Why is there a "clogged drainpipe" effect in idea generation?

post by jacobjacob · 2020-11-20T19:08:08.461Z · LW · GW · 5 comments

This is a question post.

Contents

  What on earth is going on here? Why would minds work like this? What does it tell us about the structure of human cognition? 
  ---
None
  Answers
    28 ike
    1 Slider
None
5 comments

I've noticed something curious about creativity. (Like when doing the Babble challenges [LW · GW], or  just when thinking in general.)

I have a bad idea. ("Send an object to the moon using a bird in a spacesuit") I don't write it down, instead trying to generate other ideas. Yet the bad idea keeps lingering on my mind. Eventually, I say "okay, fine, I will write you down!" and suddenly there's a sense of flow and relaxation, as if though new space has opened up for other ideas. 

Ed Sheeran describes something similar

If I was advising anyone to start songwriting, I’d just say: write anything that comes into your head. Anything. Just get stuff out there. I kinda treat songwriting as a dirty tap in an old house. When you run a dirty tap, it’s clogged up, bit of dirt in it, bits of mud. You run it, mud’s going to come out, a little bit of water, a bit more mud, and then suddenly it’s going to start flowing clean water.

You’ve got to unclog the pipes when you’re songwriting. So just start writing bad songs. Write songs for the sake of it. Write a song a day. Just sit down, pick a chord, write a song, get that song out of you, and the more and more you do it, the more and more you unclog the pipes.

I’d say, if you want to start songwriting, just start. Don’t worry too much about writing the best songs possible. If anyone wants to listen to growth, go on YouTube, type in “Ed Sheeran Addicted” and you’ll hear a song that I did when I was 12. And the singing is dreadful, the songwriting is dreadful, the guitar playing is dreadful, and now I’m here talking to you as a professional musician. 

When I did the "100 ways to light a candle" Babble challenge yesterday, something similar happened. I had the lingering idea of "I could ask someone else to do light the candle", but also thought "That's a boring answer". Yet it kept lingering. Finally, I wrote down "There's a general class of strategies around getting the help of other people". And suddenly that unblocked a whole range of similar ideas -- some of which I really liked, like: 

"Destroy a bunch of other light sources in the world, thereby strongly increasing the incentive for others to light this candle"

or

"Go around the neighbourhood, finding a bunch of lit candles, place them next to you in a beautiful pattern EXCEPT that in a symmetry-breaking place is your candle. Wait until someone fixes it." 

It was as if though the initial thought was a bunch of junk clogging the pipe. But by reifying the idea that this was an entire class of strategies, and writing the boring answer down, I suddenly opened up for the water to flow freely through the rest of the pipe. 

---

What on earth is going on here? Why would minds work like this? What does it tell us about the structure of human cognition? 

---

Also, I'm really excited about LessWrong's new prediction feature, so let's make some use of it: 

Answers

answer by ike · 2020-11-20T21:14:28.218Z · LW(p) · GW(p)

Your mind tracks the idea so as not to forget it. This reduces the effective working memory space, which makes it harder to think.

comment by ozziegooen · 2020-11-21T01:24:27.583Z · LW(p) · GW(p)

This is one of the key insights of Getting Things Done.

comment by Vladimir_Nesov · 2020-11-20T22:11:10.041Z · LW(p) · GW(p)

Not writing an idea down also helps you naturally think about it in the background. So it's more of a trade-off than a generally useful heuristic.

Replies from: NicholasKross
comment by Nicholas / Heather Kross (NicholasKross) · 2023-09-04T22:57:09.392Z · LW(p) · GW(p)

I wonder, though. I've found that writing things down makes them easier for me to reference, and thus I can consciously build on existing ideas/thoughts instead of rehearsing the same ones over and over again. E.g. codifying some AI timeline scenarios [LW · GW] before doing logic upon them [LW · GW].

(Maybe it is a tradeoff, and people with rumination and/or focus issues, like myself, find one side of the tradeoff generally-better.)

Replies from: Vladimir_Nesov
comment by Vladimir_Nesov · 2023-09-04T23:20:59.531Z · LW(p) · GW(p)

There was another exchange on this point recently [LW(p) · GW(p)]. In more detail, my preferred process is to hold off on writing things down for a bit to promote comprehension, then write them down within a few days to move on or have a better chance at partial reconstruction in the future. Not writing down at all makes forgetting to revisit in future or inability to sufficiently remember more likely. Writing down too quickly can turn an idea into a less useful caricature of itself, insufficiently familiar to settle into a natural formulation and no longer striving to do so once written down.

Replies from: NicholasKross
comment by Nicholas / Heather Kross (NicholasKross) · 2023-09-05T06:21:09.073Z · LW(p) · GW(p)

I may be able to try different "incubation periods" for ideas, to see which write-down times work well for me...

comment by jacobjacob · 2020-11-21T01:33:14.633Z · LW(p) · GW(p)

But why does my mind have such a strong need not to forget things, including things I consciously flag as "boring", that it would block off entire lines of good thoughts until I listen to it? 

Huh, maybe it somehow knows that saying the bad thought will unlock good trains-of-thoughts, and that's why it pushes so hard for it... 

Replies from: AllAmericanBreakfast
comment by DirectedEvolution (AllAmericanBreakfast) · 2020-11-23T07:44:31.034Z · LW(p) · GW(p)

The mind's a hack. So maybe it's just harnessing the same mechanism you use to suppress socially undesirable thoughts?

Rather than saying "forget about it," it's saying "ABSOLUTELY DO NOT SAY THIS THING."

Saying it shows the mind there's no bad consequences and allows it to stop focusing on avoidance.

answer by Slider · 2020-11-20T20:29:10.905Z · LW(p) · GW(p)

A person that has ot practised will have a theorethical understanding of the process and will think of the requirements ideally. Doing stuff makes it clear that things don't need to be ideal for things to happen. Taking steps proves that steps are possible instead of dreading over how can one ever cross a gap with mere stepping.

5 comments

Comments sorted by top scores.

comment by Raemon · 2020-11-20T21:57:48.431Z · LW(p) · GW(p)

Seems like maybe the OP should have a "do you experience a clogged-drain effect?" in addition to "do people generally?"

Replies from: jacobjacob
comment by jacobjacob · 2020-11-20T22:13:28.875Z · LW(p) · GW(p)

Feel free to make an Elicit question an add it as a comment :) 

comment by Slider · 2020-11-20T20:26:26.545Z · LW(p) · GW(p)

I have gone to a phase where you let them come out and just don't count them.

However there was a thing where I was pretty strict on distincness on the listings. Looking at others they were pretty lax on how different each option should be. There were like 20-30 points on others lists that would just fall under

 "chemical fuel"

in my brain. (I kinda feel that this is risking pushing against norm not to draw negatives to creative activities but I am focusing on my own here).

One could argue that is a form of blockage. And it was harder to get to 100 while simultanously that every option is not any of the other options.

And yet

 pyrokinesis and polymorph

 one could argue should collapse in some sort of phantasy option.

And yet it feels like 

use social pressures to get it done probably has a "umbrella option". Ask X, Sell X, demand X, intimidate X all seemof the formula Y X with social Y where variation of Y doesn't seem to make a fully distinct option. But it feels like my brain can't make that abstraction and somehow that fact makes it okay for them to be different.

probably has a "umbrella option".

 Ask X, Sell X, demand X, intimidate X

 all seemof the formula Y X with <property kind of interest> Y where variation of Y doesn't seem to make a fully distinct option. But it feels like my brain can't make that abstraction and somehow that fact makes it okay for them to be different.

A little like one could break down 

"nuclear fuel" into "fission fuel" and "fusion fuel" and have specific elements for fission fuels.

A superficial understanding makes <them> two phenomena but deeper understanding makes it two aspects of a single thing. With <property of the Y kind> I haven't unified it yet.

I find it easier when I ask myself "is this distinct" that more nebuloous "is this good?" questions don't bother me as much. Or like rather than thinking in terms of "quilty or not quilty" you have "probable cause", "indictable", "beoynd resonable doubt". So instead of being "crap" and "good" ideas you have categories like "explorable", "inspiring", "route", "limit-case", "fashionable""

comment by jacobjacob · 2020-11-20T19:35:31.602Z · LW(p) · GW(p)

(Seems to be a bug with the predictions. Pinged habryka about it.)

Replies from: Measure
comment by Measure · 2020-11-20T20:08:44.554Z · LW(p) · GW(p)

The seem to be working/displaying fine for me at this time.