Humans are not automatically strategic

Re­ply to: A “Failure to Eval­u­ate Re­turn-on-Time” Fallacy

Lion­hearted writes:

[A] large ma­jor­ity of oth­er­wise smart peo­ple spend time do­ing semi-pro­duc­tive things, when there are mas­sively pro­duc­tive op­por­tu­ni­ties un­tapped.

A some­what silly ex­am­ple: Let’s say some­one as­pires to be a co­me­dian, the best co­me­dian ever, and to make a liv­ing do­ing com­edy. He wants noth­ing else, it is his pur­pose. And he de­cides that in or­der to be­come a bet­ter co­me­dian, he will watch re-runs of the old tele­vi­sion car­toon ‘Garfield and Friends’ that was on TV from 1988 to 1995....

I’m cu­ri­ous as to why.

Why will a ran­domly cho­sen eight-year-old fail a calcu­lus test? Be­cause most pos­si­ble an­swers are wrong, and there is no force to guide him to the cor­rect an­swers. (There is no need to pos­tu­late a “fear of suc­cess”; most ways writ­ing or not writ­ing on a calcu­lus test con­sti­tute failure, and so peo­ple, and rocks, fail calcu­lus tests by de­fault.)

Why do most of us, most of the time, choose to “pur­sue our goals” through routes that are far less effec­tive than the routes we could find if we tried?[1] My guess is that here, as with the calcu­lus test, the main prob­lem is that most courses of ac­tion are ex­tremely in­effec­tive, and that there has been no strong evolu­tion­ary or cul­tural force suffi­cient to fo­cus us on the very nar­row be­hav­ior pat­terns that would ac­tu­ally be effec­tive.

To be more spe­cific: there are clearly at least some limited senses in which we have goals. We: (1) tell our­selves and oth­ers sto­ries of how we’re aiming for var­i­ous “goals”; (2) search out modes of ac­tivity that are con­sis­tent with the role, and goal-seek­ing, that we see our­selves as do­ing (“learn­ing math”; “be­com­ing a co­me­dian”; “be­ing a good par­ent”); and some­times even (3) feel glad or dis­ap­pointed when we do/​don’t achieve our “goals”.

But there are clearly also heuris­tics that would be use­ful to goal-achieve­ment (or that would be part of what it means to “have goals” at all) that we do not au­to­mat­i­cally carry out. We do not au­to­mat­i­cally:

  • (a) Ask our­selves what we’re try­ing to achieve;

  • (b) Ask our­selves how we could tell if we achieved it (“what does it look like to be a good co­me­dian?”) and how we can track progress;

  • (c) Find our­selves strongly, in­trin­si­cally cu­ri­ous about in­for­ma­tion that would help us achieve our goal;

  • (d) Gather that in­for­ma­tion (e.g., by ask­ing as how folks com­monly achieve our goal, or similar goals, or by tal­ly­ing which strate­gies have and haven’t worked for us in the past);

  • (e) Sys­tem­at­i­cally test many differ­ent con­jec­tures for how to achieve the goals, in­clud­ing meth­ods that aren’t ha­bit­ual for us, while track­ing which ones do and don’t work;

  • (f) Fo­cus most of the en­ergy that *isn’t* go­ing into sys­tem­atic ex­plo­ra­tion, on the meth­ods that work best;

  • (g) Make sure that our “goal” is re­ally our goal, that we co­her­ently want it and are not con­strained by fears or by un­cer­tainty as to whether it is worth the effort, and that we have thought through any ques­tions and de­ci­sions in ad­vance so they won’t con­tinu­ally sap our en­er­gies;

  • (h) Use en­vi­ron­men­tal cues and so­cial con­texts to bolster our mo­ti­va­tion, so we can keep work­ing effec­tively in the face of in­ter­mit­tent frus­tra­tions, or temp­ta­tions based in hy­per­bolic dis­count­ing;

.… or carry out any num­ber of other use­ful tech­niques. In­stead, we mostly just do things. We act from habit; we act from im­pulse or con­ve­nience when primed by the ac­tivi­ties in front of us; we re­mem­ber our goal and choose an ac­tion that feels as­so­ci­ated with our goal. We do any num­ber of things. But we do not sys­tem­at­i­cally choose the nar­row sets of ac­tions that would effec­tively op­ti­mize for our claimed goals, or for any other goals.

Why? Most ba­si­cally, be­cause hu­mans are only just on the cusp of gen­eral in­tel­li­gence. Per­haps 5% of the pop­u­la­tion has enough ab­stract rea­son­ing skill to ver­bally un­der­stand that the above heuris­tics would be use­ful once these heuris­tics are pointed out. That is not at all the same as the abil­ity to au­to­mat­i­cally im­ple­ment these heuris­tics. Our ver­bal, con­ver­sa­tional sys­tems are much bet­ter at ab­stract rea­son­ing than are the mo­ti­va­tional sys­tems that pull our be­hav­ior. I have enough ab­stract rea­son­ing abil­ity to un­der­stand that I’m safe on the glass floor of a tall build­ing, or that ice cream is not healthy, or that ex­er­cise fur­thers my goals… but this doesn’t lead to an au­to­matic up­dat­ing of the re­ward gra­di­ents that, ab­sent rare and costly con­scious over­rides, pull my be­hav­ior. I can train my au­to­matic sys­tems, for ex­am­ple by vi­su­al­iz­ing ice cream as dis­gust­ing and artery-clog­ging and yucky, or by walk­ing across the glass floor of­ten enough to per­suade my brain that I can’t fall through the floor… but sys­tem­at­i­cally train­ing one’s mo­ti­va­tional sys­tems in this way is also not au­to­matic for us. And so it seems far from sur­pris­ing that most of us have not trained our­selves in this way, and that most of our “goal-seek­ing” ac­tions are far less effec­tive than they could be.

Still, I’m keen to train. I know peo­ple who are far more strate­gic than I am, and there seem to be clear av­enues for be­com­ing far more strate­gic than they are. It also seems that hav­ing goals, in a much more per­va­sive sense than (1)-(3), is part of what “ra­tio­nal” should mean, will help us achieve what we care about, and hasn’t been taught in much de­tail on LW.

So, to sec­ond Lion­hearted’s ques­tions: does this anal­y­sis seem right? Have some of you trained your­selves to be sub­stan­tially more strate­gic, or goal-achiev­ing, than you started out? How did you do it? Do you agree with (a)-(h) above? Do you have some good heuris­tics to add? Do you have some good ideas for how to train your­self in such heuris­tics?

[1] For ex­am­ple, why do many peo­ple go through long train­ing pro­grams “to make money” with­out spend­ing a few hours do­ing salary com­par­i­sons ahead of time? Why do many who type for hours a day re­main two-finger typ­ists, with­out both­er­ing with a typ­ing tu­tor pro­gram? Why do peo­ple spend their Satur­days “en­joy­ing them­selves” with­out both­er­ing to track which of their ha­bit­ual leisure ac­tivi­ties are *ac­tu­ally* en­joy­able? Why do even un­usu­ally nu­mer­ate peo­ple fear ill­ness, car ac­ci­dents, and bo­gey­men, and take safety mea­sures, but not bother to look up statis­tics on the rel­a­tive risks? Why do most of us set­tle into a sin­gle, stereo­typed mode of study­ing, writ­ing, so­cial in­ter­ac­tion, or the like, with­out try­ing al­ter­na­tives to see if they work bet­ter—even when such ex­per­i­ments as we have tried have some­times given great boosts?