Skip to content

Philosophy

14 min left
Philosophy · List

10 Thought Experiments Explained Simply: What Each One Tests and Where It Comes From

Ten imagined scenes that philosophers and scientists use to test an idea, in plain words.

Cheat sheet ↓

Hacks Vitae artwork (AI-assisted)
The answer

A thought experiment is an imagined situation used to test an idea.

You follow what would happen, instead of running a real experiment. Some ask what we can know, some ask what a mind is or what we value, and some reason about nature. Several, such as Mary's room and the Chinese room, are still argued over.

Why it mattersOur reading: a thought experiment is a careful question, not a finished answer.

Key facts
  1. 1638Galileo set out his falling-stones argument in Two New Sciences.
  2. 1641Descartes' Meditations, with the deceiving demon, was first published.
  3. 1688William Molyneux sent his question about a blind man's sight to John Locke.
  4. 1917Einstein's Relativity, with the chest pulled upward, was first published, in German.
  5. 1980John Searle first published the Chinese room argument.
  6. 1982Frank Jackson set out the case of Mary, who has been kept in a black-and-white room.

By Hacks Vitae Published Last checked November 2026 14 min read 17 sources ↓

Start hereImagine you had only ever seen shadows on a wall. Would you take them for the real world? Plato asked that question in the Republic, and people still use scenes like it to test ideas.

A thought experiment is an imagined situation used to test an idea: you follow what would happen, instead of running a real experiment. The Stanford Encyclopedia of Philosophy calls thought experiments “devices of the imagination”.

Below are ten famous ones in four groups. Each card says what the scene is, what the sources say about it, and how settled the question is. Five more, including the ship of Theseus and the trolley problem, already have pages of their own, linked near the end.

What a thought experiment is

A thought experiment asks what would follow if a scene were true, and people still discuss how much thinking alone can teach.

The encyclopedia says thought experiments suggest we can learn about the real world “by virtue of merely thinking about imagined scenarios”. It then asks how we can learn about reality “(if we can at all), just by thinking”.

Ernst Mach is commonly credited with introducing the term. Its history goes back at least to the Danish word “Tankeexperiment”, as Hans-Christian Ørsted used it, and the encyclopedia quotes a scholar who found the word in use by 1811.

The historian of physics Pierre Duhem is “almost alone”, the encyclopedia says, in what has been understood as an outright condemnation of scientific thought experiments. In his view, a thought experiment is “no substitute for a real experiment”.

Our reading: a thought experiment is a careful question, not a finished answer.

For example, asking “what would I do if I won the lottery?” can show what you value, without buying a ticket.

Doubting what you see

Three of these scenes ask whether the world you see is the whole truth, or real at all.

1. Plato’s cave

Prisoners who have only ever seen shadows on a wall would take the shadows for the real world.

In Book VII of Plato’s Republic, Socrates tells Glaucon an allegory about people living in a cave. In Benjamin Jowett’s translation, they have been there “from their childhood”, chained so that they can only see in front of them. For them, Socrates says, “the truth would be literally nothing but the shadows of the images”.

If one of them went up into the light and came back, the others would say that “up he went and down he came without his eyes”. Socrates suggests they would even put to death anyone who tried to lead them up to the light.

He then gives his own reading, offered, he says, as “my poor belief”: the prison-house is “the world of sight”, and the climb out is “the ascent of the soul into the intellectual world”. Our page on quotes from Socrates, Plato and Aristotle looks at how the cave is quoted today.

For example, if you knew a city only from photos, you might take the photos for the place until you walked its streets.

How sure: Strong the scene and Socrates’ own reading of it are in the text, in Jowett’s translation.

2. Descartes’ deceiving demon

What if a powerful deceiver were feeding you a false world, so that even your body was an illusion?

René Descartes set this out in his Meditations, first published in 1641 with six sets of objections and replies. In John Veitch’s translation, he first notes that there are “no certain marks” to tell waking from sleep.

Then he imagines “some malignant demon, who is at once exceedingly potent and deceitful”. He even resolves to think of himself as “without hands, eyes, flesh, blood, or any of the senses”.

The doubt sweeps away almost everything, but one thing is left standing. In the second Meditation he finds that “I am, I exist” must be true each time he says it or thinks it.

The encyclopedia says the demon story itself is not what matters: the point is realising that, “for all we know, our cognitive wiring is flawed”.

For example, a vivid dream can feel completely real until you wake up.

How sure: Strong the demon and the “I am, I exist” step are in the text, in Veitch’s translation.

3. The brain in a vat

Could you be a brain kept alive in a vat, with a computer feeding you every experience?

The encyclopedia says the sceptical scenarios most widely used today are versions of this: a “bodiless brain in a vat” stimulated by “a super-computer”.

In 1981 Hilary Putnam argued that the idea that we are, and always have been, brains in vats is “self-refuting”. His reason turns on what our words can refer to. On his view, if we had always been brains in vats, our sentence “we are brains in a vat” would mean something like “brains in a vat in the image”.

Thomas Nagel replied that, even if the argument works, it raises a new worry: a brain in a vat still could not truly think that it is one. The encyclopedia says Putnam’s argument started “a stream of illuminating literature” on how to rebuild and judge it.

For example, a very detailed video game can feel like a real place while you play it.

How sure: Unsettled Putnam’s argument started a stream of work on how to rebuild and judge it, and replies like Nagel’s are still part of that discussion.

What a mind is

Three scenes test what a mind is: whether knowing every fact is enough, whether following rules is understanding, and whether the senses share what they know.

4. Mary’s room

A scientist who knows every physical fact about colour, but has only ever seen black and white, sees colour for the first time: does she learn something new?

Frank Jackson set out the case in 1982. In Jackson’s own description, which the encyclopedia quotes, Mary has had to study the world “from a black and white room via a black and white television monitor”. She has “all the physical information there is to obtain” about what happens when we see ripe tomatoes.

Jackson states the intuition this way: “It seems just obvious that she will learn something about the world and our visual experience of it”. If she does, the argument goes, then not all information is physical information. That would be a problem for physicalism, the view that everything is physical. The encyclopedia calls Mary’s room “one of the most discussed arguments against physicalism”.

Jackson himself later changed his mind, and in 1995 argued that something must be wrong with the argument. One reply says Mary gains no new fact, only “a bundle of abilities”, such as being able to imagine, remember and recognise colours.

For example, reading every review of a dish is not the same as tasting it.

How sure: Unsettled it is one of the most discussed arguments against physicalism, and even Jackson changed his mind about it.

5. The Chinese room

Someone following a rule book to answer messages in Chinese, without knowing any Chinese, could look fluent from outside.

The American philosopher John Searle first published the argument in a 1980 article. He imagines himself alone in a room, following a computer program for responding to Chinese characters slipped under the door. People outside would wrongly suppose there is a Chinese speaker in the room.

His narrow conclusion is that programming a computer may make it appear to understand language but could not produce real understanding. One reply, called the Systems Reply, says that while the man does not understand Chinese, the system as a whole does.

Since AI systems based on large language models drew wide attention in late 2022, the encyclopedia asks whether they make any difference to the argument. Our page on how large language models work explains what those systems do, in plain words.

For example, a traveller reading answers from a phrasebook can reply correctly without understanding the words.

How sure: Unsettled the encyclopedia says there is still no consensus on whether the argument is sound.

6. Molyneux’s problem

If a person born blind could suddenly see, could they tell a cube from a sphere by sight alone?

William Molyneux sent the question to John Locke in a letter on 7 July 1688. For reasons unknown, Locke never replied to that letter. Molyneux asked again in a letter dated 2 March 1693, and Locke quoted it almost word for word in the second edition of his Essay, in 1694. Molyneux’s wife had lost her sight in the first year of their marriage.

Molyneux answered no, and Locke agreed: the blind man “at first sight, would not be able with certainty to say which was the globe, which the cube”.

Surgeons had reported on newly sighted patients since 1728, but those early reports were anecdotal and hard to compare. In 2011, a study of people born blind whose sight was restored found “a lack of immediate transfer”, but the link between touch and sight “developed rapidly”. The encyclopedia describes the study: five people born blind had their vision restored by cataract surgery. All five failed the touch-to-vision test at first, barely better than guessing. Three of them succeeded when tested after a few days of “natural real-world experience”. The encyclopedia adds that critics often question how robust the new sight of these patients was.

Even so, the encyclopedia calls it an open question whether Molyneux’s original question would also be answered “no”.

For example, you might know a tune by ear and still not recognise it written down as notes.

How sure: Moderate the study tested five people, and the encyclopedia says the original question is still open.

How to live

Two scenes ask what we value: being good when no one can see, and real life over pleasant feelings.

7. The ring of Gyges

If a ring could make you invisible, would you still act fairly?

In Book II of Plato’s Republic, Glaucon tells of Gyges, “a shepherd in the service of the king of Lydia”. When he turned the ring, “instantly he became invisible to the rest of the company”. He then “seduced the queen, and with her help conspired against the king and slew him, and took the kingdom.”

Glaucon argues that no one would be of “such an iron nature” as to stay just with such a ring. He sums up the view: “wherever any one thinks that he can safely be unjust, there he is unjust.”

But Glaucon is putting the common view, not his own. He says so plainly: “since I myself am not of their opinion”. What he wants is to hear justice “praised in respect of itself”.

For example, would you tidy a shared kitchen if no one could ever know who did it?

How sure: Strong the story and Glaucon’s challenge are in the text, in Jowett’s translation.

8. The experience machine

Would you plug into a machine that gave you any experiences you wanted, for the rest of your life?

Robert Nozick set it out in Anarchy, State, and Utopia, published in 1974. The encyclopedia puts the scene this way: “Imagine I have a machine that I could plug you into for the rest of your life.” You “would not know you were on the machine”.

The encyclopedia says Nozick thinks it would be a big mistake to plug in. His reason, in his words, is that “something matters to us in addition to experience”. We want to do certain things, he says, not just have the experience of doing them.

The scene is aimed at hedonism, the view that only pleasure has worth or value. The encyclopedia calls it a weighty objection to hedonism, and to the view that well-being consists only in conscious states.

It also says that quickly dismissing hedonism on the strength of the machine “is not methodologically sound”. One reply open to hedonists is to accept that the objection has force, but to say it rests on “common sense” intuitions.

For example, watching a film of a mountain climb is not the same as climbing it, however good the film.

How sure: Unsettled the encyclopedia calls it a weighty objection, but warns against dismissing hedonism quickly on its strength.

Testing nature in your head

Scientists use thought experiments too, and two famous ones changed how people thought about falling and gravity.

9. Galileo’s falling stones

If heavy things fell faster, a heavy stone tied to a light one would have to fall both faster and slower.

Galileo set out the argument in Two New Sciences, published in Holland in 1638. In Henry Crew and Alfonso de Salvio’s 1914 translation, the character Salviati says the faster body “will be partly retarded by the slower”.

Tied together, the pair is heavier than the heavy stone alone, so it should fall faster. Yet the light stone should also hold it back, so it should fall slower too. The encyclopedia sums it up: the joined body “would have to fall both faster and slower” than the heavy one alone.

Salviati concludes that “even without further experiment” a short argument shows that a heavier body does not fall faster than a lighter one of the same material. In the same scene, another speaker, Sagredo, says he has “made the test” with a cannon ball and a musket ball. The encyclopedia notes that Galileo’s texts are “full of illustrative scenarios he does not pretend to have carried out”.

For example, tie a light ball to a heavy one in your mind and ask which way the light one should pull.

How sure: Strong the argument is in Galileo’s own text, in the 1914 English translation.

10. Einstein’s chest

A person inside a closed box pulled upward with a constant force, so that it keeps speeding up, would feel just as if gravity were holding them down.

Albert Einstein described it in Relativity, first published in German in 1917; we read Robert W. Lawson’s authorised translation. He asks us to “imagine a spacious chest resembling a room with an observer inside who is equipped with apparatus.”

Something then pulls the chest with a constant force, so that it moves “upwards” with a uniformly accelerated motion. The man inside concludes that he is in a gravitational field that does not change over time. Einstein asks: “Ought we to smile at the man and say that he errs in his conclusion?” He answers that we should not, if we wish to remain consistent.

So from inside the box, being pulled and being held by gravity feel the same. A footnote in this English edition says the bending of light the theory predicted was “first confirmed” during the solar eclipse of 29 May 1919. The encyclopedia calls the scene “Einstein’s elevator”, listing it with Galileo’s falling bodies and Newton’s bucket.

For example, in a lift that sets off upward, you feel heavier for a moment.

How sure: Strong the scene is in Einstein’s own book, in Lawson’s authorised translation.

Already on our site

Five more thought experiments have their own pages on our site, so here they get one line each.

  • The ship of Theseus. Plutarch describes an old debate about a ship whose timbers were replaced, and the second ship, built from the old planks, is Thomas Hobbes’s addition. See our page on mental models from philosophy.
  • The trolley problem. Philippa Foot wrote the case it descends from in 1967, and in her version it was a runaway tram. See our page on where the trolley problem comes from.
  • Buridan’s ass. A donkey is stuck between two equal piles of hay, but, in the encyclopedia’s words, the example is “nowhere to be found in Buridan’s writings”. See mental models from philosophy.
  • The veil of ignorance. In John Rawls’s version, the principles of justice are chosen behind a “veil of ignorance”, and John Harsanyi discussed the problem before Rawls. See mental models from philosophy.
  • The dragon in the garage. Chapter 10 of Carl Sagan’s The Demon-Haunted World opens with a thought experiment: “A fire-breathing dragon lives in my garage.” See our summary of the book.

How to use one yourself

Our suggestion: you can run a small thought experiment on any idea you feel sure of.

  1. Picture one clear case, with as few details as you can.
  2. Change one detail at a time, and ask what would follow.
  3. Check the result against the real world when you can, the way researchers in 2011 tested Molyneux’s question with people whose sight had been restored.

For example, if you are sure a new rule at work is fair, imagine you were the person it affects most, then change who that person is.

Where to learn more

Quick answers

What is a thought experiment in simple terms?

An imagined situation used to test an idea. You follow what would happen, instead of running a real experiment.

Who came up with the term "thought experiment"?

Ernst Mach is commonly credited with introducing it, and its history goes back at least to Hans-Christian Ørsted's Danish word "Tankeexperiment".

What is the difference between the brain in a vat and Descartes' demon?

Both ask how you could know the world you see is real. Descartes imagined a powerful deceiver; the version most used today imagines a bodiless brain stimulated by a super-computer.

Do scientists use thought experiments?

Yes. Galileo's falling stones and Einstein's chest are two famous examples. The historian Pierre Duhem is almost alone in what has been understood as an outright condemnation of them.

Use this

The lesson

Our suggestion: run a small thought experiment the next time you feel sure of something.

The rule

What would follow if this were true?

Quick quiz

Three questions, thirty seconds

Nothing is scored or saved. Recalling it once now is what makes it last.

1. In Plato's cave, what do the prisoners take for the truth?

Answer: The shadows of the images on the wall.Read it again ↑

2. What did the 2011 study find when people born blind first had their sight restored?

Answer: They could not match them at first, but learned to quickly.Read it again ↑

3. Why does Glaucon tell the story of the ring of Gyges?

Answer: To hear Socrates answer the common view, which he says is not his own.Read it again ↑

How we checked this · Last checked November 2026
  • We read the Stanford Encyclopedia of Philosophy entry on each thought experiment and, where the scene comes from an old book, the book itself in a named translation (Jowett for Plato, Veitch for Descartes, Crew and de Salvio for Galileo, Lawson for Einstein).
  • Nozick's book is not freely readable, so the experience machine is given in the encyclopedia's words, and his own phrases are the ones the encyclopedia quotes.
  • The ship of Theseus, the trolley problem, Buridan's ass, the veil of ignorance and the dragon in the garage already have pages on our site, so they get one line each here.

How we check every article →

Cheat sheet

All 10 at a glance

Each idea in a few words, an everyday example, and one thing to try. Tap a name for the full section.

#IdeaIn shortExampleTry this
Doubting what you see
01Plato’s caveStart herePeople who only ever see shadows would take the shadows for reality.Knowing a city only from photos, and taking the photos for the place.Next time you feel sure about something, ask what you have only seen second-hand.
02Descartes’ deceiving demonIf everything could be a deception, "I am, I exist" still holds.A vivid dream that feels completely real until you wake up.Next time you doubt everything, look for the one thing you cannot doubt.
03The brain in a vatCould you be a brain fed experiences by a computer? Still argued.A detailed video game that feels like a real place while you play.Next time a "what if it's all fake" idea comes up, ask what your words would even mean.
What a mind is
04Mary’s roomStart hereKnowing every fact about colour may not be the same as seeing it.Reading every review of a dish is not the same as tasting it.Next time you think you know something from reading alone, ask what doing it would add.
05The Chinese roomStart hereFollowing rules well may not mean understanding. No agreed answer.A traveller reading answers from a phrasebook.Next time an answer sounds fluent, ask whether it shows understanding or just good rules.
06Molyneux’s problemOne study's newly sighted failed at first to match sight to touch, then improved within days.Knowing a tune by ear but not recognising it written as notes.Next time a skill does not transfer straight away, give it a few days of practice.
How to live
07The ring of GygesStart hereWould you act fairly if no one could see you?Tidying a shared kitchen when no one will know who did it.Next time you choose something in private, ask if you would choose it in public.
08The experience machineNozick says we want to do things, not just feel we did them. Still argued.Watching a film of a mountain climb versus climbing it.Next time you pick a shortcut to a good feeling, ask what you would miss by not doing it.
Testing nature in your head
09Galileo’s falling stonesA heavy and light stone tied together can't fall both faster and slower.Tying a light ball to a heavy one and asking which way it pulls.Next time a rule seems obvious, combine two cases and see if it contradicts itself.
10Einstein’s chestFrom inside a closed box, being pulled and gravity feel the same.Feeling heavier for a moment as a lift sets off upward.Next time two situations feel identical, ask what test could tell them apart.

Sources

Primary written at the time, by a witness, or the original text Scholarly edited or peer-reviewed work Popular press, blogs and general websites
  • J. R. Brown and Y. Fehige, “Thought Experiments”, Stanford Encyclopedia of Philosophy (substantive revision 28 November 2023).Scholarly
  • Plato, Republic, Books II and VII, translated by Benjamin Jowett, Project Gutenberg eBook #1497.Primary
  • René Descartes, Meditations on First Philosophy, Meditation I and Meditation II, translated by John Veitch, Wikisource.Primary
  • L. Newman, “Descartes’ Epistemology”, Stanford Encyclopedia of Philosophy (substantive revision 27 November 2023).Scholarly
  • G. Ebbs, “Skepticism and Content Externalism”, Stanford Encyclopedia of Philosophy (first published 18 March 2025).Scholarly
  • M. Nida-Rümelin and D. O Conaill, “Qualia: The Knowledge Argument”, Stanford Encyclopedia of Philosophy (substantive revision 1 March 2024).Scholarly
  • D. Stoljar, “Physicalism”, Stanford Encyclopedia of Philosophy (substantive revision 16 September 2026).Scholarly
  • D. Cole, “The Chinese Room Argument”, Stanford Encyclopedia of Philosophy (substantive revision 23 October 2024).Scholarly
  • B. Glenney and G. Ferretti, “Molyneux’s Problem”, Stanford Encyclopedia of Philosophy (substantive revision 8 January 2026).Scholarly
  • John Locke, An Essay Concerning Human Understanding, Book II, chapter IX, section 8, Project Gutenberg eBook #10615.Primary
  • R. Held, Y. Ostrovsky, B. de Gelder, T. Gandhi, S. Ganesh, U. Mathur and P. Sinha, “The newly sighted fail to match seen with felt”, Nature Neuroscience 14(5) (2011), 551–553, abstract on PubMed.Scholarly
  • A. Moore, “Hedonism”, Stanford Encyclopedia of Philosophy (substantive revision 17 October 2013).Scholarly
  • R. Crisp, “Well-Being”, Stanford Encyclopedia of Philosophy (substantive revision 13 January 2026).Scholarly
  • D. M. Miller, “Galileo Galilei”, Stanford Encyclopedia of Philosophy (substantive revision 17 December 2025).Scholarly
  • Galileo Galilei, Dialogues Concerning Two New Sciences, First Day, translated by Henry Crew and Alfonso de Salvio (Macmillan, 1914), Internet Archive.Primary
  • Albert Einstein, Über die spezielle und die allgemeine Relativitätstheorie (Braunschweig: Vieweg, 1917), the German first edition, Project Gutenberg eBook #77850.Primary
  • Albert Einstein, Relativity: The Special and General Theory, translated by Robert W. Lawson, Project Gutenberg eBook #30155.Primary