CstineSublime's Shortform

CstineSublime

This is a special post for quick takes by CstineSublime. Only they can create top-level comments. Comments here also appear on the Quick Takes page and All Posts page.

How can you mimic the decision making of someone 'smarter' or at least with more know-how than you if... you... don't know-how?

Wearing purple clothes like Prince, getting his haircut, playing a 'love symbol guitar' and other superficialities won't make me as great a performer as he was, because the tail doesn't wag the dog.

Similarly if I wanted to write songs like him, using the same drum machines, writing lyrics with "2" and "U" and "4" and loading them with Christian allusions and sexual imagery, I'd be lucky if I'm perceptive enough as a mimic to produce some pastiches. However if I wanted to drill further, how might I 'black box' his songwriting mind, reverse engineer which cruxes and decision pivots determine what rhyming or rhythm patterns he chooses, what chord progressions he operates on. Maybe after years of doing this I'd have a model composed of testable hypotheses that I could run experiments on, either by reverse engineering songs of his at random and seeing if they hold to the patterns I observed, writing my own songs in this manner and seeing if they have that 'x-factor' (hardest and most subjective of all), and finally comparing the stated narratives in biographies and interviews about how certain songs were written in accordance with my hypotheses.

Of course someone is going to say that you can't reduce a human being, let alone a super-talented human being to a formula, and perhaps draws a long bow about why they don't like A.I. art or modern Hollywood or whatever. All sentiments I'm sympathetic too even if I'm not 100% sold on.

What I'm thinking about is not too dissimilar from what Ray Dalio advises: One shouldn't just trust an expert's conclusion or advice blindly, even if they have a unparalleled pedigree.

But because I'm pretty extreme in believing that it is important to obtain understanding rather than accepting doctrine at face value, I would encourage the new batter not to accept what [Babe] Ruth has to say as right just because he was the greatest slugger of all time. If I were that new batter, I wouldn't stop questioning Ruth until I was confident I had found the truth.

In both cases rather than just taking the end result blindly - writing parodic or pastiches songs - the tail doesn't wag the dog - there is an attempt to find out why, to question!

My problem isn't so much that Prince (or Babe Ruth) are no longer around to answer these questions, but that unlike a multi-billionaire like Ray Dalio, anyone with sufficient pedigree is unlikely to pick up the phone and answer my incessant questions about "why?" and "how come?". I have to black-box it.

2Viliam2y

Yes, my conclusion is "it's hopeless". (NLP assumes that you could reverse-engineer someone's thought processes by observing their eye movements. That looking in one direction means "the person is trying to remember something they saw", looking in another direction means "the person is trying to listen to their inner voice", etc., you get like five or six categories. And when you listen to people talking, by their choice of words you can find out whether they are "visual" or "auditive" or "kinesthetic" type. So if you put these two things together, you get a recipe like "first think about a memory that includes some bodily feelings, then turn on your auditive imagination, then briefly switch to visual imagination, then back to auditive, then write it down". They believe that this is all you need. I believe that it misses... well, all the important details.)

1CstineSublime2y

Sorry I made a mistake in my last reply: putting NLP aside, are there any effective methods of reverse engineering the decision making of people that you can't get on the phone? There's an abundance of primary evidence for many decisions, whether it be minutes of deliberations, press releases which might involve more reading of the tea-leaves. In the case of Prince one could possibly listen to different live-performances of the same song and analyze what changes are made. What words are crossed out on a lyrics sheet. Many times people have to become very good at intuiting people in their life who are loathe to actually explain their reasoning, yet build pretty useful models of how to interact with those people. From grumpy shopkeepers, to school teachers, to coworkers etc. etc. Diplomacy is an entire profession based on building such models. Negotiation builds those models under pressure - but often has the ability to speak with the other side, as per Ray Dalio's suggestion, which I'm trying to find a method for. Are there no methods of understanding and reverse engineering the reasoning, not the superficial aspects, of another person?

Dunno; it probably depends a lot on the kind of task, the kind of person, and your observation skills. Some people explain more, some people explain less. Some people are more neurotypical (so you could try to guess their patterns by observing what other people similar to them would do in a similar situation), some people are weird and difficult to predict. At some tasks people produce artifacts (a mathematician can make notes on paper while solving a problem; if you obtain the paper you could reconstruct some of their thoughts), other tasks happen mostly ... (read more)