Wanting Is Trickier Than It Looks

The Harvard psychologist Daniel Gilbert gave the problem its name: affective forecasting — predicting how we'll feel in a future we've only imagined. His research shows we're reliably bad at it. We overestimate how long both good and bad events will move us — Gilbert calls this the impact bias. The cottage by the sea is vivid in the imagining and strangely quiet by the second wet February.

This isn't a character flaw. It's how the mind works: when we picture a future, we leave out the ordinary texture of daily life — the admin, the weather, the way we adapt to almost anything. Which means “I'll know it when I feel it” is a weak test, because the feeling you're forecasting is not the feeling you'll actually get.

If the feeling can't be trusted, what can? Three tests that interrogate a want before you build a life around it.

The third test is the heart of what the Stanford designers Bill Burnett and Dave Evans call designing your life: you don't think your way into a new life, you prototype your way in. A prototype is a small, cheap, time-boxed experiment whose job is to generate real data — how it actually felt — rather than more imagining.

This matters most now precisely because, as Oliver Burkeman argues in Four Thousand Weeks, time is finite and choosing what to want (and what to give up) is the whole game. A prototype is how you choose without betting the farm.