Kshanikavada, the Buddhist doctrine of momentariness, holds that every conditioned thing, including each moment of awareness, lasts only an instant and is replaced by a causally linked successor; continuity is an appearance produced by the stream. Neuroscience has real evidence that perception samples the world in cycles, around 10 Hz for a sensory alpha rhythm and about 7 Hz for attentional sampling, and that conscious percepts may arrive as discrete "time slices". The parallel holds at the level of structure (experience is chunked, not continuous).
- Kṣaṇikavāda: every conditioned thing lasts one indivisible moment and is replaced by a causally linked successor.
- Modern estimates of an Abhidharma moment run from 0.13 to 13 ms; a 10 Hz alpha cycle lasts 100 ms.
- Perception samples in rhythms near 10 Hz (sensory alpha) and 7 Hz (attention), and percepts may follow up to 400 ms of unconscious integration.
- The "alpha frame rate" idea failed a careful 2022 test: peak alpha frequency did not predict when two flashes fuse.
- Verdict: a real structural parallel for the mind, no evidence for the strong claim that all things perish each instant.
The parallel fails on scale and on metaphysics: Abhidharma moments are estimated at 0.13 to 13 milliseconds, far shorter than any measured perceptual cycle, and no brain rhythm shows that things themselves perish every instant.
This post is for readers who know the basic Buddhist teaching of impermanence and want to know whether the much stronger Abhidharma claim, that nothing lasts more than a moment, has any counterpart in the science of perception. The comparison is an analogy and is labelled as one throughout. The Buddhist side is presented as philosophy, with its own arguments, not as a prediction about brains.
What does Buddhist momentariness actually claim?
Momentariness claims that conditioned phenomena exist for a single, indivisible moment (kshana) and then cease, each giving rise to the next. Noa Ronkin's Stanford Encyclopedia of Philosophy entry on Abhidharma, substantively revised in March 2026, describes how both the Sarvastivada and the later Theravada turned the early teaching of impermanence into this theory of discrete events.
The early texts already stress how fast the mind changes. In the Anguttara Nikaya (AN 1.48, translated by Bhikkhu Sujato), the Buddha says he sees nothing as quick to change as the mind, so quick that no simile captures it. The Abhidharma schools then gave that speed a structure:
- Sarvastivada-Vaibhashika: every conditioned thing has four characteristics, origination, endurance, decay and dissolution. The moment is the smallest definite unit of time that cannot be divided.
- Sautrantika: there is no real duration at all, only a succession of instants; a cause ceases as its effect arises. That makes direct perception hard to explain, so the Sautrantika held that we perceive objects indirectly, through representations (akara) left by moments already gone.
- Theravada: each mind-moment has three sub-moments (arising, persisting, ceasing). Only mental events are strictly momentary; material phenomena last sixteen or seventeen consciousness moments.
Continuity is explained by the stream (santāna). The SEP entry on mind in Indian Buddhism notes that terms such as citta-santāna, the "mind-stream", were introduced partly to explain why a series of separate events feels like one continuing person.
Memory works through seeds or traces carried in the stream, not through anything that endures.
Modern perception science has its own founding scene about still moments that look like motion. As it is usually told, the psychologist Max Wertheimer was travelling by train on holiday in 1910 when he noticed the blinking lights at a railway crossing, got off at Frankfurt and bought a zoetrope, a motion-picture toy, to experiment with. The story may be apocryphal, but what followed is documented.
At the Psychological Institute of Frankfurt University he used a tachistoscope to flash shapes onto a screen for precise durations. His two younger colleagues, Kurt Koffka and Wolfgang Köhler, were his subjects. His 1912 paper on the perception of movement, which his students nicknamed the "dot paper" for its abstract dot illustrations, described the movement observers saw between stationary flashes and helped launch Gestalt psychology. Separate, still events, correctly timed, are experienced as one continuous thing: the same claim the Abhidharma makes about the mind.
New World Encyclopedia: Max Wertheimer (1910 train story, Frankfurt experiments, 1912 paper)| School | Is there duration within a moment? | How we perceive | What carries continuity |
|---|---|---|---|
| Sarvastivada-Vaibhashika | Yes, four phases in sequence | Directly; past and future dharmas also exist | "Homogeneity" causes, and prapti (possession) |
| Sautrantika | No, only succession | Indirectly, through representations | Seeds (bija) |
| Theravada | Three sub-moments | A sequence of functions in a cognitive process (citta-vithi) | Life-continuum (bhavanga) |
| Dharmakirti (6th to 7th c.) | No; to exist is to be causally efficient | Non-conceptual perception of particulars | Causal series of particulars |

Why did Buddhist philosophers think things last only a moment?
The main arguments were logical, not observational. The SEP entry on Dharmakirti (revised April 2026) sets out the two classic proofs:
- The argument from perishing (vinashitvanumana). Destruction needs no cause; it belongs to a thing's nature. If perishing is intrinsic, a thing perishes as soon as it exists. The SEP entry judges this argument weak, because its steps do not follow.
- The argument from existence (sattvanumana). Largely Dharmakirti's own: to exist is to be causally efficient (arthakriya). Something that stayed the same could not produce new effects at different times, so whatever exists must be new at each moment. A permanent cause would be causally idle.
The second argument is the interesting one for a scientist, because it is a claim about change and causation, not about perception. It says nothing about how fast moments are, only that they cannot be extended.
That matters for the comparison below: no measurement of a brain rhythm can confirm or refute a thesis that is about the nature of existence.
Reviews how the Sarvastivada, Sautrantika and Theravada built a theory of momentary dharmas out of the teaching of impermanence. Records that a moment was treated as the smallest indivisible unit of time, with modern estimates ranging from about 0.13 to 13 milliseconds and no consensus, and that Theravada held material phenomena last sixteen or seventeen mind-moments. Limit: a reference work, so it reports the traditions' positions rather than testing them.
Read the entry (free)Is perception continuous or discrete?
The evidence says perception is partly discrete: the brain samples incoming information in rhythmic cycles, and conscious percepts may be updated in steps rather than a smooth stream. Rufin VanRullen's 2016 review in Trends in Cognitive Sciences concludes that perception runs on several rhythms rather than one, which may vary with sense, task, stimulus and brain region.
In vision, VanRullen describes a sensory alpha rhythm near 10 Hz alongside at least one other rhythm near 7 Hz that samples attention. At 10 Hz, one cycle lasts 100 ms; at 7 Hz, about 143 ms.
Newer work keeps finding rhythmic structure. A 2026 study in Cerebral Cortex by Noah and colleagues showed 20 participants a 50 ms target with a distractor and found that information about both oscillated in the theta range (4 to 8 Hz) for almost 1,000 ms after the display disappeared, with better performance when the two were further apart in phase.
Proposes a two-stage model of perception. Features are first processed unconsciously, quasi-continuously and with high temporal resolution, for up to about 400 ms; when processing settles, all features enter awareness together as one discrete percept. Evidence includes feature fusion, where TMS still changed how two verniers shown for 60 ms in total were combined up to 400 ms later, and apparent motion detectable with timing differences of only 3 ms. Limit: a theoretical synthesis; it argues that simultaneity tests cannot measure how long a percept lasts, so the duration of a "slice" is not pinned down.
Read the paper (free)
The closest match: Sautrāntika and the two-stage model
Herzog's model is the closest thing in science to the Sautrantika picture. In both, what you are aware of now is a packaged result of processing that is already finished, not the event as it happens.
The Sautrantika said we perceive representations of moments that have ceased; the two-stage model says conscious percepts are released after up to 400 ms of unconscious integration. That is a genuine structural parallel, and it is still an analogy: the Sautrantika argument was about causation (an effect cannot coexist with its cause), not about neural delays.
In both pictures, what you are aware of now is a report on something that has already happened.
Do Abhidharma moments match brain rhythms?
No. Even the longest modern estimate of an Abhidharma moment, about 13 ms, is roughly eight times shorter than a 10 Hz alpha cycle, and the shortest, 0.13 ms, is shorter than a single neural action potential. The chart puts the numbers on one scale.
See the data as a table
| Duration (ms) | |
|---|---|
| Abhidharma moment, short estimate | 0.13 |
| Apparent motion threshold | 3 |
| Abhidharma moment, long estimate | 13 |
| Simultaneity window for two disks | 40 |
| Alpha cycle (~10 Hz) | 100 |
| Attentional sampling cycle (~7 Hz) | 143 |
| Unconscious integration before a percept | 400 |
Source: SEP Abhidharma (Ronkin 2026); VanRullen, Trends in Cognitive Sciences 2016; Herzog et al., PLOS Biology 2016 · Cycle lengths computed as 1000/frequency from VanRullen 2016 (10 Hz, 7 Hz). Thresholds and the 400 ms figure from Herzog et al. 2016.
Use the calculator to see how many Abhidharma moments would fit into a brain rhythm of your choice.
How many Abhidharma moments fit in one perceptual cycle?
The mismatch is not a refutation, because the Abhidharma moment was never meant to be a perceptual frame. It is a unit of existence, the lifetime of a dharma. If anything maps onto a perceptual cycle, it is the Theravada cognitive process (citta-vithi), a sequence of mind-moments that together make one act of perceiving, which is closer to Herzog's "slice" than a single kshana is.
Is the alpha-rhythm "frame rate" theory settled?
No. The idea that each alpha cycle is a perceptual frame has weakened under careful tests, which is a reason not to build philosophical parallels on it. Buergers and Noppeney (Nature Human Behaviour, 2022) tested whether alpha frequency sets the window in which two flashes fuse into one, the core prediction of alpha-as-frame-rate.
Twenty adults did two-flash discrimination tasks with zero, one or two sounds, with EEG, across three experiments over about five days. Neither trial-by-trial alpha frequency nor a person's peak alpha frequency predicted their temporal binding window; Bayes factors mostly favoured no effect, and the one significant result ran the wrong way and did not replicate. Limit: one task family and 20 people; the authors say the question is not closed and needs replication with standardised methods.
Read the paper (free)So the current scientific picture is layered: there are rhythms in perception, attention samples periodically at around 7 Hz in many tasks, and awareness may be updated in steps, but there is no single fixed frame rate. That is closer to VanRullen's "several rhythms" view than to a cinema projector, and much closer to the varied Abhidharma accounts than to a single doctrine.

Where does the parallel hold, and where is it only a metaphor?
It holds on three structural points and fails on three substantive ones. The table separates them.
| Point | Buddhist claim | Perception science | Verdict |
|---|---|---|---|
| Experience is chunked | Awareness is a series of discrete mind-moments | Rhythmic sampling at ~7 to 10 Hz; discrete conscious percepts (Herzog 2016) | Structural parallel |
| Continuity is constructed | The stream (santana) creates the look of an enduring mind | Smooth experience is built from sampled, integrated input | Structural parallel |
| We see the past | Sautrantika: perception is of representations of ceased moments | Percepts released after up to 400 ms of processing | Structural parallel, different reasons |
| Size of the unit | 0.13 to 13 ms (modern estimates) | ~100 to 143 ms cycles; ~400 ms integration | No match |
| What is momentary | All conditioned things, including matter (Sarvastivada) | Only neural sampling; objects persist | Metaphor only |
| Why it is momentary | To exist is to be causally efficient (Dharmakirti) | Physiology of oscillating circuits | Different kinds of claim |
The cleanest way to put it: neuroscience supports a modest, mind-only version of momentariness (experience is sampled and assembled). It is silent on the strong Abhidharma version (all conditioned things perish each instant), which is a metaphysical thesis defended by argument. Calling one a "confirmation" of the other confuses a claim about the brain's timing with a claim about what it is to exist.
What does this mean for the self and rebirth?
If each mind-moment is new, there is no enduring subject passing from one to the next, which is why momentariness and not-self (anātman) support each other. The SEP entry on mind in Indian Buddhism describes persons as reducible to the five aggregates, with "self" a conventional label for a causal series. The obvious question, what then carries memory or karma, is answered by causal continuity: each moment conditions the next, as a flame lights a wick.
That answer is taken further in what gets reborn without a self and set against brain-imaging work on self-processing in atman, anatta and the neuroscience. Nagarjuna later pushed the critique one step further and denied that even moments have intrinsic nature; that argument is in Nagarjuna's catuskoti. For the resting brain's own rhythm of self-talk, see the default mode network, and for the wider set of essays, Manthan.
What's new
- Noa Ronkin's SEP entry on Abhidharma, including momentariness, was substantively revised Stanford Encyclopedia of Philosophy, Mar 2026
- The SEP entry on Dharmakirti, covering his proofs of momentariness, was substantively revised Stanford Encyclopedia of Philosophy, Apr 2026
- Target and distractor information oscillates at theta rhythm for nearly a second after a 50 ms display Cerebral Cortex, 2026
- A new framework proposes that breathing gates perceptual windows through coupled brain rhythms Neuroscience of Consciousness, 2026
Questions people ask
How long is a moment (kshana) in Buddhism?
There is no single answer. Abhidharma traditions treated it as the smallest indivisible unit of time, and modern estimates drawn from those texts range from about 0.13 to 13 milliseconds, with no consensus (SEP, Abhidharma). Meditation teachers sometimes speak of "moments" lasting a second or more, but that is a practical unit, not the doctrinal kshana.
Does neuroscience prove the Buddhist doctrine of momentariness?
No. It supports the idea that perception is sampled in cycles and that conscious percepts arrive as discrete packages. It cannot test the claim that all conditioned things perish every instant, which Dharmakirti defended with an argument about causal efficiency, not with observations.
If everything is momentary, how does memory work?
Buddhist schools gave different answers: seeds or traces carried in the mind-stream (Sautrantika, Yogacara), a possession factor (Vaibhashika), or the life-continuum (Theravada). Vasubandhu described memory as a capacity that arises from earlier experience, like a sprout from a seed.
How did Hindu philosophers answer the doctrine of momentariness?
Mostly through memory and moral responsibility. A Buddhism Stack Exchange thread quotes the Saiva Mrgendra Agama, which argues that consciousness only looks momentary because the instruments that manifest it work in brief bursts, and that memory of past experience and the fruition of past actions both require something that persists. The Brahma Sutras' refutation of the Buddhist realists, discussed in a related thread, presses the same point. Buddhist replies lean on causal continuity in the mind-stream.
Is the brain's "frame rate" 10 per second?
Not in any simple sense. Alpha (about 10 Hz) and attentional sampling (about 7 Hz) are both candidates, but careful tests such as Buergers and Noppeney 2022 found that a person's alpha frequency did not predict when two flashes fuse. Most researchers now talk about several rhythms, not one frame rate.
Sources
- Ronkin, Stanford Encyclopedia of Philosophy: Abhidharma (2010, revised 2026)
- Stanford Encyclopedia of Philosophy: Dharmakirti (2011, revised 2026)
- Stanford Encyclopedia of Philosophy: Mind in Indian Buddhist Philosophy (2009, revised 2012)
- Buddhism Stack Exchange: How would a Buddhist respond to the Mrgendra Agama's criticism?
- Anguttara Nikaya 1.48, translated by Bhikkhu Sujato, SuttaCentral
- VanRullen: Perceptual Cycles, Trends in Cognitive Sciences 20(10) (2016)
- Herzog, Kammer & Scharnowski: Time Slices: What Is the Duration of a Percept? PLOS Biology (2016)
- Buergers & Noppeney: The role of alpha oscillations in temporal binding within and across the senses, Nature Human Behaviour (2022)
- Noah et al.: Rhythmic sampling and competition of target and distractor representations in visual sensory memory, Cerebral Cortex (2026)
- Andrews & Dockree: The Rhythmic Embodied Perception Framework of breath, brain, and perception, Neuroscience of Consciousness (2026)
- Wikipedia: The Horse in Motion (Muybridge's 1878 Palo Alto series and the zoopraxiscope)
- New World Encyclopedia: Max Wertheimer (1910 train story, Frankfurt tachistoscope experiments, 1912 paper)
