How to stop second-guessing a decision you already made

It is the fourth night since you paid for the car. You are lying in bed with the comparison open on your phone again: the hatchback you bought, the compact SUV you didn't, the same eleven reviews you have already read. Nothing in them has changed since Saturday. You read them anyway.

If a decision is made and the mind keeps reopening it, the useful question is whether anything new has arrived since you decided. A new fact deserves a proper look, on a date you set. A new feeling about old facts is a rerun, and running it again mostly produces doubt. The rest of this piece is how to tell the two apart at midnight, and what to do with each.

Why does the case keep reopening?

Partly because reopening feels responsible. A careful person checks their work. The trouble is that a decision is not a sum you can re-add, and the second pass does not run on the same inputs as the first.

In 1991 Timothy Wilson and Jonathan Schooler compared college students' preferences for brands of strawberry jam with experts' ratings of the same jams. Some of the students were asked to analyse why they felt the way they did. Others were not. "Students who analyzed why they felt the way they did agreed less with the experts than students who did not" (Wilson and Schooler, Journal of Personality and Social Psychology, 1991). A second study, on choosing college courses, found the same direction. Their explanation is the useful part: "Analyzing reasons can focus people's attention on nonoptimal criteria, causing them to base their subsequent choices on these criteria."

That study was about forming a preference, not doubting one afterwards, so what follows is reasoning built on it. The midnight rerun is that kind of reason-analysis. The reasons that come easily at midnight are the ones that fit into words: mileage, resale value, boot space in litres. The thing that decided it on Saturday, how the car felt in your own narrow lane, does not fit into a sentence. So each rerun weighs the case a little more by the wordable criteria and a little less by the one that mattered. The doubt grows without any new information arriving.

Does keeping it open cost anything?

Yes. Daniel Gilbert and Jane Ebert ran a study with photography students choosing which prints to keep. Some had the chance to change their minds afterwards; some did not. The ones who could "liked their prints less than those who did not". The students had expected the option to make no difference to how they felt, and "most participants wanted to have that opportunity" anyway (Gilbert and Ebert, Journal of Personality and Social Psychology, 2002). The authors write that the chance to change their minds "impaired the postdecisional processes that normally promote satisfaction".

In that study the option to swap was real. The dealer is not taking your car back. Reopening the comparison every night looks like a homemade version of the same thing, though that is an inference and not something the study tested: the decision is closed on paper and held open in your head. The car has not got worse since Saturday. You have just not let it be yours yet.

What counts as new?

A new fact is something you did not know when you decided. A fault the dealer did not mention. Your office announcing a move to the other side of the city.

A new feeling about an old fact is not new. "The boot really is small" is not new if you saw the boot. A review you had not read that says what the others said is not new. Your brother-in-law's opinion is usually not new either, unless he knows something about the car you didn't.

Most of what arrives at midnight is the second kind. It feels like news because it comes with a jolt. The jolt is the only new part.

What to do when the loop starts

Step one belongs on the day you decide, while the reasons are still in front of you. If that day has passed, do it tonight from memory and treat it as a rough copy.

1. Write the decision record. Three lines on why you chose it, in your own words, including the reason that is hard to defend ("it felt right in the lane"). Then a fourth line: I will reopen this if: followed by one or two specific facts that would genuinely change the answer. Not "if I start to feel bad". A fact, with a number on it if it can have one.
2. When the loop starts, ask one question. What is new since I decided? If you cannot name a fact, it is a rerun. Say so, in those words. "Rerun." Then make a mark on a tally on the same page. It sounds silly. It is useful later.
3. If something is new, check it against your reopen line. If it matches, the decision is open again, properly, in daylight. If it doesn't match, write it under the record with the date and leave it there.
4. Set one review date. A day when you will have lived with it long enough to know something: three months for a car, the end of probation for a job. Put it in the calendar. Reviews happen on that day, against the written record, not in bed.

The record keeps Saturday's reasons in Saturday's words. The review date gives the doubt somewhere to go that is not tonight.

The car, carried through

An illustration, not anyone's real car. Say you bought a four-year-old petrol hatchback for ₹5.4 lakh on a Saturday, over a compact SUV at ₹7.9 lakh. On Sunday you write the record. One: no loan, and ₹2.5 lakh stays in the bank. Two: it fits the parking spot in your building, which the SUV only just did. Three: it felt easy in the lane, and the SUV felt like work. Reopen if: a repair over ₹40,000 in the first year, or we need to carry more than four people regularly.

The first week is bad. You reread reviews three nights out of five. On Wednesday a colleague says the SUV holds its resale value better. On Friday you sit in a friend's SUV and the seats are higher. Each time, you ask the question. Resale value was in the reviews you read before you bought. You had sat in that same SUV at the dealer. Nothing new. Rerun, rerun, rerun. By the end of the second week there are nineteen marks on the tally, and you notice you never set a review date. You put one in the calendar for three months out.

In week three something new does arrive. At the first service the mechanic says the clutch plate will need replacing within a year, roughly ₹12,000. A real fact, so you check it against the reopen line. ₹12,000 is not ₹40,000. You write it under the record with the date and leave it.

With a date to send them to, the marks thin out. Four in week three. One in week five.

Three months later, on the review date, you open the record. Reason one held: the money is still in the bank. Reason two held, and more than you expected, because a neighbour's SUV has been scraped twice backing out of the same row. Reason three you cannot assess any more; you just drive the car. The honest verdict is that it was a good decision and probably not the best possible one. The SUV would have been nicer on the highway to your in-laws' place, and on that drive you still think so.

You keep the car. You write one line at the bottom, reviewed, keeping it, and close the notebook. Ho gaya. The highway thought still comes back on that drive, and it is allowed to. It just does not get the comparison tab any more.

When is reopening the right thing to do?

Sometimes the doubt is right. The test is built to let that through.

When a new fact matches your reopen line, reopen it fully. That is what the line is for.

When the decision is cheap to undo and the window is closing (a return period, a cancellation date, a lease with an exit clause), a quick honest look before the deadline costs little. Pushing it to a review date three months away would be avoidance. Do the review early, once, and let the window close.

When it costs more every month, as with a course you keep paying for and never attend or a job that is wearing you down, "I already decided" turns into a reason to keep paying for a mistake. Bring the review date forward.

And when the doubt is about something else. If you second-guess every purchase over a certain amount, the car is probably not the subject. Money worry is, and the car is where it lands this month. A decision record will not fix that, though it may help you see it.

Where this advice does not work

When you never really decided. If the choice was made for you, or you agreed to end an argument, the reopening may be the first honest look you have had. There are no reasons of your own to check against. The tools for that are in deciding when you are genuinely torn.

When the reopen line was written to be safe. If your condition can never happen, you have built a lock and called it a review. If you cannot think of a fact that could plausibly arrive, you were not sure on the day, and it is better to know that.

When the tally keeps climbing. If the marks do not thin out after the review date is set, and the same choice still takes hours of every week, a notebook is not enough. Talk it through with someone who will not just reassure you. If it happens with nearly every decision, that is beyond what a blog post can do.

When it affects someone else. A decision about a parent's care or a child's school has another person in it, and their experience is new information every week. Review sooner and more often, and ask them.

If you want somewhere to say the doubt out loud and hear back whether anything in it is actually new, that is a reasonable use of Cabin. If the thing keeping you up is a conversation rather than a choice, replaying conversations is the closer piece.

This article is for everyday reflection and isn't a substitute for professional mental-health care. If you're struggling or in crisis, please reach out to a professional or a helpline. In India you can call Tele-MANAS at 14416 (24/7, free, 20 languages) or KIRAN at 1800-599-0019 (24/7).