Put two buckets on the floor of a long hallway, one close to a person’s starting point and one further along, both meant to be carried the rest of the way to the far end. Logic says pick whichever one means carrying less weight for less distance. In the original experiments, people didn’t do that. They grabbed whichever bucket was nearer, even when it was heavier, and even when that meant carrying more total weight over a longer stretch of the walk than if they’d waited a few extra steps for the lighter one.
The bucket experiments
The behaviour was named and formally documented by David Rosenbaum, Lanyun Gong and Cory Potts in a 2014 paper in Psychological Science, “Pre-Crastination: Hastening Subgoal Completion at the Expense of Extra Physical Effort.” Participants walked down a hallway toward two buckets set at different points along the route, one lighter, one heavier, and were told to pick up either one and carry it to the end. Repeatedly, participants chose the closer bucket, accepting more total physical effort in exchange for picking it up sooner and getting it over with.
The original paper reported this same pattern across nine separate experiments, and the researchers, drawing on both the behavioural results and participants’ own reports afterward, attributed the choice specifically to a desire to reduce the load on working memory, not to any of the more obvious alternative explanations.
Rosenbaum’s own explanation skipped past the two most obvious guesses, laziness and impatience. People were willing to spend more physical effort specifically to stop holding an unfinished task open in their head. . Carrying the decision, which bucket, when, for how long, apparently cost something on its own, and finishing the choice early relieved that cost even at a real physical price. Rosenbaum called the behaviour precrastination, the mirror image of procrastination: not delaying a task to avoid effort, but rushing to complete it to avoid the mental load of leaving it unresolved.
What later work added and complicated
Follow-up research has mostly reinforced the core finding while sharpening what’s actually driving it. A 2018 paper by Fournier and colleagues in Attention, Perception, & Psychophysics tested what they called the cognitive-load-reduction hypothesis directly, and found support for the idea that people are trading physical effort for a measurable reduction in mental effort, not simply acting on habit or impatience.
The behaviour isn’t obviously unique to humans, either. In a 2015 touchscreen task, Edward Wasserman and Stephen Brzykcy found pigeons showed a comparable pattern, pecking to complete an earlier step of a multi-part task sooner rather than later even without any extra reward for doing so, raising the possibility that the underlying tendency has some evolutionary depth to it rather than being a peculiarly human, culturally learned habit. Thomas Zentall, testing pigeons under a differently structured task, found birds instead procrastinating, delaying rather than rushing, and argued the result was better explained by delay-reduction and conditioned reinforcement, the value an animal places on a signal that a reward is imminent, than by anything resembling working-memory relief. Rather than settling the disagreement in a rival paper, Rosenbaum, Wasserman and Zentall ended up co-authoring a joint review together in 2019, in Current Directions in Psychological Science, laying out precrastination as a genuinely multi-causal phenomenon: partly about efficient foraging-style task selection, partly about the cognitive load of holding a decision open, with individual differences in how strongly either factor applies.
A more recent experiment brought the task closer to an actual household chore. Rosenbaum, working with Julian Dettling, had participants carry a set of balls between bins in multiple trips, deliberately designed to echo carrying groceries in from a car in stages, and published the results in 2023 in Psychological Research. Participants consistently carried more per trip early on rather than saving the heavier loads for later, the reverse of what a purely efficiency-minded plan might suggest, and asking participants to simultaneously hold a set of digits in memory increased how strongly they front-loaded their trips. That last detail lines up with the cognitive-load explanation specifically: making working memory more crowded made people even more eager to empty a task out of it early.
Why this looks familiar outside a lab
The original experiments were never about kitchens or mugs, but the behaviour they documented shows up constantly in ordinary domestic life, which is presumably why the finding keeps getting rediscovered outside academic psychology. Rosenbaum and Dettling’s own grocery-carrying framing makes the same point without needing a separate example: carrying one mug down at a time, rather than waiting for three or four, is the same trade-off their dodgeball-carrying participants made — more total trips, more total physical effort, in exchange for not having to keep “bring the mugs down” sitting open as an unfinished item in mind.
That’s worth sitting with for a moment, because it reframes what looks, from the outside, like poor planning. A person doing this isn’t failing to notice that four trips cost more than one. They’re noticing, correctly, that an open task has its own cost, one that doesn’t show up if you’re only counting steps or dishes, and deciding that cost is worth paying down immediately rather than carrying it a little longer for a more efficient outcome later. Rosenbaum’s buckets and a kitchen counter with one abandoned mug on it are, it turns out, the same finding wearing different clothes.
The researchers involved don’t frame this as a flaw to correct. Rosenbaum and his co-authors have suggested the same tendency that makes someone grab the nearer bucket, or the nearer mug, might also explain more consequential kinds of rushing: starting an email reply before fully reading the message, or beginning a task before gathering everything needed to do it properly, in both cases trading a bit of accuracy or efficiency for the relief of no longer holding the thing open. The mechanism that makes a kitchen slightly less efficient is, on this account, the same one occasionally responsible for a reply sent one read too soon.
