preference for whiskey.
Ainslie's other point is that this rule needs to really be hard-and-fast. If his rule is "I will drink less whiskey", then that leaves it open for him to say "Well, I'll drink some whiskey now, and none later; that counts as 'less'", and then the whole problem comes back just as bad as before. Likewise, if he says "It's my birthday, I'll let myself break the rule just this once," then soon he's likely to be saying "It's the Sunday before Cinco de Mayo, this calls for a celebration!" Ainslie has some much more formal and convincing ways of framing this, which is why you should read the article instead of just trusting this summary.
The stuff by Ainslie I read (I didn't spring for any of his dead-tree books) didn't offer any specific pointers for increasing your willpower2, but it's pretty easy to read between the lines and figure out what applied picoeconomics ought to look like. In the interest of testing a scientific theory, not to mention the ongoing effort to take control of my own life, I've been testing picoeconomic techniques for the last two months.
The essence of picoeconomics is formally binding yourself to a rule with as few loopholes as possible. So the technique I decided to test3 was to write out an oath detailing exactly what I wanted to do, list in nauseating detail all of the conditions under which I could or could not be released from this oath, and then bind myself to it, with the knowledge that if I succeeded I would have a great method of self-improvement and if I failed I would be dooming myself to a life of laziness forever (Ainslie's theories suggest that exaggeration is good in this case).
I chose a few areas of my life that I wanted to improve, of which the only one I want to mention in public is my poor study habits. I decided that I wanted to increase my current study load from practically never looking at a book after school got out, up to two hours a day.
I wrote down - yes, literally wrote down - an oath in which I swore to study for two hours a day. I detailed exactly the conditions that would count as "studying" - no watching TV with an open book placed in my lap, for example.
I also included several release valves. The theory behind this was that if I simply broke the oath outright, the oath would no longer be credible and would lose its power (again, see Ainslie's article), and there would be some point where I would be absolutely compelled to break the oath (for example, if a member of my family is in the emergency room, I refuse to read a book for an hour and a half before going to check up on them). I gave myself a whole bunch of cases in which I would be allowed to not study, guilt-free, and allowed myself five days a month when I could just take off studying for no reason (too tired, maybe). I also limited the original oath to a month, so that if it didn't work I could adjust it without completely destroying the effectiveness of the oath forever. Finally, I swore the oath in a ceremonial fashion, calling upon various fictional deities for whom I have great respect.
One month later, I find that I kept to the terms of the oath exactly, which is no small achievement for me since my previous resolutions to study more have ended in apathy and failure. On an introspection level, the need to study each day felt exactly like the need to complete a project with a deadline, or to show up for work when the boss was expecting you. My brain clearly has different procedures for dealing with vague responsibilities it can weasel out of, and serious responsibilities it can't, and the oath served to stick studying on the "serious" side of the line.
I am suitably cautious about other-optimizing and the typical mind fallacy, so I don't promise the same method will work for you. But I'd be interested to see if it did4. I'd be especially interested if everyone who tried it would post, right now, what they're trying so that in a month or so we can come back and see how many people kept their oath without having too much response bias.
1: I'm split on the value of picoeconomic theory. A lot of it seems either common-sense if taken as a vague model or metaphor, or obviously false if taken literally. But sometimes it's very good to have a formal model for common sense, and I'm optimistic about someone developing a more literal version of it that explains what's actually going on inside someone's head.
2: Ciphergoth, as far as you know does Ainslie ever start making practical suggestions based on his theory anywhere, or does he leave it entirely as an exercise for the reader?
3: I don't read a lot of stuff on productivity, so I might be reinventing the wheel here.
4: For people trying this, a few suggestions and caveats from my experience: