Is free will real, or does the brain decide before we do?

What did the famous brain experiments actually show?

The readiness potential, known before Libet’s work, is a slow electrical rise recorded with EEG (electroencephalography), sensors worn on the scalp. Averaged over many movements, it builds up slowly before movement begins.

In a 1983 study in the journal Brain, Benjamin Libet and colleagues added a twist. Five volunteers watched a spot circle a clock face and made a quick movement whenever they wished. Afterward, each person reported where the spot was when they first felt the wish to move. In the sets of trials where people said every movement felt spontaneous, the main rise began about 350 milliseconds, on average, before that reported moment. Libet concluded that the brain’s start of a freely chosen act “can begin unconsciously,” before any awareness that a decision has been made.

Libet himself saw limits to his conclusion. As a 2021 review in Trends in Cognitive Sciences explains, he suggested that people might still veto, or cancel, an act at the last moment, an idea often called “free won’t.” The review adds that later studies make this idea look untenable, because vetoes are also preceded by a readiness potential. It calls his second point more helpful: many actions come from slower, conscious thinking.

In 2008, a brain-scan study by Soon and colleagues reported predicting a decision up to 10 seconds before people were aware of it. The same review notes those predictions were only a little better than a random guess.

Why many researchers now read the signal differently

The catch: the readiness potential is an average built only from stretches of brain activity that ended in a movement.

In 2012, Aaron Schurger, Jacobo Sitt and Stanislas Dehaene built a simple model of this. As an analogy, picture waves at a beach, with a line drawn in the sand. Each time a wave happens to reach the line, the person moves. If you line up all those moments and average the waves before them, you get a smooth rise, even though no wave planned to reach the line. The analogy breaks in one way: in the model, the instruction to move adds a small, steady push, so it is not pure chance. The model matched the real signal and passed a new test: fast reactions to clicks at random times also came after a similar slow rise. In this account, the real neural decision to move now comes late, perhaps about 150 milliseconds before the movement, close to when people report the urge.

In 2019, Uri Maoz and colleagues tested choices that mattered. In some rounds, people picked which of two charities should get money, and a lottery meant each choice could lead to a real $1,020 donation. In other rounds, the key press made no difference. The readiness potential showed up before the meaningless presses but was “strikingly absent” before the meaningful ones.

The 2021 review sums up: if these newer models are right, the timing of the readiness potential cannot show that we lack conscious free will, and even if they are wrong, the signal still fails to support the old conclusion that conscious will does nothing. It also notes that a few studies support the older reading, and it lists which model is correct as an open question. The Stanford Encyclopedia of Philosophy says that after extensive study, neuroscientists are uncertain what the signal means. For more on the brain and awareness, see What Is Consciousness?

What do philosophers mean by free will?

Even perfect brain data would not settle the question, because people mean different things by free will. The Stanford Encyclopedia calls it a significant kind of control over one’s actions. The debate is over what that control requires. It starts with determinism: the idea that everything that happens results from the laws of nature and the state of the world long ago. Does that leave room for free will?

• Compatibilists say yes. In the classic version described in the encyclopedia’s entry on compatibilism, you are free when you can do what you want without being forced. Some newer versions say free actions are the ones that respond to good reasons.
• Libertarians (nothing to do with politics here) say no. For them, free will requires that determinism be false, so more than one future is truly open when you choose.
• Free will skeptics say we lack the kind of free will that would make us deeply deserve praise or blame. The older version, hard determinism, has few defenders today, the encyclopedia notes. A newer version, called hard incompatibilism and defended by philosophers such as Derk Pereboom, says we lack that kind of free will whether or not determinism is true.

In the 2020 PhilPapers Survey, 1,785 professional philosophers took part. Of those who answered this question, 59.2% accepted or leaned toward compatibilism, 18.8% toward libertarianism and 11.2% toward no free will. A vote does not make a view right, but it shows the question is far from closed.

Religious thinkers have wrestled with this for centuries. The Stanford Encyclopedia notes that much Western writing on free will was done within a theological framework. It points to thinkers such as Al-Ghazali, Aquinas, Scotus and, later, Jonathan Edwards in 1754. Believers, skeptics and the never-religious all have a place in this long conversation.

Does quantum randomness make room for free will?

Many people hope quantum physics, the physics of atoms and particles, can help. It usually gives only the odds of an outcome. If nature is not fixed in advance, maybe our choices are not either.

It weakens strict determinism, but does not refute it. The encyclopedia’s entry on skepticism about moral responsibility says quantum mechanics has led many to doubt that everything is determined, yet determinism has not been refuted. The entry on nondeterministic theories of free will adds that quantum theory is very well confirmed, but there is nothing approaching a consensus on how to interpret it.

It may not reach the brain. Libertarians need openness in the right places, at the moment of choice. On that, the same entry says, our current understanding of the brain gives “little evidence one way or the other.” Some argue that everyday brain activity runs on cause and effect even if atoms are chancy.

Chance is not control. Philosophers call this the problem of luck. If a choice comes down to chance, then whether it happened may not have been up to you. It was luck, as if nature rolled dice.

So quantum randomness offers, at most, an opening. Many compatibilists think it is not needed at all. For more on whether nature truly plays dice, see Is the Universe Truly Random?

What is the strongest objection?

One of the strongest challenges to free will does not come from brain scans. The Stanford Encyclopedia calls it the most radical argument of its kind: the philosopher Galen Strawson’s case, built on reasoning alone, that free will is impossible. You act the way you do because of how you are, mentally speaking. To be truly responsible for a choice, you would have to be responsible for how you are. But to choose how you are, you would already need a character and reasons to choose with, and so on forever.

The encyclopedia also notes evidence that our choices can be swayed by factors we are not aware of. The social psychologist Daniel Wegner argued that the experience of consciously willing is “an illusion.”

Defenders of free will reply. The philosopher Alfred Mele argues that freedom belongs mainly to our actions, and only secondarily to the characters they come from. Mele and Timothy O’Connor suggest that freedom comes in degrees and grows over time, as our own past choices shape who we are.

What do people often get wrong?

• “Libet proved free will is an illusion.” Libet timed movements made at arbitrary moments, with no reason to pick one moment over another. His reading has come in for severe criticism, and later models read the signal differently.
• “Scientists can read your choices 10 seconds early.” Those predictions were barely better than chance.
• “Random means free.” A coin flip inside your head would not make a choice more yours.
• “Without free will, nothing matters.” Spinoza denied free will, yet the encyclopedia notes he was keen to deny that this has the dire effects many assume.

How can you check this yourself?

• Ask what kind of choice was studied. Was it a random finger lift, or a decision with reasons and stakes? Maoz and colleagues found the two can look different in the brain.
• Ask what “predict” means. Better than chance is not the same as mind reading.
• Remember that curves are averages. The 2021 review warns that a smooth average may not match what happens in any single trial.
• Ask which free will is meant. A headline that says free will is dead usually means one definition.

The answer stays the same. The famous experiments do not show that your brain decides before you do, and whether free will is real is still an open question.