No, You Can’t “Always Change Your Mind”

On sticky defaults

Image created with ChatGPT

There often comes a moment, around the end of a big restaurant order, when eyes move searchingly around the table. “You think that’s enough?” someone asks, half to the group—then turns to look at the server.

The server taps their pad and pretends to do some math. “I think so!” they say. “How hungry are you?”

At this point the would-be voice of reason steps in: “Let’s hang onto a menu,” this unassuming hero says. “We can always order more!”

Here’s the thing, though: You actually can’t order more. Or, rather: You can theoretically order more—there’s no law that says you can’t order more—but in practice you won’t. In my almost-three-decades on this planet, I can think of maybe one time I’ve been at a restaurant where this has happened (and that was in all likelihood just an extra side of fries!). I expect if you search your memory, you’ll find the same.

The problem here is that at the moment of decision, we fail to understand what class of choices we’re actually dealing with. In the midst of our initial order, we are selecting between option A (sticking with the current dishes) and option B (adding more food); this may be a “hard” choice in that it is based on imperfect information, but it is one the group is highly motivated to make. For this reason, we’re able to reach a consensus with relative speed and conviction.

Once the order is placed, though, we’re no longer deciding between A and B; we are switching from A to B. And that is an entirely different decision. The mistake of “We can always order more!”, in other words, is that it falsely projects favorable decision-making conditions into a future where they no longer hold.

The truth is, choices create defaults, and defaults are surprisingly sticky—a facet of human behavior that repeatedly emerges in psychology and behavioral economics. In its most general form, our preference for defaults is known simply as status quo bias. Various researchers have identified related concepts like loss aversion (the perception that a given loss is more painful than an equivalent gain is pleasant) and the endowment effect (the tendency to value what we own more highly than if it were owned by someone else). In one recent study, participants flipped a coin to decide at random whether or not to make a change in their life; those who changed were significantly happier for many months afterwards. All of this evidence converges on a consistent story: We are systematically slower to make changes than is rational, preferring what we’ve got to what we might otherwise have.

To be sure, there can be good reasons to prefer the status quo; switching is not costless. In fact, situations that are worse in the short-run can actually sometimes benefit us in the long-run: Discomfort gives us the much-needed push to overcome the friction of switching. (Psychologist Dan Gilbert refers to this common “bad-but-not-so-bad-I’m-willing-to-change” trap as the region-beta paradox.¹) Nor am I arguing that growth is impossible or that people can never be persuaded—just that the ease and predictability of change are illusions. Certainly, it’s an error to assume you can “always” correct course.

One implication of status quo bias, then, is that when we recognize we’re making a choice based on fuzzy or incomplete information, we should set up explicit nudges to revisit the decision. This might mean, for example, committing to two six-month trial periods for a couple different careers we’re considering; or simply writing down the reasoning behind our initial choice and setting a reminder to see if that reasoning still holds a year from now. Without these nudges, we’re much more likely to sleepwalk down paths we initially took almost at random.

Admittedly, telling yourself you can always change your mind can be a useful bit of self-delusion if it spurs you to make a hard choice you wouldn’t otherwise have made. While certainty about future flexibility is on-its-face irrationally optimistic, it may be meta-rational in the sense of accurately accounting for your own irrationality.²

But sometimes—and if I’m being honest, this is the part that bothers me most—“you can always change your mind” is just a shorthand for “the choice doesn’t matter.” That, of course, is the greatest lie of all. Choices matter, and they always will.

1.  Some other examples of the region-beta paradox, lifted from the Wikipedia page:

• If someone is currently in a non-ideal romantic relationship, they will be less likely to end it to find an ideal relationship than if their current relationship was further non-ideal, thus making their current overall situation worse.

• The same would be true if they were in a not-preferred job which is still acceptable, where the perceived strain in finding a new job makes them stuck in region alpha.

• Another useful example is if one prefers living in a tidy home, a small mess might not be enough to start cleaning up. A slightly bigger mess would cause enough discomfort to start cleaning up, thus reaching the end goal of a tidy house faster.

2. In other words: If it’s psychosocially difficult to choose A over B, but A is in fact the better choice, saying “I can always switch to B later!” is descriptively false but pragmatically effective in producing the best outcome.