Sam Gentle.com

Goal substitution

It's difficult to overstate how good Daniel Kahneman's Thinking, Fast and Slow is. It's a book with so much to offer for understanding human behaviour and decision making. One particularly eye-opening phenomenon described early in the book is attribute substitution, where you take a hard question and replace it with an easier, hopefully equivalent one.

If asked "is Dave nice?", a good answer would involve some kind of deep analysis of Dave's character, but maybe that's too hard to do on the spot. Instead, you substitute an easier question like "can you easily remember a time when Dave did something nice?". So you answer the substitute question but, crucially, you still feel like you're answering the original question. This optimisation works well most of the time, but can lead to some pretty wacky results when it fails.

A similar idea I've been thinking about is goal substitution. Let's say you want to have fun. One good way to do that is to call up some friends, go out and have a good time. But maybe you don't do that, and instead lie on the couch all evening watching TV or reading junk internet. If you tried both options, you'd rate the first one as much more enjoyable. So why don't you do it? My theory is that you substituted the goal of "do something fun" with "do a lesiure activity". The second goal is easier than the first, so you achieve it instead.

This also happens with work. You want to get some good work done, but it's easy to substitute that goal for "do something that feels like work". The problem is, lots of things can feel like work even if they aren't work. Reading emails, checking up on news, researching some technology you might use – these things definitely feel like work, in the sense that they're mentally stimulating and related to work, but may not actually move you closer to your real goal.

In both cases, the issue isn't necessarily redefining your goals. Often that can be useful to avoid overloading yourself or taking an unnecessarily amount of context. It's fair to redefine "go to space" as "assemble and manage the best team of people for going to space". The problem is when this happens subconsciously. You still feel like you're achieving the original goal when in fact you're doing something different. Much like with attribute substitution, this works well when it works, but can misfire badly.

I think goal substitution is a particular issue because a lot of entertainment does not necessarily have to be fun, merely compelling. The main evaluation that entertainment creators make is to measure consumption, not enjoyment. So in a sense entertainment has evolved to target the "feels like leisure" substitute goal very effectively. With such a glut of available fun-like and work-like activities, it's very easy not to notice whether you're actually having fun or doing work.

As for solutions, the best I can offer is the same advice I've heard about attribute substitution: make sure the decision takes time. Substitution works because it takes a hard problem and gives you an easy alternative. If you force yourself to spend five minutes thinking about "is Dave a nice guy?" you won't feel the urge to substitute because there is no easy alternative.

Similarly, taking time before doing something to figure out if it meets your actual goal should remove the allure of the easier substitute goal. I admit that's easier said than done; mostly the decision of what to do next happens on autopilot. We seem particularly prone to optimisation, even when it does us more harm than good.