What Survived: The Food Environment Research That Held Up After Cornell
KEY TAKEAWAYS
- A lot of popular kitchen advice traces back to a single Cornell lab whose director was found to have committed academic misconduct. By 2022, Retraction Watch counted 18 retracted papers, one of them retracted twice.
- Independent evidence that distance and the number of options change what people pick up does exist, but a Cochrane review rates it low certainty, and most of the trials were short and run in laboratories.
- A 2026 hallway experiment on 533 strangers found candy was taken more often when it sat on the walking route than when it sat farther along. That’s evidence from a real building, though it measured taking, not eating.
- None of it speaks to long-term habits, and this piece deliberately leaves out portion and plate-size research. What’s left is a layout variable worth testing in your own kitchen, with the people who share it in the room.
Somewhere in your kitchen there’s a rule you never chose. The snack cabinet sits up high. The fruit bowl lives by the window. Maybe the good bowls got pushed to the back because a book, a magazine or a friend said that what you can see is what you reach for. It came with a confident phrase, studies show, and almost nobody asked which studies.
A lot of that advice came from one lab. That lab no longer holds up.
This piece sorts what’s left: which kitchen findings have independent data behind them, how strong that data really is, and what a reasonable person does with evidence that’s real but thin. It skips plate sizes and portions on purpose, because this site doesn’t deal in them. What remains is one idea about distance, one about how many choices are within reach, and a short test for the next kitchen tip you meet.
The lab that shaped how we talk about kitchens
Brian Wansink ran the Food and Brand Lab at Cornell and wrote Mindless Eating, a book that made “your environment decides” a household phrase. In 2018, a Cornell investigation found problems in his statistical methods, and he announced his resignation. By that October, 15 of his papers had been retracted, according to the Cornell Daily Sun [1]. Retraction Watch’s 2022 count is 18, one of them retracted twice [2]. He has acknowledged mistakes and poor record-keeping and has denied fraud. He was out of Cornell by 2019.
None of that proves every idea his lab floated was wrong. A claim can be true while its first evidence is unusable, and that’s the uncomfortable part: you can’t tell which is which from the headline.
Retraction also works less dramatically than it sounds. A retracted paper doesn’t prove its claim false; it removes the paper as proof. The claim goes back into the pile of things that might be true and haven’t been shown. That’s a fair place for much of the lab’s kitchen advice to sit: not debunked, not established, waiting for someone with clean data.
So the useful question isn’t “did Wansink say it?” It’s “did anyone else find it, without him?”
The food environment research that someone else checked
The best answer to that question is a Cochrane review, the kind of study that gathers every comparable trial and grades how far to trust them. A team led by Gareth Hollands pooled 24 randomized trials, roughly three thousand participants, asking two things: does moving a food farther away change how much people pick up or eat, and does offering fewer options change what they choose [3].
The answers are cautious. Fewer options meant fewer picked, and food placed farther away meant less eaten in the short term, but the authors rate the certainty of both as low. For the effect of distance on what people chose, they rate it very low. Most of the trials ran in laboratories. Most were in high-income countries, mainly the United States. Most measured what happened in a single visit, not over weeks.
One more thing, and I’d treat it as something to check rather than settle. Reading the review’s list of included studies, I found papers by Wansink among the trials it pooled. As far as I could tell, the review doesn’t re-run its results without them, and its literature search appears to date from mid-2018, before most of the retractions. I haven’t been able to confirm either point from the full text, so read this paragraph as a flag, not a finding.
Which means the strongest summary of the distance idea may lean, in part, on the lab it was supposed to be independent of.
A hallway, a bowl of candy, and 533 strangers
The most recent independent test I found is from 2026. Psychologists Brian Meier and Adam Fetterman set a bowl of candy in a building hallway, either on the route people were already walking or farther along it, and had observers quietly record who took some [4]. Of 533 passersby, more took candy when it sat on their path than when it sat farther away.
That’s a real building, real strangers, and no one knew they were in a study. It’s also one hallway, one kind of snack, and a measure of taking rather than eating. Whether the candy got eaten, and whether any of it carries into a kitchen where the same person stands every day, is exactly what this design can’t say.
There’s another gap. Everyone in these studies is a stranger passing a bowl, rested enough to be walking down a hallway. Nobody is carrying groceries at 6 p.m., hungry and tired and being asked what’s for dinner. Real kitchens are loaded with context that experiments subtract on purpose, because subtracting it is how you isolate one thing. That’s a good way to run a study. It’s also why a clean result in a hallway should make you curious about your kitchen, not confident about it.
Two things can be true at once. Distance appears to matter a little in the moment. Nobody has shown what it does over a year.
What “held up” can honestly mean
Here’s a rough yardstick, mine and not a standard. A finding has held up if someone outside the original lab found it, if it showed up somewhere other than a lab, and if it lasted longer than a single sitting.
Distance and availability clear the first bar, with the caveat above. The hallway study clears most of the second. The third, nobody has cleared yet, at least in the sources I could read.
That’s one and a half out of three. It’s a lead, not a rule.
What to do with a finding that’s real but small
Treat distance as a layout variable, not a verdict. The research says nothing about which foods belong where, and nothing about whether a farther shelf is a good idea for you. If there’s something you’d like to reach for more often, closer is the idea on the table. If there’s something you’d like to reach for less often, farther is. That’s your list, not the study’s.
The sensible way to use weak evidence is to run it as an experiment on your own counter. Move one thing. Leave everything else alone. Notice, after a couple of weeks, whether the reaching changed. If it didn’t, you’ve learned something no study could tell you about your own kitchen.
Keep the moral weather out of it. Farther isn’t punishment and closer isn’t a reward. A cabinet is a cabinet.
And a kitchen is rarely one person’s. A rearranged shelf is a message to everyone who lives there, and some of them will notice before you do. If you share the space, ask before you move their things, and ask what they’d like closer. Otherwise the experiment ends up measuring how people feel about being rearranged.
Three questions for the next kitchen tip
The tip arrives the way they all do, in a headline or a caption, with no study attached. Ask three things before it earns a shelf.
Who ran it, and has anyone else? One lab’s result is a rumor with a methods section.
Where was it done? A university lab and your Tuesday evening are different rooms.
For how long? A finding about one snack in one visit is not a finding about a habit.
Most tips fail at least one. That isn’t a reason to ignore them, only to hold them loosely.
The hallway study is about a bowl and a path. Your kitchen has a person standing in it, and often a second one who never agreed to the new layout. That’s where the evidence runs out entirely, and where the more useful question starts.
References
- The Cornell Daily Sun. Food researcher Brian Wansink faces two more retractions. October 25, 2018. cornellsun.com
- Retraction Watch. Cornell food marketing researcher who retired after misconduct finding is publishing again. May 31, 2022. retractionwatch.com
- Hollands GJ, Carter P, Anwer S, King SE, Jebb SA, Ogilvie D, Shemilt I, Higgins JPT, Marteau TM. Altering the availability or proximity of food, alcohol, and tobacco products to change their selection and consumption. Cochrane Database Syst Rev. 2019;(9):CD012573. doi:10.1002/14651858.CD012573.pub3
- Meier BP, Fetterman A. Mind over distance: the food proximity effect on candy taking in a field setting and moderation by trait mindfulness in a hypothetical setting. Food Qual Prefer. 2026. cupola.gettysburg.edu/psyfac/136
Diana Woods researches and writes about sleep, environment and everyday habit. She has no clinical training, and that shapes what this site does: instead of explaining what might be wrong with you, she reads the behavioural and environmental research and writes about the parts you can act on — the light in the hallway, the fifteen feet to the bathroom, what the morning does to the night that follows.
Every article here cites its sources, and she tries to include the parts researchers themselves flag as uncertain. Studies disagree, sample sizes are small, and correlation gets reported as cause more often than it should. Saying so is part of the job.