(None of the following should be particularly surprising. I just want to provide additional personal confirmation that well-established cognitive techniques work as advertised.)

I have fairly strong social anxiety and get overstimulated by loud noises/bright lights. I've previously conditioned myself out of most ordinary and small-group anxiety through a)unavoidable practice with socializing and b)getting a black belt. However, until recently, I continued to have problems with a)a strong stress response to crowds and b)inability to hit on women (at least outside of OKCupid, where the invitation to do so is implicit).

Over the last month, I've successfully applied a couple of standard techniques to deal with this. First, about once or twice a week I would go to a crowded, noisy bar alone and just nurse a drink and people watch for an hour. Exposure therapy worked exactly as expected -- my pulse no longer elevates, I don't perspire excessively, etc. in the crowded/noisy environment.

However, I didn't actually talk to anyone, and even though I can generally be about 95% confident that I will be at least the second-most muscular person in any bar I walk into, I have found that the "strong silent type" approach does not work at all. So it was time for step two. Before going to the bar, I made a commitment on Facebook to make a small extra charitable donation if I did not strike up a conversation with an attractive woman within 60 minutes of arrival (with the intention of increasing the donation amount each time in case of failure. Note that I did not choose a donation to a cause I disagree with, because that would have been adding a source of additional stress in an already stressful situation. The actual motivator here was more "not fail publicly in front of my FB friends," since I'll end up donating to AMF or GiveDirectly eventually regardless.)

And it turns out that pre-commitment works as intended, too. I made my first attempt about 20 minutes in. I did strike up a brief but extremely awkward conversation for a minute or two. I consider this a useful outcome, because it reinforces {failed awkwardly --> no serious consequences} on a subconscious level. I tried again another 20 minutes later (with arguably the most attractive woman in the bar at the time), had a pleasant conversation for 5 or 10 minutes, and got her name.

I don't regard any of this as a particularly heroic accomplishment. I just want to reinforce that, as they say, useful technique is useful.

I made a plan to help me take the stress out of responding to a desire for more desirable mental states. I'd like to post it in the Discussion but I do not have enough karma.

I've been meaning to cut down on snacks, and it's been harder since one coworker set up an informal mini-shop at the office where I can indulge in cookies just by walking to the next cubicle. But two weeks ago she made an inoffensive comment that was only slightly annoying, and I took that chance to use that tiny amount of discomfort as an excuse to not buy any more snacks from her. As a result, it's been two full weeks without buying one single candy, and I'm starting to not miss them.

• Established a useful new habit

Taking zinc on the first signs of a cold. Seems to have avoided three cases but whether it shortens a cold remains to be seen.

• Obtained new evidence that made you change your mind about some belief

• Decided to behave in a different way in some set of situations

I read "Getting to Yes" and used that to improve negotiations (one with a mechanic). The change mostly consists in asking questions instead of making my own position transparent (transparency is good, but position is not).

• Optimized some part of a common routine or cached behavior

I made a habit to consciously reflect on the future merit of web-pages I'm browsing instead of just following my intuition (and impulse). It led to skipping quite a few low to medium value pages.

• Learned something new about your beliefs, behavior, or life that surprised you

My model of my own learning was that I adapt quickly to new information - but often overshoot the target. I believed that I this was a sign of quickness of mind but when thinking about the differences in reasoning and how fast we adapt to new information I realized that my overadapting really might mean that I take facts too literally. That I don't sufficiently question the underlying assumptions - and thus overshoot. I'm still not clear how to deal with that. The good thing is that after overshooting I return equally fast - if given further input/feedback.