Skip to content

Metaplasticity

Prime #
988
Origin domain
Life Sciences
Subdomain
neuroscience → Life Sciences

Core Idea

Metaplasticity is the structural pattern in which a system's capacity to change is itself modulated by prior activity, context, or accumulated history — the changeability is one level higher than the change itself. The signature commitment is that the system has at least two timescales of adaptation: a fast process, the change, and a slow process, the adjustment of how readily the fast process can operate. The slow process makes the system more or less willing to change the next time a change-trigger arrives, without itself producing a directly observable change in the system's output.

The signature has four parts: a base adaptive process — a learning rule, an update rule, a feedback gain — that produces first-order change in response to input; a slower governance process that modifies the parameters of the adaptive process (its rate, threshold, sign, gating) based on history; a latency such that the second-order process leaves no immediate trace until the next first-order event encounters the altered parameters; and a dependence on the adaptation-readiness profile so the system's behavior at any moment turns not just on its current input and state but on the history-shaped readiness to update. This is what makes the same stimulus elicit different magnitudes of update in the same system at different times. What the pattern changes in a reader's view is the recognition that learning rate and update threshold are themselves dynamic state: most systems analysis treats adaptation parameters as design constants, while metaplasticity reveals them as a third layer mediating between input and the adaptation layer, with its own dynamics and pathologies.

How would you explain it like I'm…

The Hidden Knob

Imagine a door that can swing open easily or stiffly. Normally we think about whether the door is open or shut — but there is also a hidden knob that slowly sets *how easily* the door swings next time. Metaplasticity is that hidden knob: it does not move the door itself, it changes how willing the door is to move when you next push it.

Changing How You Change

Metaplasticity is when a system's *ability to change* is itself changed by what happened before. There are two speeds: a fast process — the actual change, like learning something — and a slow process that adjusts *how easily* that fast change can happen next time. The slow part leaves no visible mark on its own; you only notice it when the next change-trigger arrives and the system reacts more, or less, than it would have. So the same nudge can produce a big change one day and a small change another day, depending on the system's history.

Plasticity Of Plasticity

Metaplasticity is the pattern in which a system's *capacity to change* is itself modulated by prior activity, context, or accumulated history — the changeability sits one level higher than the change itself. The signature commitment is two timescales of adaptation: a fast process (the change) and a slow process (the adjustment of how readily the fast process can operate). The slow process makes the system more or less *willing to change* the next time a trigger arrives, without itself producing a directly observable change in the system's output. That is what makes the same stimulus elicit different magnitudes of update in the same system at different times. The reframe is recognizing that learning rate and update threshold are themselves dynamic *state*, not fixed design constants — a third layer mediating between input and the adaptation layer, with its own dynamics and pathologies.

 

Metaplasticity is the structural pattern in which a system's capacity to change is itself modulated by prior activity, context, or accumulated history — the changeability is one level higher than the change itself. The signature commitment is that the system has at least two timescales of adaptation: a fast process, the change, and a slow process, the adjustment of how readily the fast process can operate. The slow process makes the system more or less willing to change the next time a change-trigger arrives, without itself producing a directly observable change in the system's output. The signature has four parts: a base adaptive process — a learning rule, an update rule, a feedback gain — that produces first-order change in response to input; a slower governance process that modifies the parameters of the adaptive process (its rate, threshold, sign, gating) based on history; a latency such that the second-order process leaves no immediate trace until the next first-order event encounters the altered parameters; and a dependence on the adaptation-readiness profile, so the system's behavior at any moment turns not just on its current input and state but on the history-shaped readiness to update. This is what makes the same stimulus elicit different magnitudes of update in the same system at different times. What it changes in a reader's view is the recognition that learning rate and update threshold are themselves dynamic state: most systems analysis treats adaptation parameters as design constants, while metaplasticity reveals them as a third layer mediating between input and the adaptation layer, with its own dynamics and pathologies.

Structural Signature

the fast base adaptive process — the slow governance process over its parameters — the latency before the meta-change shows — the history-shaped adaptation-readiness state

The pattern is present when each of the following holds:

  • A base adaptive process. A first-order rule — a learning rule, an update rule, a feedback gain — produces observable change in response to input.
  • A governance process over its parameters. A slower process modifies the parameters of the adaptive process — its rate, threshold, sign, or gating — on the basis of accumulated history, rather than producing output directly. The two processes operate on separable timescales.
  • A latency. The second-order change leaves no immediate observable trace; it manifests only when the next first-order event encounters the altered parameters.
  • A readiness-dependence. The system's response at any moment depends not only on its current input and state but on a history-shaped readiness to update, so the same stimulus elicits different magnitudes of update at different times.

The defining commitment is that learning rate and update threshold are themselves dynamic state, not design constants. These compose into a two-tier adaptive system: a fast tier that changes and a slow tier that governs how readily the fast tier changes — and absent a separable meta-state with its own slow dynamics, the system is merely adaptive, not metaplastic.

What It Is Not

  • Not adaptation. adaptation is the first-order change in response to input; metaplasticity is the second-order modulation of how readily that change happens. Without a separable slow governor over the adaptive parameters, a system is merely adaptive.
  • Not adaptive capacity. adaptive_capacity is the property of being able to change; metaplasticity is the motion of that capacity — its history-driven rise and fall — with its own dynamics.
  • Not learning. learning is the first-order layer that metaplasticity modulates; metaplasticity is learning-to-learn — change in the rules of change, not the change itself.
  • Not hysteresis. hysteresis is one effect a metaplastic state can produce (history-dependent response), but it is not the definition; metaplasticity names the two-tier governor, hysteresis names a path-dependence it can cause.
  • Not feedback. Negative feedback routes output back to modify input; metaplasticity modifies the gain or threshold of the adaptive process itself — it changes how the system learns, not what its output currently is.
  • Common misclassification. Reading a non-responding system as having "lost the ability to change." Plasticity loss (a degraded base rule) and metaplastic saturation (an intact rule behind a raised threshold) look identical from the output but demand opposite fixes — restore the base process versus reset the meta-layer.

Broad Use

In neuroscience, the origin, synaptic plasticity itself adapts: prior strong activity raises the threshold for further potentiation and lowers it for depression — a sliding threshold — without which runaway potentiation would destabilize networks. In machine learning, meta-learning, learning-rate schedulers, and adaptive optimizers maintain per-parameter or per-task adaptation rates that themselves update on history, and continual-learning regularizers raise stiffness on weights important to prior tasks. In organizational learning, an organization's appetite for change — its willingness to overturn process, hire heterodoxly, rewrite roadmaps — is shaped by prior change history, with repeated successful reorganizations raising the meta-rate and a botched change lowering it for years. In habit formation and behaviour change, motivation and self-efficacy function as a metaplastic state: recent small successes lower the threshold for new commitments, recent failures raise it. In education a student's learning-to-learn — study skills, prior-knowledge density, metacognitive monitoring — modulates how much they update per exposure. In the immune system, trained immunity reprograms innate cells to respond more or less strongly to subsequent stimuli. In policy and regulation, an agency's enforcement appetite shifts with leadership, recent scandals, and political pressure. And in software, adaptive auto-scalers and rate-limiters adjust their own sensitivity parameters based on recent load history — slow controllers wrapped around fast ones.

Clarity

Metaplasticity sharpens an otherwise-blurred distinction between current behavior and current willingness to adapt. Two systems that look identical right now may diverge wildly in response to the next disturbance because one is primed and the other saturated; without this vocabulary analysts attribute the difference to luck, hidden state, or noise, while with it the difference is read as a second-order state with its own laws of motion. It also distinguishes plasticity loss — the base process is degraded, as in aged neurons or ossified organizations — from metaplastic suppression — the base process is intact but the threshold has shifted to make it non-firing; resistance from intrinsic stiffness from resistance from a temporarily raised threshold that recent context will eventually relax; and hyper-plasticity — the base rule is too aggressive — from metaplastic over-priming — the threshold has dropped too far, inviting overfitting to the next stimulus. Drawing these lines is what keeps a diagnosis of "this system won't change" from collapsing several structurally distinct conditions into one, each of which calls for a different intervention.

Manages Complexity

The pattern compresses a sprawling class of phenomena — saturation, ceiling effects, sensitization, tolerance, burnout, change fatigue, refractory period, momentum, sliding scales — into a single structural template: a first-order adaptive process with a second-order parameter governing it. The diagnostic question becomes what the current state of the meta-layer is and what would change it, which routes attention away from the immediate input-output relation toward the slower dynamics that determine how that relation operates. It also rationalizes why fixed-parameter models of adaptive systems consistently mispredict: they ignore the second-order layer and so cannot account for context-dependent gain, and recognizing metaplasticity tells the modeller they need at least one more state variable. By naming the meta-layer as separable state, the pattern lets an analyst reason about a system's readiness to change independently of its current output, which is exactly the dimension that single-layer models collapse.

Abstract Reasoning

Several structural moves become available. Two-timescale separation: when analyzing a learning system, separate what is being learned from what is being learned about how to learn, since the two have different dynamics and require different interventions. Saturation diagnosis: a system that has stopped responding may have lost plasticity or may be metaplastically saturated, and the interventions differ — restoring plasticity versus resetting the meta-layer through novelty, perturbation, or rest. Priming engineering: deliberately raising or lowering the meta-state to alter the next adaptation, as in curriculum sequencing, warm-up tasks, or pre-mortems. Anti-runaway design: in any system with positive-feedback adaptation, a metaplastic governor that raises the threshold as activity rises is the standard defense against runaway, and designing it in is a generic pattern. Hysteresis interpretation: when a system shows different responses to identical stimuli depending on history, the natural model is a metaplastic state variable rather than a hidden parameter. And reversibility audit: changes induced by metaplastic shifts decay on the meta-layer's timescale, not the first-order timescale, so a "this change is sticky" claim should distinguish first-order from meta-order persistence. These are structural inferences about two-tier adaptive systems, independent of substrate.

Knowledge Transfer

Because the abstract structure — two-tier adaptation with the upper tier governing the lower — is substrate-free, the transfer is well-attested and runs both ways. The neuroscience sliding-threshold inspired adaptive learning-rate methods and the explicit meta-learning paradigm, the correspondence "adjust the learning rule based on accumulated history" preserved across the port; meta-learning algorithms returned the favor by giving neuroscientists a precise computational language for what metaplasticity might be computing. The language of priming, refractory periods, and saturation has been borrowed by organizational-change literatures, where double-loop learning is essentially metaplastic — change in the change rules. The recognition that change fatigue is a metaplastic state, not a moral failing, has reshaped behaviour-change interventions such as commitment-device design and nudge sequencing. Trained immunity is metaplasticity in the innate immune system, and recognizing it as a generic two-timescale structure has informed prime-boost vaccine schedules. And the failure modes port across domains as one diagnostic template: uncontrolled meta-saturation — cellular tolerance, organizational change fatigue, catastrophic forgetting under continual updates — follows the same shape everywhere. Across every port the prescription is to separate the fast rule from the slow governor, to diagnose whether a non-responding system has lost plasticity or saturated its meta-layer, and to engineer a governor against runaway. The transfer carries a decisive boundary: a receiving domain must commit to the two-tier model, distinguishing metaplasticity from plain adaptation (first-order change), from adaptive capacity (the property rather than its motion), from learning (the first-order layer it modulates), from hysteresis (one effect it can produce but not its definition), and from feedback (a related routing that modifies output rather than gain). If one cannot point to a separable meta-state and a slow process governing it, the system is merely adaptive, not metaplastic — and holding that line is what keeps the prime sharp as it travels from a synapse to a learning-rate schedule to an organization's appetite for change.

Examples

Formal/abstract

Consider the BCM (Bienenstock-Cooper-Munro) sliding-threshold rule for synaptic plasticity — the prime's home case, where the two-tier structure is mathematically explicit. The fast base adaptive process is Hebbian plasticity: a synapse strengthens (long-term potentiation) when postsynaptic activity exceeds a modification threshold \(\theta_M\) and weakens (long-term depression) when activity falls below it. The slow governance process over its parameters is the sliding of \(\theta_M\) itself: the threshold moves as a function of the time-averaged postsynaptic activity — formally, \(\theta_M \propto \langle a^2 \rangle\), the average of recent squared activity. A synapse driven hard for a long stretch raises its own potentiation threshold, so the same presynaptic stimulus that would have potentiated it earlier now produces less change or even depression — the prime's readiness-dependence made precise. The latency before the meta-change shows is intrinsic: sliding \(\theta_M\) produces no immediate change in synaptic weight; it only manifests when the next plasticity-inducing event arrives and finds the threshold relocated. This is exactly why fixed- threshold Hebbian rules are unstable — they permit runaway potentiation — and why the metaplastic governor is the standard anti-runaway design: the slow upward slide of the threshold as activity accumulates is a homeostatic brake that keeps the network in a usable dynamic range. The diagnostic the prime forces: a synapse that has "stopped potentiating" may have a degraded base rule (plasticity loss) or an intact rule with a raised threshold (metaplastic saturation) — structurally distinct conditions needing different fixes, and the BCM model makes the second one a computable state variable.

Mapped back: Hebbian potentiation is the fast base process, the sliding \(\theta_M\) the slow governance process over its parameters, the no-immediate- weight-change the latency, and the activity-averaged threshold the history-shaped readiness state — the two-tier governor that prevents runaway.

Applied/industry

Consider an organization's appetite for change as a metaplastic state, alongside the directly analogous learning-rate scheduler in machine learning. In the organizational case the fast base adaptive process is the organization's first-order ability to alter process, reorganize teams, and rewrite roadmaps in response to a trigger (a market shift, a failed launch). The slow governance process over its parameters is the slow drift of the organization's willingness to undertake such changes, shaped by its change history: a string of well-run reorganizations that paid off raises the meta-rate — people green-light the next change readily — while one botched, painful change lowers the threshold for years, so the same proposal that would have sailed through before now meets reflexive resistance. The latency is characteristic: a demoralizing reorg leaves no immediate visible change in how the organization operates day-to-day; its effect appears only when the next change proposal arrives and is met with change fatigue. The readiness-dependence is the practical payoff: two organizations with identical current structure respond oppositely to the same disruption because one is primed and the other saturated — and recognizing change fatigue as a metaplastic state, not a moral failing, reroutes the intervention toward resetting the meta-layer (small early wins to lower the threshold, rest between changes) rather than exhorting people to "embrace change." The ML parallel is exact and was a documented two-way port: an adaptive optimizer (Adam, learning-rate warmup-and-decay) maintains per-parameter learning rates that themselves update on the history of gradients — a slow controller wrapped around the fast weight-update rule, raising or lowering how much each weight moves per step based on accumulated gradient statistics. Both diagnose the same way: separate what is being learned from what is being learned about how readily to learn.

Mapped back: First-order reorganizing (or weight updates) is the fast base process, the history-shaped change appetite (or adaptive learning rate) the slow governor, the delayed onset of change fatigue (or rate decay) the latency, and the primed-versus-saturated stance the readiness state — the same two-tier structure ported between an organization and an optimizer.

Structural Tensions

T1 — Two Timescales versus One (temporal). The prime requires a separable slow governor over a fast base process — and the whole structure dissolves if the two timescales are not actually distinct. The failure mode is modeling a genuinely two-tier system with a single-layer, fixed-parameter model, which cannot account for context-dependent gain and so mispredicts whenever the meta-state has shifted. Conversely, positing a meta-layer where the apparent history-dependence is just slow first-order dynamics over-fits a phantom tier. Diagnostic: ask whether you can point to a meta-state that persists across first-order events and evolves on its own slower clock — if no separable slow variable exists, the system is merely adaptive, not metaplastic, and the second tier is imaginary.

T2 — Current Behavior versus Current Readiness (sign/direction). Metaplasticity insists that what a system does now and how willing it is to change now are different states pointing potentially opposite ways. Two systems identical in current output can respond oppositely to the next disturbance because one is primed and the other saturated. The failure mode is reading readiness off behavior: inferring that a quiescent system is unwilling to change, or that an active one is primed, when the meta-state is orthogonal to the output. Diagnostic: ask what would happen on the next change-trigger, not what is happening now — a readiness claim grounded only in present behavior ignores exactly the second-order state the prime names.

T3 — Plasticity Loss versus Metaplastic Saturation (scopal). A system that has stopped responding admits two structurally distinct diagnoses: the base rule is degraded (plasticity loss — aged neurons, ossified org) or the base rule is intact but the threshold has shifted to make it non-firing (metaplastic saturation). They look identical from the output and demand opposite interventions — restore the base process versus reset the meta-layer through novelty, rest, or perturbation. The failure mode is collapsing them into "this system won't change" and applying one fix to both. Diagnostic: ask whether the base mechanism still can fire under a different meta-state — if a reset or novel context restores responsiveness, it was saturation, not loss.

T4 — Latency versus Apparent Inertness (temporal). The second-order change leaves no immediate observable trace; it manifests only when the next first-order event encounters the altered parameters. The failure mode is concluding a meta-intervention had no effect because nothing changed right after it — a demoralizing reorg, a curriculum warm-up, a threshold shift all look inert until the next trigger arrives. Reasoning that requires immediate feedback to confirm an effect will systematically under-detect metaplastic changes. Diagnostic: ask whether the absence of an immediate response means no change or latent change awaiting the next event — the meta-layer is invisible precisely between events, so its effects must be probed by the next stimulus, not by inspecting the system at rest.

T5 — First-Order Persistence versus Meta-Order Persistence (measurement). A "this change is sticky" claim is ambiguous: the persistence may live in the first-order state (the weight, the process) or in the meta-layer (the readiness), and these decay on different timescales. The failure mode is auditing reversibility at the wrong tier — assuming a behavior change is durable because the meta-state is primed, when the underlying first-order change was never consolidated, or vice versa. Diagnostic: ask which layer the persistence is claimed for, and measure decay on that layer's clock — a change that is sticky at the meta-level may evaporate at the base level the moment the readiness relaxes, and the two must be tracked separately.

T6 — Anti-Runaway Governor versus Over-Damping (sign/direction). The metaplastic governor that raises the threshold as activity accumulates is the standard defense against runaway potentiation — but the same brake, set too aggressively, suppresses legitimate adaptation, leaving a system that cannot learn when it should. The governor and the base process pull against each other by design, and the calibration is the whole game. The failure mode is either an absent/weak governor (runaway, instability, overfitting to the latest stimulus) or an over-strong one (chronic saturation, change fatigue, frozen learning). Diagnostic: ask whether the meta-layer's threshold tracks activity tightly enough to prevent runaway without clamping below the level needed for real updates — a governor tuned only against instability will silently cost the system its adaptiveness.

Structural–Framed Character

Metaplasticity is a mixed-structural prime, sitting just on the structural side of the structural–framed spectrum. Its skeleton is mathematically clean — a base adaptive process produces first-order change, and a slower governance process modulates the parameters of that process (rate, threshold, sign, gating) from history, so that changeability is itself dynamic state one level up. That two-timescale, adaptation-of-adaptation structure recurs in adaptive learning-rate schedules, gain-of-gain control loops, and second-order optimizers. The neuroscience name is the only thing pulling it in from the bare end.

The diagnostics read structural with one translatable seam. The pattern carries no evaluative weight: a meta-layer that raises an update threshold is neither good nor bad — over-damping and runaway potentiation are symmetric failure modes, value-neutral until you say what the system should learn. It is not human-practice-bound (human_practice_bound 0): an Adam optimizer adapting its per-parameter step sizes, or an automatic-gain-control circuit adjusting its own responsiveness, instantiate the second-order-modulation pattern in pure computational and physical substrates with no human in the loop. And invoking it largely recognizes a layered-dynamics structure already present — that learning rate and update threshold are themselves history-shaped state is a fact about the system, not an imported frame. What pulls it to the center is the home vocabulary: "plasticity," "potentiation," "metaplasticity" arrive from synaptic neuroscience and must be translated into rate-of-a-rate terms when the substrate is an optimizer or a control loop (vocab_travels and import_vs_recognize each 0.5, institutional_origin 0.5 for the field of origin). The second-order-adaptation core is clean and medium-neutral; the neuroscience label is a thin overlay — exactly the mixed-structural reading the aggregate of 0.3 records.

Substrate Independence

Metaplasticity is a maximally substrate-independent prime — composite 5 / 5 on the substrate-independence scale. On domain breadth, the second-order-adjustment-of-adaptation-parameters pattern recurs with identical force across neuroscience (its origin — the BCM sliding threshold), machine learning (meta-learning, adaptive optimizers like Adam, learning-rate schedules, continual-learning regularizers), organizational learning (an organization's history-shaped appetite for change; double-loop learning), habit and behavior change (motivation and self-efficacy as a metaplastic state), education (learning-to-learn), immunity (trained immunity reprogramming innate cells), and software (adaptive auto-scalers wrapping fast controllers) — a wide span of biological, computational, and institutional substrates, a clear 5. On structural abstraction, the skeleton is mathematically clean — a fast base adaptive process governed by a slower process that modulates its rate, threshold, sign, or gating from history — and an Adam optimizer or an automatic-gain-control circuit instantiates it with no human in the loop; the neuroscience vocabulary ("plasticity," "potentiation") is a thin overlay over a medium-neutral two-timescale structure, a 5. On transfer evidence, the prime scores a 5 — the transfer is well-attested and explicitly bidirectional: the sliding-threshold rule inspired adaptive learning-rate methods and the meta-learning paradigm, and meta-learning algorithms returned a precise computational language to neuroscience, with the failure modes (saturation, change fatigue, catastrophic forgetting) porting as one diagnostic template. Every component reads maximal, anchoring the composite at 5.

  • Composite substrate independence — 5 / 5
  • Domain breadth — 5 / 5
  • Structural abstraction — 5 / 5
  • Transfer evidence — 5 / 5

Relationships to Other Primes

One-hop neighborhood: parents above, mutual partners to the right, children below.Metaplasticitysubsumption: AdaptationAdaptation

Parents (1) — more general patterns this builds on

  • Metaplasticity is a kind of Adaptation

    metaplasticity is the SECOND-ORDER specialization of adaptation: a slow governor over the parameters (rate, threshold, sign, gating) of a fast first-order adaptive process — 'learning to learn.' More-specific CHILD of adaptation, not a reparent.

Path to root: Metaplasticity → Adaptation

Neighborhood in Abstraction Space

Metaplasticity sits among the more crowded primes in the catalog (31st percentile for distinctiveness): several abstractions describe nearly the same structure, so a description that fits it will tend to fit its neighbors too — transporting it usually means disambiguating within this family rather than landing on it exactly.

Family — Generative Rules & Stage-Wise Change (19 primes)

Nearest neighbors

Computed from structural-signature embeddings · 2026-06-14

Not to Be Confused With

The most fundamental confusion is with plain adaptation, because metaplasticity is a kind of adaptation — only one level up. Adaptation is first-order change: a system alters its state, weights, or behavior in response to input, on the timescale of the input. Metaplasticity is the second-order modulation of how readily that first-order change occurs — a slower process that shifts the rate, threshold, sign, or gating of the adaptive rule based on accumulated history, leaving no immediate trace until the next adaptive event encounters the altered parameters. The line is sharp and load-bearing: if you cannot point to a separable meta-state that persists across first-order events and evolves on its own slower clock, the system is merely adaptive and the second tier is imaginary. Conversely, modeling a genuinely two-tier system with a single-layer fixed-parameter model cannot account for context-dependent gain and mispredicts whenever the meta-state has shifted. The practical test is whether the same stimulus elicits different magnitudes of update at different times because of history — that history-shaped readiness is the metaplastic signature, absent in plain adaptation.

It must also be distinguished from hysteresis, with which it is easily conflated because both produce history-dependent behavior. Hysteresis is the phenomenon: a system's current response depends on its past trajectory, so identical inputs yield different outputs depending on the path taken to reach them. Metaplasticity is a mechanism that can produce hysteresis — one of several — but it is defined by its specific architecture (a slow governor over a fast adaptive rule), not by the path-dependence it generates. Many hysteretic systems have no meta-layer at all (a magnetic material's hysteresis loop is not metaplasticity). The error of identifying them is to posit a two-tier governor wherever history-dependence appears, over- fitting a phantom meta-state to what may be plain slow first-order dynamics. When a system shows different responses to identical stimuli, metaplasticity is a candidate model, not a conclusion.

A third confusion is with negative feedback, since both are regulatory loops that keep an adaptive system in a usable range. The distinction is in what is routed back. Feedback routes a function of the system's output back to modify its input, correcting the output toward a setpoint. Metaplasticity does not touch the output directly at all; it modifies the gain or threshold of the adaptive process — how much the system will change next time — on the basis of history. A metaplastic governor that raises the potentiation threshold as activity accumulates is the standard anti-runaway design, and it can look like negative feedback, but it acts on the learning rule rather than on the output, and its effect is latent until the next adaptive event. A practitioner who models a gain-modulating governor as an output-correcting loop will look for the wrong setpoint and miss the latency that is metaplasticity's hallmark.

These distinctions decide the intervention. A non-responding system that is merely adaptation-limited needs its base rule restored; one that is metaplastically saturated needs its meta- layer reset through novelty, rest, or perturbation; a hysteretic one may need neither; and a feedback problem needs the loop retuned, not the learning rate. The prime earns its keep by isolating the readiness-to-change as a separable state with its own slow dynamics, distinct from the change, the capacity, the path-dependence, and the output-correcting loop it is so easily confused with.

Solution Archetypes

No catalogued solution archetypes reference this prime yet.