by Zvi
4 min read22nd Jul 20171 comment


Epistemic Status: Spent way too long deciding what to put here

Kind of a Follow-Up to: Change is Bad

Will Hopefully Lead To, But No Promises: Some combination of, when and if I write them: Complexity is Bad, Choices Are Really Bad, Frontload Choices, Delay Choices, Avoid Choices, Simplify Choices, Destroy Choices, Make Good Choices, On Choice Algorithms

Note that the title is not: Against Choices

We all know about The Paradox of Choice. There have been posts and lectures, and a book by that name.

It’s even two of the ten principles of economics, as is explained here:

I recommend it. It’s quite fun and is more insightful than you would think. I’ll assume below that you’ve watched at least the explanation of the first two laws.

Choices Reduce Perceived Value

If I give you a snickers bar you value at a dollar, you feel like you got a free dollar. It’s easy, all you’re thinking is ‘free Snickers bar, sweet.’ If I give you a choice between a snickers bar and some M&Ms that you value at sixty cents, you still feel like you got a free snickers bar, which is a lot like getting a free dollar (since in this example you value it at a dollar). The trick is, now you had to start with ‘free candy, sweet’ to ‘which of these do I want more?’ and then to ‘the Snickers bar is forty cents better.’

When you finish this sequence, even if you know exactly what you want, you are left with the feeling of making a choice where you did better by forty cents. Most people typically don’t quantify these effects, but in my model they often still do have the gut feelings involved.

Even worse, you might be thinking ‘I had a choice of candy and I chose the Snickers bar’ and comparing the value of the Snickers bar not to the value of the M&Ms, but to the value of a choice of two candies. Now things get dangerous. You could reasonably value a choice of candies very close to or higher than a dollar! If you are thinking Choices Are Good rather than Choices Are Bad, you could even think of the optionality as being worth more than the choice you eventually make, and now you got a free candy bar and potentially actually feel bad about it, because you lost value. You might even get tricked into thinking about how what you really wanted was a Milky Way, or a banana.

The more you were forced into choosing mode, the more the choice was a real choice requiring consideration, the more you end up thinking in this way. This is why the two identical Snickers bars is mostly just a funny joke; your brain hopefully does not go into choosing mode, because it notices there is no reason to do so, and stays in ‘free candy, sweet.’ Similarly, if you have a well-established rank order and your brain didn’t register the option to take the M&Ms, you mostly get to avoid the problem and enjoy your Snickers bar. However, if you stopped and thought for a few seconds, you could be in trouble, and if you think for a while, the cost of that thinking starts to factor in as well, making things even worse. Time spent on a choice makes you feel worse about all the wasted time, and you feel worse because you judge the final outcome against a higher standard.

A movie you would have enjoyed if it was just on makes you feel like a schmuck if you take twenty minutes choosing it from thousands at the video store.

Search your feelings. You know it to be true.

Choice Hijacks You into Choosing Mode

When you have a choice, you must stop what you’re doing, and choose.

Choosing is hard. You probably need to figure out what all your choices are, figure out at least vaguely how good they are, and choose one of them.

That sounds like work, which is bad (citation needed, but writing it seemed like work, and again, work is bad).

At a minimum, you need to figure out what one of your choices is, and choose that one.

Making a random slash arbitrary choice might not be hard (although sometimes it is) but first you have to choose to choose at random. If you choose not to decide, you still have made a choice. There’s no escape! You certainly don’t want to be making that choice at random times!

While this sounds petty, it’s important because making a choice takes you into choosing mode. You reorient, gather information, decide if the choice matters, decide how much it matters, then actually work on making the choice.

When you finish, you’ve often lost your old thread entirely.

If the choice could really matter, this process could get really expensive. It is very easy to not only get taken out of what you are doing, but then to spend quite a bit of time and effort on decisions even when those decisions barely matter. Making a wrong decision is not only risky and bad due to object level consequences, it also risks being wrong and being wrong is just awful. 

When you force someone to choose, even a trivial choice, you’re imposing a real cost even if they do not care. Forcing them to change modes is expensive.

Avoid this, and staying out of choosing mode, is vital to many tasks. Stop choosing how to do the thing and do the thing!

An important task in this class is relaxation. This is a lot of why hypnosis is so relaxing. Relaxing requires extended periods without choice. I have a lot of trouble relaxing, and I realized recently that a lot of this is about my being bad at deliberately front loading and then avoiding choices. Choices are bad.


New to LessWrong?

New Comment
1 comment, sorted by Click to highlight new comments since: Today at 6:25 AM

I think a somewhat better title for this would have been "Choices are Costly." (Roughly as accurate, roughly as easy to work into conversation, less value laden, more appropriate in more circumstances)