Channel factors

Or, “how not to make a fundamental attribution error on yourself;” or, “how to do that thing that you keep being frustrated at yourself for not doing;” or, “finding and solving trivial but leveraged inconveniences.”

Note: cross-posted from my blog, so some of this may be rather elementary to LessWrong readers or CFAR workshop attendees. If that’s you, feel free to skip or skim to the end, where I try to crowdsource a list of interesting channel factors.


One of the key insights of social psychology is that our reactions to events are hugely dependent on the fine details of the situation in question, and often pretty much independent of personality. For instance, suppose you have a bunch of people to playing the Prisoner’s Dilemma with each other, and you want to figure out who will defect. Most people’s theory of mind here says something like, “people defect because they’re self-interested or grumpy or vindictive.” So you might ask a player’s friends how cooperative they were, and use that to guess who will cooperate.

Unfortunately, this approach is totally useless. People are equally likely to cooperate whether or not their friends think they’re cooperative. In the Prisoners’ Dilemma, your friends’ assessment of your personality does not correlate at all with your behavior.

Fortunately, there’s something else which is a pretty good predictor! In particular, it matters a lot whether the instructions of the game called it the “Community Game” or the “Wall Street Game.”1

A graph of Liberman's results.

Yep, a single phrase of the instructions, repeated twice, causes cooperation rates to double. If you ever like to think of yourself as some kind of agent whose decisions are controlled by a rational ego (instead of some random words you heard once upon a time), you might find that a bit worrying.

On the other hand, if you like to think of yourself as the kind of person who prefers to have true beliefs, you might be excited because your beliefs just got truer! You probably sometimes fail to impose egoistic control on your own decisions, and if you understand your brain purely as a consciously-controlled ego then you will be really confused when this happens, and spit out solutions like “try harder next time” or “have more willpower” or “don’t be dumb” or “don’t get distracted” or something.

These are fake solutions. They would totally solve your problem, except that they’re not really actions you can take.

So you should change your model. There are examples everywhere of seemingly trivial changes to circumstances that disproportionately change your behavior. Another example in the literature is that students advised to get tetanus inoculations were far more likely to do so if they were given a map to the university health center and times of operation, which the majority of the students already knew.2 (Of course, intent to get an inoculation had almost no predictive power unless students were given a map.)

The psychology term for such things is a “channel factor,” and it’s probably the most useful psychology concept I’ve learned this semester. Since acquiring it, I’ve noticed it cropping up a lot. For instance:

  • For flossing, a channel factor for me is availability of those little plastic flossy things. Obviously flossing is good enough that I should do it even without handy newfangled devices to help. But while I know this on some level, flossing is time-consuming and unpleasant enough without the plastic aids that I can never really be bothered.

  • For listening to music and calling friends/​family/​people who put me on hold for a while: wireless headphones. I didn’t realize how uncomfortable it was to walk around with a phone jammed up to my ear until I bought a pair of these and suddenly became much more excited about phone calls.

  • For doing laundry routinely, detergent packets. I used to have the problem of leaving my detergent in my dorm’s public laundry room, where it would get removed by well-meaning people and disappear before I realized that it was gone. Then I wouldn’t have any detergent, and I would feel bad about doing laundry and be too lazy to buy new detergent.

The common thread is that the inconvenience is so trivial that I didn’t even notice it until I was specifically on the lookout for channel factors. My usual model of myself is too sane to do things like not floss because I didn’t have the thing that made my fingers slightly more comfortable while I do it. Whoops.

Anyway, this results in a heuristic of, when I am frustrated by the fact that I ‘end up’ not doing something I ‘want’ to do, looking for channel factors. So far, it’s been pretty successful at uncovering small modifications with large effects:

  • closing browser tabs as soon as I’m done with them

  • avoiding checking email like the plague

  • using mightytext to send texts from my computer3

Have you discovered any channel factors of your own? What am I missing?


  1. Lieberman, Varda, et al. “The Name of the Game: Predictive Power of Reputations versus Situational Labels in Determining Prisoner’s Dilemma Game Moves.” Personality and Social Psychology Bulletin, September 2004 vol. 30 no. 9: 1175-1185. doi:10.1177/​0146167204264004.

  2. Leventhal, Howard, et al. “Effects of fear and specificity of recommendation upon attitudes and behavior.” Journal of Personality and Social Psychology, Vol 2(1), Jul 1965: 20-29. doi:10.1037/​h0022089.

  3. Apparently staying in touch with classmates is just not worth the extra 10 seconds per text on my phone keyboard.