Why Hard Decisions Should Be Easy (But Aren't)
We make dozens of decisions on a daily basis: what to have for breakfast, which task to complete first, which article to read.
Most of these decisions are easy.
But then there are the hard decisions — the ones we agonize over, the ones that lead to sleepless nights. These decisions are hard for two reasons: because no single option clearly dominates the alternatives, and because we expect our choice to have significant consequences. It's these two elements that explain why hard decisions should be easy — but are not.
Let's start with the first reason.
Decisions are hard when no single option clearly dominates the alternatives. If no option clearly dominates, then two or more options must be roughly matched in terms of their appeal. This seems intuitive enough, yet it prompts a compelling paradox. If the options are equally appealing (or equally unappealing, as the case may be), does it matter much which one we choose? Why not simply flip a coin and be done with it?
This paradox is spelled out in an influential book by Marvin Minsky called Society of Mind, where he attributes it to Edward Fredkin and formulates it like this: "The more equally attractive two alternatives seem, the harder it can be to choose between them — no matter that, to the same degree, the choice can only matter less."
It seems to follow that hard decisions should be easy, because our choice matters so very little — the most it can matter is the "difference" between the value of the best option and the value of the alternative that we find nearly indistinguishable. We might be so bold as to propose "Fredkin's formula" for hard decisions: Just flip a coin. The choice doesn't matter much anyway.
And yet, Fredkin's paradox is unlikely to soothe the sleepless. Fredkin's formula isn't a popular approach to making decisions about work or marriage. Hard decisions should be easy, perhaps, but they certainly don't feel that way. That's due to the second reason hard decisions are hard: We perceive the choice to be consequential.
Consider the choice between two flavors of ice cream: chocolate or vanilla. You find the two options equally appealing (say), so the decision is hard in the sense that one option doesn't clearly dominate the other. But here the stakes are low; getting it wrong isn't life-changing. Even if you choose vanilla and discover that it's much worse than you expected, even if your friend who ordered chocolate moans with pleasure, your regret will be short lived.
Not so for truly hard decisions. When it's a matter of life partners or surgical procedures rather than ice cream flavors, the stakes are much higher. Correspondingly, the cost of getting it wrong, and the possibility for regret, loom much larger.
Consider the decision to move to Los Angeles versus New York. Consider the decision to pursue medical intervention A versus medical intervention B. These might be equally (un)appealing based on the information you have available right now, such that their expected utilities are roughly the same. And so Fredkin's paradox applies: the decision will be hard but maybe shouldn't be; the decision "doesn't matter" in the sense that it doesn't affect your expected utility.
The trouble is the future, when that expected utility is realized.
It's possible that your life in Los Angeles will be different, but just as good, as your counterfactual life in New York. It's possible that your health will be the same after medical intervention A or medical intervention B. But it's also possible that one of these options would in fact be much better for you than the other. Their expected utilities could be the same, and yet their actual utilities could diverge. (Consider a choice between two bets: I'll give you $5 if my quarter lands heads, or I'll give you $5 if my dime lands heads. The expected utility for each bet is the same, but in fact, it might be that only one of these bets yields a win.)
I think this is why Fredkin's paradox isn't soothing; why Fredkin's formula might rub us the wrong way. What's true of the decision process — that our choice "doesn't matter" — isn't true of the decision outcome. If we could only tell the future, we might see that the decision does matter — that it matters quite a lot. This is the sense in which we perceive the decision to be consequential.
What matters for hard decisions, then, isn't only a decision procedure that will maximize the odds of choosing the (perhaps marginally) better choice, but also one that will minimize future regret.
Hard decisions are hard because the process keenly matters, not just the choice or its outcome.
Tania Lombrozo is a psychology professor at the University of California, Berkeley. She writes about psychology, cognitive science and philosophy, with occasional forays into parenting and veganism. You can keep up with more of what she is thinking on Twitter: @TaniaLombrozo