Libet's Mind Time: Readiness Potential, Unconscious Initiation, Delayed Awareness, Antedating Consciousness

Neurophysiologist Benjamin Libet's experimental theory of consciousness holds that phenomenal experience is real but slow-forming: a substantial minimum duration of neuronal activation, up to roughly half a second, is required before any experience arises at all, and its apparent immediacy is manufactured after the fact by an automatic backward referral of subjective timing to an early cortical marker. Applied to volition, the same temporal method yielded his most cited result—that the readiness potential precedes the reported onset of conscious intention—from which he concluded that voluntary acts are initiated unconsciously, leaving conscious will a veto rather than an originating role.

The most consequential current challenge is Aaron Schurger's stochastic-accumulator reinterpretation, which denies that the readiness potential marks any real neural event and thereby dissolves the timing comparison on which the volitional argument rests. A further and counterintuitive point runs through the entry: although the readiness potential is widely enlisted as experimental support for a purely physicalist account of consciousness, Libet's own considered position was that phenomenal experience is non-physical, a view developed in Libet's Conscious Mental Field.

Neural Time and the Emergence of Conscious Experience

Libet approached consciousness experimentally by asking when neural activity becomes phenomenally experienced. Drawing especially on direct electrical stimulation and recording from the somatosensory cortex, medial lemniscus, and ventrobasal thalamus of awake neurosurgical patients, while the patient reported what was felt. The central result, developed across two decades, was that eliciting a conscious somatosensory experience by direct cerebral stimulation required a pulse train lasting up to approximately 500 milliseconds at liminal intensity; shorter trains produced nothing the patient reported feeling (Libet et al., 1964; Libet, 1973, 1993; summative account in Libet, 2004, 2006). Libet named the required state “neuronal adequacy,” and treated it as the physiological condition for phenomenal experience specifically, rather than for discrimination, detection, or response. The explanandum is explicitly the raw feel in Feigl's sense, as distinguished from behavior that may proceed with no accompanying experience. If the finding holds, phenomenal consciousness is not a rapid readout of sensory arrival but a cerebral state that takes time to constitute, and much of ordinary mental life must be transacted below it.

Neuronal Adequacy and the Half-Second Delay

He argued that appropriate neural activity must persist for a minimum duration before achieving what he called “neuronal adequacy” for conscious sensation. Under weak near-threshold cortical stimulation, this could require activity extending for as much as approximately 500 ms (Libet, 1993, 2004, 2006). Neural events could therefore occur, and potentially influence subsequent processing, without yet being accompanied by reportable phenomenal awareness.

Libet generalized this temporal dissociation cautiously but provocatively. If comparable requirements hold across modalities, conscious mental events do not originate simultaneously with the cerebral processes responsible for them; they emerge only after an unconscious neural phase. In his mature formulation, “unconscious cerebral processes precede a subjective sensory experience,” perhaps implying that mental events generally begin unconsciously (Libet, 2006).

Antedating and the Subjective Referral of Time

This raised an immediate phenomenological problem. The delay generates an immediate difficulty: a touch on the skin does not feel delayed. A tactile stimulus does not ordinarily seem to enter awareness half a second after it occurs. Libet and colleagues therefore proposed “subjective referral” or antedating: once the neural activity required for consciousness has become adequate, the resulting experience may be subjectively assigned to the time of an earlier cortical event—the initial evoked response produced by the incoming sensory signal (Libet et al., 1979). The phenomenal time of the experience would consequently differ from the cerebral time at which the neural conditions sufficient for awareness are finally achieved.

In other words, Libet's resolution was that the brain retroactively corrects the felt timing. He proposed that the primary evoked potential—the sharply localized cortical response arriving within tens of milliseconds of peripheral stimulation—serves as a time-marker, and that once neuronal adequacy is reached the subjective timing of the experience is referred backward to coincide with it (Libet et al., 1979). The test compared peripheral skin stimuli against cortical stimuli, which generate no such primary response and therefore should not be antedated; subjects' ordering judgments matched the prediction. If antedating is real, the felt temporal order of experience is not the order of the neural events producing it, and phenomenal timing becomes a represented content rather than a transparent record.

If correct, this would make phenomenal temporality itself partly constructed. The brain could process information unconsciously during an extended interval and nevertheless generate an experience whose apparent onset is referred backward toward the initiating sensory event. Objective neural chronology and subjective chronology would not map one-to-one.

The proposal remains contested. Dennett's Multiple Drafts Model rejects the framing rather than the data, arguing that referral of experience to an earlier moment presupposes the very inner theatre in which such a referral could be observed, which he famously argued against. Gomes argued that the duration required for successful cortical stimulation does not establish a corresponding latency for consciousness and therefore does not require retrospective referral (Gomes, 1998). Pockett likewise reinterpreted Libet’s evidence as compatible with substantially shorter conscious latencies and without antedating (Pockett, 2002, 2004). Libet’s specific half-second threshold should therefore not be treated as established neuroscience. His broader contribution is more secure: the timing of neural processing, the timing of reportable phenomenal awareness, and the subjectively experienced timing of an event are distinct variables requiring independent investigation.

The Time-On Factor

Libet then asked what converts unconscious cerebral processing into phenomenal experience. His answer was duration. Trains of 0 to 750 milliseconds were delivered to the ventrobasal thalamus through electrodes implanted for pain control, and subjects made forced-choice judgments about which of two intervals contained the stimulus while separately rating their awareness of it (Libet et al., 1991). They performed above chance at durations they reported not feeling at all. Libet's "time-on" theory takes the parameter governing the transition from unconscious detection to conscious sensation to be the duration of appropriate neuronal activation, with intensity trading off against it. The proposal is unusual in offering a single continuous quantity as the phenomenality-conferring variable rather than an architectural event such as the ignition described in Baars's and Dehaene's Global Workspace Theory.

The Readiness Potential and Conscious Will

The best-known experiments applied the same temporal method to volition. Subjects flexed a wrist or finger whenever they wished while watching a rapidly revolving clock spot, reporting the position at which they first became aware of the wish or urge to move, a quantity Libet called W (Libet et al., 1983). Scalp EEG showed the readiness potential first described by Kornhuber and Deecke (1965) beginning roughly 550 milliseconds before muscle activation for unplanned movements, while W fell only about 200 milliseconds before it. Libet concluded that cerebral initiation of a spontaneous voluntary act begins unconsciously, but preserved a role for conscious will in the residual interval, proposing that awareness can veto an already-prepared act—commonly called "free won't"—and restricting the finding to unplanned acts rather than to deliberation over longer timescales (Libet, 1985). Sapolsky's Hard Incompatibilism and Mitchell's Free Agents mark the two poles of the dispute this experiment provoked.

Schurger's Reinterpretation of the Readiness Potential

Schurger's challenge is directed at the ontology of the signal rather than at Libet's timing measurements. Because Libet-style paradigms average EEG epochs selected only because they culminate in movement, the resulting waveform may be an artifact of biased sampling. Schurger, Sitt, and Dehaene (2012) modeled spontaneous movement initiation as a leaky stochastic accumulator: with no external reason to move now rather than later, the moment of onset is fixed partly by ongoing, temporally autocorrelated fluctuations in neural activity, so time-locking to movement is equivalent to time-locking to crests in that noise. Averaging then produces a slow ramp resembling the readiness potential without any preparatory process having occurred. On this account, the "onset" of the readiness potential is physiologically meaningless: it marks no causally efficacious event, and comparing it with W is comparing a real time to a statistical artifact. Schurger, Hu, Pak, and Roskies (2021) develop the point as a late-decision account against Libet's early-decision one, and note that it renders the reported W of roughly 150 to 200 milliseconds not a lag but a reasonably accurate report of the threshold crossing. Supporting evidence includes rodent secondary motor cortex ramping fitted by the same model (Murakami et al., 2014) and an estimated point of no return near 200 milliseconds before movement (Schultze-Kraft et al., 2016). The authors do not claim the matter settled—whether a discrete readiness potential exists on single trials remains open (Travers et al., 2020)—but argue that even were the stochastic models wrong, the classical inference from the readiness potential to the inefficacy of conscious will would not follow. One limit deserves emphasis: the critique leaves Libet's sensory findings untouched, since neuronal adequacy, antedating, and time-on stand or fall on separate evidence.

Conscious Veto: “Free Won’t”

Libet also resisted the claim that his experiment eliminated conscious causal control. Although conscious intention appeared after the onset of the readiness potential, it appeared approximately 200 ms before movement. He proposed that consciousness might therefore retain an opportunity to allow the developing action to proceed or to veto it before execution (Libet, 1985, 1999).

Free will, if it exists, would thus operate less as conscious initiation than as conscious control—sometimes described as “free won’t.” Libet argued explicitly that his findings constrain models of free will but do not experimentally disprove it. Indeed, as noted, he rejected the view that conscious intentions should simply be regarded as causally impotent epiphenomena and considered a genuinely nondetermined conscious veto scientifically permissible (Libet, 1999).

There are important objections. The experiments did not independently establish that the veto itself originates consciously rather than from prior unconscious neural activity. The method for retrospectively reporting W is also subject to uncertainties involving attention, temporal estimation, and introspective access. More fundamentally, the simple purposeless movement used in Libet tasks differs radically from ordinary decisions involving reasons, alternatives, consequences, deliberation, or competing values.

The Materialist Appropriation—and Libet’s Rejection of It

The readiness potential has had an afterlife its author did not intend. The intuitive philosophical interpretation is powerful. If measurable brain activity begins preparing an action before the conscious subject experiences deciding to perform it, then conscious intention can appear to be a late-arriving representation of something the physical brain has already set in motion. On a stronger interpretation, the brain does the causal work while consciousness merely becomes aware afterward of what neural processes have already determined.

Libet’s findings have accordingly been invoked repeatedly against libertarian free will and in debates over whether conscious intention has genuine causal efficacy. They readily fit an epiphenomenalist or robust physicalist picture: unconscious physical processes produce both the action and the subsequent conscious experience of willing it, while phenomenal consciousness itself contributes nothing causally indispensable. Contemporary philosophical discussions explicitly identify Libet’s work as a major empirical source of the epiphenomenalist challenge to conscious agency (Moore, 2020; Robinson, 2023).

In other words, because Libet’s readiness potential appears to show that the brain commits to an act before the agent knows of it, the result is routinely enlisted not only against libertarian free will but for a physicalist conclusion about the nature of experience: that phenomenal consciousness is a causally idle accompaniment of physical processes, generated by the brain and explanatorily dispensable. Wegner (2002) built the most influential version, in which the feeling of willing is produced by a system distinct from the one producing action and interprets behavior rather than causing it; Harris (2012) presses the material to similar effect. Positions of that family are represented on the Landscape by Halligan and Oakley's Species-Enhancing Epiphenomenalism and, in a more radical register, by Frankish's Illusionism. Libet's data have become a standard exhibit for physicalism about consciousness.

But the striking, counterintuitive—and frequently obscured—fact is that this was not Libet’s own theory of consciousness. Libet himself drew the opposite ontological conclusion from his own findings. He defended free will and moral responsibility, insisted the veto was a real conscious power, and objected that those citing him against free will omitted it (Libet, 2004). More directly on point here, he held that subjective experience is a nonphysical phenomenon and that mind and brain are independent variables in causal interaction. What irony! The experimentalist whose signature result underwrites much contemporary physicalism about consciousness was himself an interactionist non-physicalist about phenomenal experience.

The gap between the data and their appropriation is worth stating precisely, since noticing it does not require siding with either side. When a decision is initiated is a question about temporal order within a causal chain; what makes any state phenomenally conscious is a question about the nature of that state. Nothing in the timing result bears directly on the second. Even granting that unconscious cerebral activity precedes awareness of intention, what follows is that conscious intention does not initiate the act, not that experience is physical. Schurger's critique compounds the difficulty from the other direction: if the readiness potential marks no real neural event, the evidential base for the epiphenomenalist reading contracts, or even dissolves, before the ontological inference is even reached.

Libet’s experimental results can certainly be appropriated by materialists; empirical findings do not belong metaphysically to the scientist who discovers them. Yet the fact remains: Libet himself did not infer from neural priority that phenomenal experience is reducible to neural activity. Quite the contrary: he eventually described subjective consciousness as nonphysical and developed his Conscious Mental Field theory precisely to explain how unified phenomenal experience could be generated by brain activity yet remain phenomenologically irreducible and potentially capable of acting back upon the brain (Libet, 1994, 2006). (See Libet’s Conscious Mental Fields.)

Such irony is therefore fundamental to understanding Libet. The experiment that became emblematic of “the brain decides before consciousness” was conducted by a neuroscientist who rejected the more mainstream conclusion that consciousness is therefore nothing but brain activity. For Libet, neural causation could precede consciousness without exhausting the ontology of consciousness.

Assessment

Libet's standing rests on having made phenomenal consciousness experimentally tractable in awake human subjects at a time when few thought it could be. The sensory findings introduced a temporal dimension that later work continues to develop, and the volitional experiment set the terms of a debate now four decades old. The vulnerabilities are equally clear. The stimulation studies rest on small clinical samples that cannot readily be replicated, and antedating remains contested at the level of interpretation rather than of data. The volitional result faces a serious alternative in which its central measurement refers to nothing, and its exposure is asymmetric: if Schurger’s stochastic account is right, the argument against conscious initiation fails, whereas if it is wrong, the argument survives to face its other critics.

What is not in dispute is that Libet established duration and delay as measurable variables of phenomenal consciousness, and that this remains true whatever becomes of the readiness potential. It is also not in dispute, though it is frequently overlooked, that he personally took those variables to characterize something he regarded as non-physical. That counterintuitive conjunction is essential to his place in Landscape. Libet supplies one of the most bracing historical cases in which neurobiological evidence commonly interpreted as supporting physicalist consciousness was produced by a theorist who explicitly rejected physicalism about phenomenal consciousness. Libet’s experiments establish a temporal problem for consciousness and agency; they do not settle the ontology of consciousness.

References

Harris, S. (2012). Free will. Free Press.

Kornhuber, H. H., & Deecke, L. (1965). Hirnpotentialänderungen bei Willkürbewegungen und passiven Bewegungen des Menschen: Bereitschaftspotential und reafferente Potentiale. Pflügers Archiv, 284, 1–17.

Libet, B. (1973). Electrical stimulation of cortex in human subjects, and conscious sensory aspects. In A. Iggo (Ed.), Handbook of sensory physiology, Vol. 2 (pp. 743–790). Springer.

Libet, B. (1985). Unconscious cerebral initiative and the role of conscious will in voluntary action. Behavioral and Brain Sciences, 8(4), 529–539.

Libet, B. (1993). The neural time factor in conscious and unconscious events. In Experimental and Theoretical Studies of Consciousness, Ciba Foundation Symposium 174, 123–146.

Libet, B. (1994). A testable field theory of mind-brain interaction. Journal of Consciousness Studies, 1(1), 119–126.

Libet, B. (1999). Do we have free will? Journal of Consciousness Studies, 6(8–9), 47–57.

Libet, B. (2004). Mind Time: The Temporal Factor in Consciousness. Cambridge, MA: Harvard University Press.

Libet, B. (2006). Reflections on the interaction of the mind and brain. Progress in Neurobiology, 78(3–5), 322–326.

Libet, B., Alberts, W. W., Wright, E. W., Delattre, L. D., Levin, G., & Feinstein, B. (1964). Production of threshold levels of conscious sensation by electrical stimulation of human somatosensory cortex. Journal of Neurophysiology, 27(4), 546–578.

Libet, B., Gleason, C. A., Wright, E. W., & Pearl, D. K. (1983). Time of conscious intention to act in relation to onset of cerebral activity (readiness-potential): The unconscious initiation of a freely voluntary act. Brain, 106(3), 623–642.

Libet, B., Pearl, D. K., Morledge, D. E., Gleason, C. A., Hosobuchi, Y., & Barbaro, N. M. (1991). Control of the transition from sensory detection to sensory awareness in man by the duration of a thalamic stimulus: The cerebral "time-on" factor. Brain, 114(4), 1731–1757.

Libet, B., Wright, E. W., Feinstein, B., & Pearl, D. K. (1979). Subjective referral of the timing for a conscious sensory experience: A functional role for the somatosensory specific projection system in man. Brain, 102(1), 193–224.

Murakami, M., Vicente, M. I., Costa, G. M., & Mainen, Z. F. (2014). Neural antecedents of self-initiated actions in secondary motor cortex. Nature Neuroscience, 17(11), 1574–1582.

Schultze-Kraft, M., Birman, D., Rusconi, M., Allefeld, C., Görgen, K., Dähne, S., Blankertz, B., & Haynes, J.-D. (2016). The point of no return in vetoing self-initiated movements. PNAS, 113(4), 1080–1085.

Schurger, A., Hu, P. B., Pak, J., & Roskies, A. L. (2021). What is the readiness potential? Trends in Cognitive Sciences, 25(7), 558–570.

Schurger, A., Sitt, J. D., & Dehaene, S. (2012). An accumulator model for spontaneous neural activity prior to self-initiated movement. PNAS, 109(42), E2904–E2913.

Travers, E., Friedemann, M., & Haggard, P. (2020). Do readiness potentials happen all the time? NeuroImage, 206, 116286.

Wegner, D. M. (2002). The illusion of conscious will. MIT Press.