Let There Be Light

Se­quence in­dex: Liv­ing Lu­mi­nous­ly
Pre­vi­ously in se­quence: You Are Likely To Be Eaten By A Grue
Next in se­quence: The ABC’s of Luminosity

You can start from psych stud­ies, per­son­al­ity tests, and feed­back from peo­ple you know when you’re learn­ing about your­self. Then you can throw out the stuff that sounds off, keep what sounds good, and move on.

You may find your un­der­stand­ing of this post sig­nifi­cantly im­proved if you read the first story from Seven Shiny Sto­ries.

Where do you get your pri­ors, when you start mod­el­ing your­self se­ri­ously in­stead of do­ing it by half­hearted in­tu­ition?

Well, one thing’s for sure: not with the cal­iber of in­tro­spec­tion you’re most likely start­ing with. If you’ve spent any time on this site at all, you know peo­ple are rid­dled with bi­ases and mechanisms for self-de­cep­tion that sys­tem­at­i­cally con­found us about who we are. (“I’m splen­did and brilli­ant! The last five hun­dred times I did non-splen­did non-brilli­ant things were out­ra­geous flukes!”) Hu­mans suck at most things, and obey­ing the edict “Know thy­self!” is not a spe­cial case.

The out­side view has got­ten a bit of a bad rap, but I’m go­ing to defend it—as a jump­ing-off point, any­way—when I fill our lu­minos­ity toolbox. There’s a ma­jor body of liter­a­ture de­signed to figure out just what the hell hap­pens in­side our skulls: it’s called psy­chol­ogy, and they have a rather im­pres­sive track record. For in­stance, learn­ing about heuris­tics and bi­ases may let you de­tect them in ac­tion in your­self. I can of­ten tell when I’m about to be sub­ject to the by­stan­der effect (“There is some­one sit­ting in the mid­dle of the road. Should I call 911? I mean, she’s sit­ting up and ev­ery­thing and there are non-alarmed peo­ple look­ing at her—but gosh, I prob­a­bly don’t look alarmed ei­ther...”), have made some progress in re­duc­ing the ex­tent to which I gen­er­al­ize from one ex­am­ple (“How are you not all driven in­sane by the spat­ters of oil all over the stove?!”), and am sus­pi­cious when I think I might be above av­er­age in some way and have no hard data to back it up (“Now I can be con­fi­dent that I am in fact good at this sort of prob­lem: I an­swered all of these ques­tions and most peo­ple can’t, ac­cord­ing to some­one who has no mo­ti­va­tion to lie!”). Now, even if you are a stan­dard psych study sub­ject, of course you aren’t go­ing to al­ign with ev­ery psy­cholog­i­cal find­ing ever. They don’t even al­ign perfectly with each other. But—con­trol­ling for some huge, ob­vi­ous fac­tors, like if you have a men­tal ill­ness—it’s a good place to start.

For nar­row­ing things down be­yond what’s been turned up as typ­i­cal hu­man re­ac­tions to things, you can try per­son­al­ity tests like My­ers-Briggs or Big Five. Th­ese are not fan­tas­ti­cally re­li­able sources. How­ever, some of them have some abil­ity to track with some parts of re­al­ity. Ac­cord­ingly, sat­u­rate with all the test data you can stand. Filter it for what sounds right (“gosh, I guess I do tend to be rather both­ered by things out of place in my en­vi­ron­ment, com­pared to oth­ers”) and dump the rest (“huh? I’m not open to ex­pe­rience at all! I won’t even try es­car­got!”) - these are rough, first-ap­prox­i­ma­tion pri­ors, not pos­te­ri­ors you should ac­tu­ally act on, and you can af­ford a clumsy pro­cess this early in the game. While you’re at it, give some thought to your in­tel­li­gence types, cat­e­go­rize your love lan­guage1 - any­thing that carves up per­son-space and puts you in a bit of it.

Ad­di­tion­ally, if you have hon­est friends or rel­a­tives, you can ask for their help. Note that even hon­est ones will prob­a­bly have a rosy pic­ture of you: they can stand to be around you, so they prob­a­bly aren’t pay­ing ex­cru­ci­at­ingly close at­ten­tion to your flaws, and may ex­ag­ger­ate the im­por­tance of your virtues rel­a­tive to a neu­tral ob­server’s hy­po­thet­i­cal opinion. They also aren’t around you all the time, which will con­strict the cir­cum­stances in which their model is tested and skew it to­wards what­ever in­fluence their own pres­ence has on you. Their out­side per­spec­tive is, how­ever, still valuable.

(Tips on get­ting friends/​fam­ily to provide feed­back: I find mus­ing aloud about my­self in an ob­vi­ously ten­ta­tive man­ner to be fairly use­ful at elic­it­ing some do­main-spe­cific in­put. Some of my friends I can ask point-blank, al­though it helps to ask about spe­cific situ­a­tions (“Do you think I’m just tired?” “Was I over the line back there?”) rather than gen­eral traits that feel more judg­men­tal to dis­cuss (“Am I a jerk?” “Do I use peo­ple?”). When you com­mu­ni­cate in text and keep logs, you can send peo­ple pastes of en­tire con­ver­sa­tions (when this is per­mis­si­ble to your origi­nal in­ter­locu­tor) and ask what your con­sul­tant thinks of that. If you do not re­mem­ber some event, or are will­ing to pre­tend not to re­mem­ber the event, then you can get who­ever was with you at the time to re­count it from their per­spec­tive—this pro­cess will au­to­mat­i­cally paint what you did dur­ing the event in the light of out­side scrutiny.)

If dur­ing your prior-hunt­ing some­thing turns up that seems wrong to you, whether it’s a whole test re­sult or some spe­cific sup­posed fea­ture of peo­ple in a group that seems oth­er­wise gen­er­ally fit­ting, that’s great! Now you can rule some­thing out. Think: what makes the model wrong? When have you done some­thing that falsified it? (“That one time last week” is more promis­ing than “back in eighty-nine I think it might have been Jan­uary”.) What are the small­est things you could change to make it sit right? (“Change the word “rapid” to “metic­u­lous” and that’s me to a tee!”) If it helps, take in the in­for­ma­tion you gather in small chunks. That way you can in­spect them one at a time, in­stead of only holis­ti­cally ac­cept­ing or re­ject­ing what a given test tells you.

If some­thing sounds right to you, that’s also great! Ask: what pre­dic­tions does this idea let you make about your cog­ni­tion and be­hav­ior? (“Should you hap­pen to meet a tall, dark stranger, you will make rapid as­sump­tions about his char­ac­ter based on his body lan­guage.”) How could you test them, and re­fine the model? (Where do the tall, dark strangers hang out?) If you’ve be­haved in ways in­con­sis­tent with this model in the past, what ex­cep­tions to the rule does that im­ply and how can you most con­cisely, Oc­cam-es­que-ly sum­ma­rize them? (“That one tall, dark stranger was wear­ing a very cool t-shirt which oc­cluded pos­ture data.”)

Nota bene: you may be tempted to throw out things be­cause they sound bad (“I can’t be a nar­cis­sist! That wouldn’t be in keep­ing with the story I tell about my­self!”), rather than be­cause they sound wrong, and to keep things be­cause they sound good (“ooh! I’m funny and smart!”), rather than be­cause they sound right. Re­cite the Li­tany of Tarski a few times, if that helps: if you have a trait, you de­sire to be­lieve that you have the trait. If you do not have a trait, you de­sire to be­lieve that you do not have the trait. May you not be­come at­tached to be­liefs you may not want. If you have bad fea­tures, know­ing about them won’t make them worse—and might let you fix, work around, or miti­gate them. If you lack good fea­tures, de­lud­ing your­self about them won’t make them ap­pear—and might cost you op­por­tu­ni­ties to de­velop them for real. If you can’t an­swer the ques­tions “when have you done some­thing that falsified this model?” or “list some ex­am­ples of times when you’ve be­haved in ac­cor­dance with this model”—sec­ond guess. Try again. Think harder. You are not guaran­teed to be right, and be­ing right should be the aim here.

1It looks cheesy, but I’ve found it re­mark­ably use­ful as a first-pass ap­prox­i­ma­tion of how to deal with peo­ple when I’ve got­ten them to an­swer the ques­tion.