What ways? Say and pray?
No, that doesn't sound effective at all. Instead, try to think of some way of changing your behavior that has a chance in hell of actually working.
For example, suppose you eat a lot of candy (or drink lots of booze, or whatever) and you want to cut down on that for health reasons. Just trying to eat less candy will probably not work -- your willpower can break down a little bit at a time, and before you know it, you've failed to make any serious change.
So, here comes the trick: mentally redefine yourself as someone who hardly ever eats candy (or drinks alcohol, or whatever) and put as much effort as necessary into staying away from it entirely for a few weeks. The initial enthusiasm for a new plan can carry you for the first few weeks, and it gets easier after that because the new habits are set in place and you've learned how to snack on fruits and nuts, and eat meals that won't leave you with low blood sugar later.
I've found this to be a very effective strategy for shaking off bad habits. No prayer involved.
I would like to propose this as a thread for people to write in their New Year’s Resolutions (goals and sub-goals) as instrumental rationalists.
Here are mine.
Goals
I resolve to try to enable all pareto-optimizing trades between my sub-agents can be made between them. For instance, I have agents which would like to maximize my success via improving my cognition and energy levels via eating more healthily via eating less. Other agents would like enjoyment from food. I note that these agents aren't cooperating even though they both benefit from the same changes in behavior, largely because it hasn't been pointed out to them that they are on the same side. If I make a serious effort to eat the hedonic utility maximizing amount, this will probably involve eating less than I default to. After all, food is better when one is hungry. Most of my eating is probably driven by simple non-reflective systems that tell me to eat. These systems are probably promoted by hedonistic systems which are failing to understand the consequences of doing so. In practice, this resolutoin means paying attention to the experience of eating anything that wasn’t chosen for social or nutritional purposes, rarely clearing my plate, and rarely eating more than one would get served in a European restaurant, but above all, it means paying attention (and thus sending this information to many of my sub-agents) to the pleasure of eating when one is actually hungry.
I resolve to find a new home, get fully moved in, get a car, touch base with all interested Bay Area supporters, and get started on the 2010 Summit by the end of February despite this involving many boring activities.
I resolve to stop trying to keep up with a significant part of the blogosphere. My web-browsing will be limited to a) actually seeking specific information, b) checking email no more than 5-times-per-day and c) the keeping up, via google reader, with fewer than ten sites.
Also, in February I will try to use dual-n-back every day and in March I will try to publish a Less-Wrong piece every day.