I put 32 classic psychology experiments online. Here's what 1,400 choices revealed
November 5, 2026 · 5 min read
Mind Lab packs 32 famous psychology and economics experiments into one sitting. Players make a choice, then guess what everyone else chose. Nobody signs in and nothing personal is stored — just the answer.
I built it because these experiments are usually described in textbooks, not played. Reading that most people cooperate in the prisoner's dilemma is one thing; making the choice yourself, with a stranger's decision hidden from you, feels very different.
So far that adds up to roughly 1,400 choices from about 150 players. Here are five results that stood out. Never played one of these? Each section below starts with the set-up in two lines.
1,400
choices recorded
32
classic experiments
~150
anonymous players
1. Most people cooperate — even when betraying pays more
The set-up: you and another player each choose, without knowing what the other chose, whether to cooperate or betray. Betraying pays more as long as the other person cooperates.
Betraying is the mathematically safer move. Still, nearly two in three chose to cooperate.
That matches decades of lab research: people tend to give strangers the benefit of the doubt, at least the first time. Trust seems to be the default setting, and betrayal is something people learn after being burned.
2. Given $10 for free, a third kept every cent
The set-up: you are handed $10 and decide how much to give a stranger. They cannot refuse, argue or get even — your call is final.
The average gift was about $2.90 — almost exactly what lab studies have found since the 1980s. But the crowd split into two camps: keep everything, or split it down the middle.
Very few landed in between. It suggests that people don't calculate a "fair-ish" amount — they pick a rule ("it's mine" or "half each") and stick to it.
3. Over a third would rather lose money than accept unfairness
The set-up: one player proposes how to split a sum of money, the other accepts or rejects. Reject and both walk away with nothing.
A purely self-interested player would take any offer above zero. 37% said no anyway.
Rejecting an unfair offer costs you money, but it also punishes the person who made it. Many people are clearly willing to pay to make a point — a small, very human piece of evidence that fairness matters more than profit.
4. Monty Hall still fools almost everyone
The set-up: three doors, one car. You pick a door, then the host — who knows what is behind them — opens a losing door and asks whether you want to switch.
Switching wins two thirds of the time. Still, 69% stayed with their first pick. If you want to see why switching wins, try the Stay or Switch simulator.
5. Most players refuse to pull the lever
The set-up: a runaway tram is heading toward five people on the track. You can pull a lever to divert it onto a side track where one person is standing.
The surprise of the set. In classic surveys most people pull the lever to save five lives at the cost of one. In Mind Lab, 72% left the trolley alone. One possible reason: players had already seen the "obvious" answer so many times that they pushed back on it.
Another possibility is that the online setting changes things. With no researcher watching, choosing not to act may feel easier — doing nothing doesn't feel like a decision, even when it is one.
What it adds up to
Taken together, the results paint a fairly generous picture of people: most cooperate, many share, and a large group would rather lose money than accept being treated unfairly. At the same time, intuition is easy to fool — the Monty Hall result shows how hard it is to override a gut feeling, even when the maths is simple.
None of this is new to psychologists. What is interesting is how well the classic findings hold up when strangers play them for fun on their phones, years after the original studies.
How the data was collected
All numbers come from anonymous choices made in Mind Lab between October and November 2026. Players chose for themselves — there is no sampling or weighting, and each experiment has between 45 and 65 answers, so treat the exact percentages as a snapshot, not a scientific finding. The live numbers keep moving as more people play.