Viliam_Bur comments on The Power of Reinforcement - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (467)
I do accept this kind of reinforcement from my significant other, assuming that:
Actually I consider it very useful, and for me it would be a waste not to use this kind of cheap "external willpower". YMMV.
Note that I consider the reinforcement you are describing to be entirely different in kind (not "this kind"). The boundaries around the kind I accept are approximately the same as yours:
I go by what my intuition tells me but when formalizing those intuitions something similar is generated.
I make a point of rewarding desired reinforcement (while attempting 'extinction' on less desirable influence tactics like nagging or punishment.)
I supposed the reason why the husband in the story didn't put his clothes in the hamper was that he was too lazy to do that, not that he (terminally) valued that the clothes stayed outside the hamper.
Having a terminal value for clothes outside the hamper isn't the point. It is whether given the negotiated relationship boundaries and typical behaviors as they currently are the person being modified would prefer "status quo except I do <influenced behavior> more" over "status quo".
"Too lazy" can be left out of such considerations. That doesn't distinguish between akrasia and considered intent not to do the thing (for whatever reason). For most part judgements like "too lazy" are just another method of attempting influence - usually a method that is inferior to reinforcement.
Well, making judgments like "too lazy" can also provide valuable social cover for other kinds of reinforcement (or punishment), within communities where deliberately altering the behavior of others is seen as unacceptable unless I can frame it as being for their benefit.
More generally, motivated speculation about other people's best interests (including but not limited to positing that they possess unexpressed "terminal values" that happen to align better with what I seem to want than with what they seem to want) can be a very useful way to ignore people's stated preferences without feeling (or being seen by third parties as) indebted to them.