The Bag Decides When You're Done. Not Your Stomach. Not Your Willpower.

You didn't finish it because you wanted it. You finished it because it was the whole unit. Open a bag of chips telling yourself you'll have a few, and somehow you're staring at an empty bag twenty minutes later, genuinely unsure where the rest went. You didn't choose to eat all of it. You were just finishing what someone else, a long time before you ever opened the bag, decided counted as one.
The Study That Found Your Stomach Isn't Actually in Charge
In 2006, researchers Andrew Geier, Paul Rozin, and Gyorgy Doros published a paper in Psychological Science called "Unit bias: A new heuristic that helps explain the effect of portion size on food intake," and the finding was stranger than the polite academic title suggests. People don't stop eating when they're full. They stop when they reach the end of whatever was handed to them as a single unit — a plate, a bag, a scoop, a slice. In their experiments, participants offered M&Ms, Tootsie Rolls, and pretzels with a larger serving spoon or container consistently ate more, without any corresponding increase in reported hunger or fullness. The container size, not the body's actual satiety signal, was setting the stopping point. Your stomach was never running that calculation. The unit was.
This matters because it demolishes a very comfortable, very common piece of self-blame: the idea that finishing the whole bag means you lack discipline. Discipline was never the variable being tested. The size of what got handed to you was.
Restaurants Figured This Out Decades Ago. Streaming Figured It Out Faster.
Once you know the mechanism, you start seeing who's been quietly exploiting it. A bigger plate doesn't just physically hold more food — it recalibrates what your brain treats as a normal amount, so you clean it, reliably, every time, regardless of appetite. Restaurant portion sizes in the US have grown substantially over the past several decades, and the growth wasn't accidental. Bigger plates sell more food and generate more perceived value per dollar, and unit bias guarantees the customer will eat closer to the plate's capacity than to their body's actual need.
Streaming platforms found the exact same lever, just in a different currency: time instead of food. A season isn't ten individual episodes to your brain. It's one unit. So you don't stop at episode one, evaluate how you feel, and decide whether to continue — you stop when the unit ends, which might be four hours later, at 1am, having made zero individual stopping decisions along the way, because the platform never gave you a natural place to make one.
This isn't speculation. A 2025 experimental study out of the University of Chicago, published through ACM CHI Proceedings, tested this directly with 76 real Netflix users, half with autoplay disabled. Disabling autoplay cut daily viewing by 21 minutes and shortened individual sessions by 17 to 18 minutes. Participants who had to manually choose to continue took measurably longer breaks between episodes — meaning they were, for the first time in the session, actually making a decision instead of defaulting to whatever the platform had pre-selected as the unit. The feature isn't neutral. It's engineered to keep you inside a unit you never chose the size of.
This is the same architecture behind why your phone rewired your attention span in the first place — a feed with no natural stopping point is unit bias with the unit deleted entirely.
Infinite Scroll Is Unit Bias With the Floor Removed
Food and streaming both still technically have a unit, even if it's an oversized one — a bag, a season, a plate. Social feeds took the next logical step and removed the unit altogether. Research presented at CHI 2025 ("Scrolling in the Deep") and a 2025 paper on scroll-interruption interventions both document that infinite scroll interfaces exploit unit bias's underlying mechanism even without a defined unit: users perceive an undefined scroll session as something that should have a natural completion point, and the absence of that point — rather than eliminating the compulsion to find one — makes the compulsion to keep going worse, because the brain keeps searching for a stopping cue that was deliberately engineered not to exist. A 7-day user study on scroll interventions found something particularly bleak: contextual factors like being at home or in a low mood measurably reduced how well any intervention worked, meaning the platforms' design advantage compounds exactly when your own resistance is already lowest.
None of this is a coincidence discovered after the fact. It's the business model, built on a heuristic identified in a food-psychology lab in 2006 and ported, almost unchanged, into every attention product you use.
The Turn: You Don't Fight the Bag. You Redraw It First.
The instinct here is to treat this as one more thing to feel bad about — one more proof that you have no self-control. That's backwards. Unit bias isn't a character flaw being exposed. It's a heuristic every human brain runs by default, discovered in a lab, then reverse-engineered by an entire industry that profits every time you let someone else's container decide your stopping point instead of deciding it yourself.
The fix isn't willpower deployed at the moment of temptation — that's the losing move, because by the time the bag or the autoplay countdown is in front of you, the unit has already been set by someone else, and you're negotiating from the wrong position. The fix is redrawing the unit before you start, while you still have the leverage. Decant the snack into a small bowl before you sit down, so the unit you're actually looking at is the one you chose, not the one the manufacturer chose to maximize shelf appeal. Pick the number of episodes before you press play, not after episode one ends and the countdown starts doing the deciding for you. Decide what "one" means, on your terms, before someone else's default gets to decide it on theirs.
You're not weak-willed, and your brain isn't broken. It's following a rule set by the size of the box, not the size of your hunger — a rule that was documented, studied, and then sold back to you as a UI feature. Once you see who actually drew that line, you get to pick up the pen and redraw it yourself. That's the whole difference between finishing a bag because it was handed to you that size, and finishing it because you actually decided that's what you wanted.
Cover photo by Srattha Nualsate via Pexels.
Follow The Shadow Self Media on TikTok · Instagram · Facebook