What sleeping on it actually does.

A night between a strong feeling and a choice gives the feeling time to fade and your first take time to loosen its grip. That is what helps. The evidence does not show that sleep itself solves the problem, makes you surer, or that your mind quietly works it out while you are away.
The heat fades
In an ultimatum game, one player proposes how to split money and the other accepts or rejects. Rejecting an unfair split means both get nothing, so it is mostly a reaction to feeling insulted. In one study, people who answered at once accepted unfair offers less than one time in five. Made to wait ten minutes, about seven in ten accepted (Grimm & Mengel, 2011).
A day’s wait has a weaker record when the matter is one of principle and the stakes are real. In a negotiation study, a cooling-off period cut rejections when the stakes were lottery tickets, but not when they were cash (Oechssler, Roider & Schmitz, 2015). Waiting takes out heat. It does not change what you care about.
A second look is a different look
Ask people the same question twice and average their two answers, and the average tends to beat either answer on its own. This “crowd within” has held up in several studies, including very large ones (Vul & Pashler, 2008; van Dolder & van den Assem, 2018).
The second answer helps more when it comes later, but that part is only partly repeated (Steegen et al., 2014). And asking yourself again is worth much less than asking another person. A night’s gap gives you a second opinion from a slightly different you. It is a modest gain, and a real one.
What it doesn’t do
The idea that your unconscious makes better choices while you are distracted came from a striking 2006 study. A large meta-analysis and replication found no such advantage (Nieuwenstein et al., 2015).
Sleep is not a reset button for feelings either. A meta-analysis found no effect of sleep on how strongly people react to emotional material (Lipinska et al., 2022), and the one direct test of sleeping on a choice found no gain in confidence (Karmarkar, Shiv & Spencer, 2017). Sleep does seem to help with spotting hidden patterns in some lab tasks. That is about new ideas, not about picking between options.
Wait before you commit, not after
Cooling-off works when it comes before the choice. Rights to cancel afterwards are rarely used (Sovern, 2014), because once you have chosen, the choice starts to feel like yours. So the order matters: put the decision down first, decide later, and don’t decide now with a plan to reconsider.
More questions
Will I wake up knowing what to do?
Probably not with any new certainty. The one direct test found no gain in confidence after sleeping on a choice. What you get is a calmer, less anchored second look.
How long should the wait be?
A night to a week. Lab effects appear within minutes to a day, real waiting periods run for days, and there is no known best length. Waits much longer than two weeks start to look like putting it off.
Sources
- Grimm V, Mengel F (2011). Let me sleep on it: delay reduces rejection rates in ultimatum games. Economics Letters 111(2).
- Oechssler J, Roider A, Schmitz PW (2015). Cooling off in negotiations. JITE 171(4).
- Vul E, Pashler H (2008). Measuring the crowd within. Psychological Science 19(7).
- van Dolder D, van den Assem MJ (2018). The wisdom of the inner crowd in three large natural experiments. Nature Human Behaviour 2.
- Steegen S et al. (2014). Measuring the crowd within again. Frontiers in Psychology.
- Nieuwenstein MR et al. (2015). Meta-analysis and large-scale replication of the unconscious thought advantage. Judgment and Decision Making 10(1).
- Lipinska G et al. (2022). Sleep and emotional reactivity, a meta-analysis. Frontiers in Behavioral Neuroscience.
- Karmarkar UR, Shiv B, Spencer RMC (2017). Should you sleep on it? Journal of Behavioral Decision Making 30.
- Sovern J (2014). Written notice of cooling-off periods. University of Pittsburgh Law Review.