Punching a pillow was supposed to drain the anger out, but the experiments found venting rehearses rage instead, and doing nothing at all cooled people down faster

The advice is everywhere, and it sounds like common sense. You’re furious about something, so you go find a pillow and hit it until your arms give out. Get it out of your system. Blow off steam. Better the pillow than the person.

It shows up in movies, in self-help books, on billboards telling parents to hit a wall instead of a child. It has been repeated so often that almost nobody stops to ask whether it actually works.

In 2002, a psychologist decided to find out. He built an experiment designed to make six hundred people genuinely, specifically angry, then handed a third of them boxing gloves and told them to swing.

The results ran in the wrong direction. And the group that walked out calmest was the one that had been left sitting in a room with nothing to do at all.

Six hundred students were insulted on purpose, then handed a pair of boxing gloves

Brad Bushman, then at Iowa State University, told his participants he was studying first impressions. Each one wrote a short essay on abortion, arguing whichever side they actually believed.

The essay was taken away to be evaluated by another participant, but there was no other participant.

What came back was withering. Negative marks on organization, originality, clarity, persuasiveness, overall quality. At the bottom, handwritten, a line telling them it was one of the worst essays the reader had ever gotten through.

That procedure works. People get angry.

Then they were split three ways. One group pulled on boxing gloves and hit a 70-pound punching bag while a photo of the student who had supposedly savaged their essay sat on a monitor in front of them. They were told to think about that person while they swung.

A second group hit the same bag, but they were told to think about getting physically fit, and the photo on the screen was an athlete from a fitness magazine.

The third group hit nothing. They sat quietly for two minutes while the experimenter pretended to be fixing a computer down the hall. No breathing exercise, no pep talk, no advice of any kind.

Afterward, everyone rated their mood on a checklist and then got the chance to blast their critic with noise through a pair of headphones, choosing both the volume and how long it lasted.

The group that did nothing at all came out the least angry

Catharsis theory makes a clean prediction here. The people who swung at the bag while picturing the person who insulted them should have purged the anger and come out the calmest of the three.

They came out the angriest. They were also the most aggressive when handed the noise controls. The people who had sat in a room doing nothing scored lowest on both.

Bushman titled the paper Does venting anger feed or extinguish the flame?, and his own one-line summary of the finding was that doing nothing at all beat venting.

Two things deserve saying plainly. The effects were small to moderate rather than dramatic. And the distraction group, the ones hitting the bag while thinking about fitness, landed in the middle without separating cleanly from the control group on either measure.

But the direction is what matters, and the direction was the opposite of what everyone expects. The control group got no help whatsoever. They sat there for two minutes, and sitting there beat swinging.

One detail from the paper tends to stick. A man in the rumination group got so worked up hitting the bag that he put a hole in the laboratory wall.

Venting is rehearsal, and the body gets good at whatever it rehearses

The explanation Bushman offers comes from cognitive neoassociation theory, and once you hear it, the whole pillow idea starts to look strange.

Aggressive thoughts, aggressive feelings, and aggressive impulses are linked together in memory. Activate one and the activation spreads along the connections and wakes up the others.

So think about what venting actually is. You picture someone’s face. You imagine hurting them. You perform the motion of hurting them, over and over, with your whole body, for several minutes.

That is not draining anything. That is practice.

The hydraulic metaphor everyone carries around, anger as steam building in a sealed container until it blows, is doing a lot of damage here. There is no container. Anger is a state your nervous system enters, and once you’re in it, you can be pulled deeper.

Every rep you do while furious teaches the same lesson: when I feel this, I do this, and then I feel better for a minute. That association gets stronger with repetition, and the next provocation will not be a punching bag in a lab. It will be your manager, or your kid, or the driver who cut you off.

Ranting into a keyboard follows the same pattern as swinging at a bag

Most people are not punching anything. The everyday version of venting is verbal, and it turns out the pattern survives the translation.

Ryan Martin, a psychologist at the University of Wisconsin-Green Bay, studied the users of sites built specifically for ranting. His team found that reading other people’s rants pushed mood downward, and that writing your own for five minutes left people less happy and more angry than they started.

The survey side of that work turned up something bleaker. People who used rant sites reported averaging nearly one physical fight a month and more than two verbal ones.

That last figure is correlational, and Martin has said as much. It is unclear whether ranting produces the fights or whether people already fighting are drawn to rant sites. He suspects the two feed each other.

Martin is careful not to treat anger itself as the villain. Emotions, he told PBS NewsHour, are the body’s way of telling us things, and anger is the one that tells you something is wrong. His objection is to the method, not the feeling.

Rage rooms sit in the same category. You pay to smash plates and old televisions with a bat, and the rush is real, because intense exertion produces it. Scott Bea, a clinical psychologist at the Cleveland Clinic, has described that discharge as probably short-lived, and noted that for someone with a real anger problem, it mostly reinforces the coping style that was already causing trouble.

Part of what keeps all of this going is that people have been told it works. In a 1999 study, participants who read a message endorsing catharsis wanted to hit a punching bag more than those who read a message against it, and when they actually hit it, they were subsequently more aggressive.

So the belief had exactly one measurable effect. It made people more willing to do the thing that backfired on them.

A review of 154 studies found the fix was lowering the heat, not raising it

The punching bag experiment was one study with modest effects, which is a fair thing to hold against it. So in 2024, Sophie Kjærvik and Bushman went looking at everything.

Their meta-analytic review pulled together 154 studies and 10,189 participants, published and unpublished, across ages, genders, and cultures. They sorted anger interventions into two piles: the ones that raise physiological arousal, and the ones that lower it.

Arousal-decreasing activities worked, and they worked consistently. Deep breathing, progressive muscle relaxation, mindfulness, meditation, slow flow yoga, and simply taking a timeout all reduced anger and aggression, in labs and out in the world, with college students and with people who had criminal records.

Arousal-increasing activities did not. As a category they had no effect on anger at all.

The surprise was jogging, which came out as the activity most likely to make anger worse. Bushman put it bluntly in the university’s write-up of the findings, saying there is not a shred of scientific evidence supporting catharsis theory.

There was one wrinkle the researchers found genuinely interesting. Physical education classes and ball sports actually had a calming effect, which suggested to them that adding play to physical activity changes what it does to you.

Which fits the rest of the picture. What seems to decide the outcome is where your attention sits while your body works, and a game points it somewhere other than the grievance.

The anger is still telling you something worth hearing

None of this is an argument for swallowing it. The research tested one narrow claim, that acting out anger discharges it, and found that claim false. Anger itself came through the review intact, still doing the job it has always done, which is to tell you a line got crossed.

What the evidence keeps pointing at is sequence. Bring the arousal down first. Then deal with the thing.

An angry brain is a bad negotiator, a bad problem solver, and a bad judge of what someone actually meant. Two minutes of nothing, or twenty slow breaths, or a walk around the block with no rehearsing of the argument, buys you back the part of yourself that can handle the conversation.

Some people never get the chance to make that choice, because their anger arrives in disguise. It comes out sideways, as tears in the middle of an argument, routed there by a nervous system that learned early which feeling was safe to show.

But for everyone who has ever been handed a pillow and told to let it out, the research offers something oddly freeing. You do not have to perform your rage to be done with it.

You are allowed to just sit down for two minutes. In the one experiment that tested it directly, that was the group that came out ahead.