I wrote a coupleposts on my personal blog a while ago about creativity. I was considering cross-posting them here but didn’t think they were LessWrong-y enough. Quick summary: I think because of the one-way nature of most problems we face (it’s easier to recognize a solution than it is to generate it), pretty much all of the problem solving we do is guess-and-check. That is, the brain kind of throws up solutions to problems blindly, and then we consciously check to see if the solutions are any good. So what we call “creativity” is just “those algorithms in the brain that suggest solutions to problems, but that we lack introspective access to”. The lack of introspective access means it’s difficult to pass creative skills on—think of a writer trying to explain how to write well. They can give a few basic rules of thumb, but most of their skill is contained within a black box that suggests possible sentences. The actual writing process is something like “wait for brain to come up with some candidate next sentence”, and then “for each sentence, make a function call to ‘is-sentence-good?’ module of brain” (in other words, guess and check). Good writers/creative people are just those people who have brain algorithms that are unusually good at raising the correct solution to attention out of the vast possible space of solutions we could be considering. Of course, sometimes one has insights into a rule or process that generates some of the creative suggestions of the brain. When that happens you can verbalize explicitly how the creative skill works, and it stops being “creative”—you can just pass it on to anyone as a simple rule or procedure. This kind of maps nicely onto the art/science divide, as in “more of an art than a science”. Skills are “arts” if they are non-proceduralizable because the algorithms that generate the skill are immune to introspection, and skills are “sciences” if the algorithms have been “brought up into consciousness”, so to speak, to the point where they can be explicitly described and shared (of course, I think art vs science is a terrible way to describe this dichotomy, because science is probably the most creative, least proceduralizable thing we do, but what are you gonna do?)
Anyway, I don’t know if all of this is just already obvious to everyone here, but I’ve found it a very useful way to think about creativity.
Edit: I missed your last sentence somehow. The above is definitely just plausible and/or fun to read.
That’s very interesting but I think that some people can explain their creative process. Bruce Adolphe is musician who hosts a piano puzzler radio show. If you are a classical music buff I highly recommend it. He takes well known songs or pieces of music and then rewrites them in the style of an old composer. (The challenge is find the popular song buried within and figure out which composer it is in the style of). Anyways in each of his segments, he explains exactly what he is doing to imitate the style of that composer and how he incorporated the popular song into it.
Your approach looks quite unscientific to me. What empirical evidence do you have to support this? How would you go about codifying these ideas into a proper scientific theory?
I mean, I don’t really disagree; it’s not a very scientific theory right now. It was just a blog post, after all. But if I was trying to test the theory, I would probably take a bunch of people who varied widely in writing skill and get them to write a short piece, and then get an external panel to grade the writing. Then I would get the same people to take some kind of test that judged ability to recognize rather than generate good writing (maybe get some panel of experts to provide some writing samples that were widely agreed to vary in writing quality, and have the participants rank them). Then I would see how much of the variation in writing skill was explained by the variation in ability to recognize good writing. If it was all or most of the variation, that would probably falsify the theory—the theory would say the most difficult part of “guess and check” is the guessing part, but those results would say it’s the checking.
Assuming this general type of theory is vaguely correct (which I find plausible), it suggests that creativity depends on both some potentially innate creativity algorithms combined with lots of knowledge. Acquiring as domain knowledge is important for two reasons: firstly it gives one more insights/ideas to recombine, and secondly it probably indirectly trains the creativity algorithms themselves (assuming the brain is constantly trying to improve its ability to predict new novel ideas it encounters).
I wrote a couple posts on my personal blog a while ago about creativity. I was considering cross-posting them here but didn’t think they were LessWrong-y enough. Quick summary: I think because of the one-way nature of most problems we face (it’s easier to recognize a solution than it is to generate it), pretty much all of the problem solving we do is guess-and-check. That is, the brain kind of throws up solutions to problems blindly, and then we consciously check to see if the solutions are any good. So what we call “creativity” is just “those algorithms in the brain that suggest solutions to problems, but that we lack introspective access to”. The lack of introspective access means it’s difficult to pass creative skills on—think of a writer trying to explain how to write well. They can give a few basic rules of thumb, but most of their skill is contained within a black box that suggests possible sentences. The actual writing process is something like “wait for brain to come up with some candidate next sentence”, and then “for each sentence, make a function call to ‘is-sentence-good?’ module of brain” (in other words, guess and check). Good writers/creative people are just those people who have brain algorithms that are unusually good at raising the correct solution to attention out of the vast possible space of solutions we could be considering. Of course, sometimes one has insights into a rule or process that generates some of the creative suggestions of the brain. When that happens you can verbalize explicitly how the creative skill works, and it stops being “creative”—you can just pass it on to anyone as a simple rule or procedure. This kind of maps nicely onto the art/science divide, as in “more of an art than a science”. Skills are “arts” if they are non-proceduralizable because the algorithms that generate the skill are immune to introspection, and skills are “sciences” if the algorithms have been “brought up into consciousness”, so to speak, to the point where they can be explicitly described and shared (of course, I think art vs science is a terrible way to describe this dichotomy, because science is probably the most creative, least proceduralizable thing we do, but what are you gonna do?)
Anyway, I don’t know if all of this is just already obvious to everyone here, but I’ve found it a very useful way to think about creativity.
Edit: I missed your last sentence somehow. The above is definitely just plausible and/or fun to read.
That’s very interesting but I think that some people can explain their creative process. Bruce Adolphe is musician who hosts a piano puzzler radio show. If you are a classical music buff I highly recommend it. He takes well known songs or pieces of music and then rewrites them in the style of an old composer. (The challenge is find the popular song buried within and figure out which composer it is in the style of). Anyways in each of his segments, he explains exactly what he is doing to imitate the style of that composer and how he incorporated the popular song into it.
Your approach looks quite unscientific to me. What empirical evidence do you have to support this? How would you go about codifying these ideas into a proper scientific theory?
I mean, I don’t really disagree; it’s not a very scientific theory right now. It was just a blog post, after all. But if I was trying to test the theory, I would probably take a bunch of people who varied widely in writing skill and get them to write a short piece, and then get an external panel to grade the writing. Then I would get the same people to take some kind of test that judged ability to recognize rather than generate good writing (maybe get some panel of experts to provide some writing samples that were widely agreed to vary in writing quality, and have the participants rank them). Then I would see how much of the variation in writing skill was explained by the variation in ability to recognize good writing. If it was all or most of the variation, that would probably falsify the theory—the theory would say the most difficult part of “guess and check” is the guessing part, but those results would say it’s the checking.
That’s the first thing to come to mind, anyway.
Assuming this general type of theory is vaguely correct (which I find plausible), it suggests that creativity depends on both some potentially innate creativity algorithms combined with lots of knowledge. Acquiring as domain knowledge is important for two reasons: firstly it gives one more insights/ideas to recombine, and secondly it probably indirectly trains the creativity algorithms themselves (assuming the brain is constantly trying to improve its ability to predict new novel ideas it encounters).