Volition, Libet, and the Conscious Veto

In Libet's original protocol, subjects watched a fast-moving clock and were asked to flex a finger 'whenever they felt like it,' then report the clock position at which they first became aware of the urge. EEG measured a readiness potential (RP) — a slow buildup over motor cortex — preceding movement. The headline finding: the RP began about 550ms before movement, while subjective awareness of the urge arrived only ~200ms before movement. The unconscious brain appeared to begin preparing the action a third of a second before 'you' decided.

Two qualifications matter and are routinely omitted. First, Libet himself argued for a 'conscious veto' window: even if initiation is upstream of awareness, the conscious channel still has ~150ms to abort the action before it executes. He explicitly used this to defend a form of free won't if not free will. Second, the RP is a noisy, statistical signal — averaged over many trials it looks decisive, but Schurger et al. (2012) reanalyzed it as essentially stochastic accumulation toward a threshold, with the apparent 'decision' being whenever noise happens to cross. On that reading the RP doesn't precede the decision; it is the decision, made by a process whose timing isn't sharply bracketed.

Soon et al. (2008) extended the lead time dramatically using fMRI: for binary button-press decisions, patterns in medial prefrontal cortex predicted the choice (above chance, ~60%) up to seven seconds before subjective awareness. This is often quoted as 'the brain decides seconds before you do,' but the actual prediction accuracy was only modestly above chance — it tells you the brain is biased toward an answer, not that the answer is fixed. Subsequent work (Bode et al. 2011) has shown the predictability disappears in genuinely high-stakes or novel decisions.

Recent reviews (Brass et al. 2019; Schurger, Mylopoulos & Rosenthal 2016) argue that the original Libet framing was probably the wrong frame. Decisions in real life are not arbitrary finger-flexes; they are extended, multi-cue, value-weighted processes. The question 'when does the conscious decision happen?' may not have a well-defined answer because decisions are not point events. Conscious deliberation, working memory, and metacognitive monitoring are real causal contributors — they just don't initiate from nowhere.

The practical implication aligns with Life Is Perfect's frame. Will is conditioned but not strictly determined. Most action initiates from substrate processes the conscious channel cannot see, but the meta-representational operator — sustained awareness — is itself a causal node that biases subsequent distributions. Freedom is not the absence of causes. It is the recursion of awareness over them.