[SEQ RERUN] Teaching the Unteachable
Today’s post, Teaching the Unteachable was originally published on 03 March 2009. A summary (taken from the LW wiki):
There are many things we do that we can’t easily understand how we do them. Teaching them is therefore a challenge.
Discuss the post here (rather than in the comments to the original post).
This post is part of the Rerunning the Sequences series, where we’ll be going through Eliezer Yudkowsky’s old posts in order so that people who are interested can (re-)read and discuss them. The previous post was Unteachable Excellence, and you can use the sequence_reruns tag or rss feed to follow the rest of the series.
Sequence reruns are a community-driven effort. You can participate by re-reading the sequence post, discussing it here, posting the next day’s sequence reruns post, or summarizing forthcoming articles on the wiki. Go here for more details, or to have meta discussions about the Rerunning the Sequences series.
The Moore Method seems relevant.
Vague thought, drawing on (!) Betty Edwards’ approach to drawing: it seems that communication between people is verbal by default. Betty Edwards has some anecdotes suggesting that groups can learn some things from communicating to each other visually (drawing representations of their thoughts) that they can’t easily learn from communicating to each other verbally. Have people followed up on this idea? It seems like potentially valuable lowish-hanging fruit in improving interpersonal communication, especially for people who don’t think primarily verbally.
(I asked Val if CFAR had done anything along these lines. Apparently the 2011 megacamp had a drawing portion and then it was never tried again. He didn’t know more about why though.)
Drawing can be also used to teach a skill of “being specific”.
Divide students in pairs. The students sit with their backs to each other, no one can look at what the other one is doing. Prepare a picture and give it to one student. The student is supposed to describe it and the other student is supposed to draw the same picture based on the description. At the end, compare the two pictures. Then look at the pictures produced by other pairs, and discuss why sometimes things went right or wrong.
(An example of a typical mistake: One student says “there is a monster with a big oval head”. But he did not specify how big, so the other student draws a head large 80% of the paper size. But the first student simply meant “big in proportion to the body size”, not even “greater than the body”. Or the first student says the monster has a shirt with stripes, but does not mention the direction, number, or thickness of the stripes. Etc. By the way, various monsters are typically used to reduce the shared knowledge among participants, so you need to transfer more bits of information.)
There are also two variants of the game, depending on whether the second student is allowed to ask questions, or just has to draw silently. This explains the usefulness of feedback.
Some of the expert elicitation stuff I’ve read recommends having experts draw distributions for some uncertain variable, then show each other their distributions, and then talk about why they did what they did- this’ll often reveal where they have different information quickly and easily, and this works much better than having them compare point estimates or verbal descriptions.
Similarly, drawing causal models and Bayes nets of situations seems like an easy visual representation of things that may be difficult to quickly convey verbally.
What sort of things are you imagining communicating?
I’m imagining something less precise. Betty Edwards describes running corporate training seminars where she helped groups in corporations address their work problems, starting with drawing exercises and then using drawing to describe the problems they’re working on. She reports:
I’m not sure exactly what’s happening in these examples, but I think this might fall under “communicating with the emotional brain” or something like it.