Jack_LaSota comments on The correct response to uncertainty is *not* half-speed - Less Wrong

77 Post author: AnnaSalamon 15 January 2016 10:55PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Article Navigation

Comments (40)

Comments (40)

You are viewing a single comment's thread.

Comment author: Jack_LaSota 16 January 2016 06:33:38PM 4 points [-]

Anyone have a better procedure for fixing this than the following?

  1. Notice the feeling.
  2. Treat it as a signal that your S1 wants you to search for cheaper ways to figure out which option is right than continuing to drive. Search for cheaper ways and execute them. Make it a thorough search and show your S1 the thoroughness of your search. Acknowledge the awfulness of "drive back and forth in an expensive search pattern" and only choose that as a last resort.
  3. If you don't immediately become much more certain of which way the hotel is in, and the "go 30mph" feeling does not go away, treat it as a signal that your S1 thinks the thought process by which you chose (under evidence-starvation) is wrong, which does not necessarily mean that the conclusion is wrong.
  4. List the ways your S1 thinks you're biased which are screwing up your evidenced-starved reasoning.
  5. Perform sanity-inducing rituals to counter those biases. (Think about your actual goal of getting to the hotel as soon as possible, forgive yourself for maybe driving past it, imagine all 4 outcomes (60mph forward, 60mph backward) x (get to hotel on next try after this, don't get to hotel on next try after this) and how you would feel about them)
  6. If the feeling is still there, this procedure has failed.
Comment author: SatvikBeri 16 January 2016 08:24:32PM 5 points [-]

My procedure is probably similar cost, but more general:

  1. State my goal(s), e.g. "get to the hotel"
  2. Find the point of highest uncertainty towards the goal, e.g. "not sure if the hotel is ahead or behind me"
  3. Come up with plans for reducing the uncertainty, e.g. "find the next gas station and ask someone"
  4. Check whether the plan I have actually feels like it'll work

Note that this can be applied pretty broadly, e.g. to business strategy, software design, making friends etc.

Comment author: RomeoStevens 16 January 2016 11:45:20PM *  2 points [-]

My current model is that closed/exploitation mode just can not handle ambiguity. So the generalized cue becomes: notice ambiguity->pop into open/exploration mode and figure out what heuristic I actually endorse->back to closed mode.

Open mode: generating checklists

Closed mode: executing checklists

Open and closed research summary: https://www.youtube.com/watch?v=Qby0ed4aVpo

Comment author: RichardKennaway 17 January 2016 11:05:06AM 1 point [-]

Anyone have a better procedure for fixing this than the following?

When the implications of the situation are clearly perceived, the right action is effortless.

Comment author: Jack_LaSota 17 January 2016 08:01:46PM 0 points [-]

I agree, for a certain sense of the word "clearly". The procedure is to make you clearly understand the implications of the situation, which can be harder for some instantiations of the situation.

Powered by Reddit Powered by Reddit