Glossary of
Framework Terms

The current minimum capacity threshold for each node in the fitness adaptation graph. The fitness system equivalent of the education system's schema floor — the level below which that adaptation type cannot be safely loaded. Stored per node in the Dynamic Adaptation Profile. If an adaptation floor is not met, higher-load nodes that depend on it are locked until the floor is stabilized.

The process of continuously adjusting the difficulty or load of a session based on real-time performance data, rather than using a fixed schedule. In education: adjusting problem difficulty after every problem based on accuracy and latency. In fitness: adjusting workout load based on readiness score, HRV trend, and recovery state. The goal in both systems is to keep the user in the productive zone — challenged enough that real development is happening, not so challenged that the system breaks down.

A person's emotional state. In educational contexts: whether the student is engaged, anxious, bored, frustrated, confident, or curious. In fitness contexts: whether the user feels motivated, fatigued, dreading the session, or energized. Affective state is not a side concern — it directly determines whether productive engagement happens. Anxiety consumes working memory. Boredom produces disengagement. Both systems address affective state structurally through calibration and design, not through encouragement alone.

In philosophy: a statement is analytic if it is true by definition — true because of what the words mean, not because of what the world happens to be like. "A bachelor is unmarried" is analytic. Mathematical truths are analytic in this sense: 2 + 2 = 4 is true because of what 2, +, =, and 4 mean, not because we have checked many times and it keeps working out. This is why mathematical errors are always logical contradictions rather than empirical surprises — the truth was already determined by the definitions.

The ability to perform an operation so fluently that it requires almost no conscious attention or working memory. In education: automatically recalling that 7×8 = 56 without having to calculate it. In fitness: executing a movement pattern without having to consciously think through each component. Automaticity frees cognitive or attentional resources for higher-level tasks. A student who has to think about basic arithmetic while doing algebra is using cognitive resources for arithmetic that could be used for the algebra. A lifter who has to think about foot position can't also be thinking about bar path. Development of automaticity is the goal of fluency training in both systems.

The nervous system's ability to efficiently regulate the body's response to stress and recovery. Measured primarily through heart rate variability (HRV). A well-adapted autonomic system recovers faster between efforts, tolerates higher training loads, and produces more consistent physiological responses. Node 9 in the fitness DAG — a foundational node with no prerequisites, meaning it can be trained even when other adaptations are compromised. When readiness drops to the lowest level (STATE_LEVEL_3), autonomic recovery work is the mandatory response because everything else depends on the nervous system being functional.

A measurable degradation in movement quality during a session, detected when concentric (lifting) velocity drops more than 15% outside expected fatigue curves. The fitness system equivalent of a wrong-schema flag in education — a signal that the structural pattern underlying the movement has broken down, not just that the user is tired. When detected, the system automatically routes to closed-chain remediation exercises. Distinguished from normal fatigue-related slowing: normal fatigue produces proportional velocity reduction that matches predicted curves; form drift produces irregular, asymmetric, or structurally inconsistent patterns.

The specific constraint that is most limiting performance at any given moment. In education: the reading comprehension ceiling, a fluency gap, or a schema floor — whichever is most constraining progress right now. In fitness: the recovery node, the weakest adaptation in the prerequisite chain, or the readiness state. Identifying the actual bottleneck rather than the apparent one is the primary job of the diagnostic system. Working on the wrong constraint — drilling mathematics when the bottleneck is reading comprehension, or adding training volume when the bottleneck is recovery — produces no progress and often makes things worse.

Adjusting the difficulty or load of a task to match the current capability of the specific person doing it. Not a one-time setup — a continuous process that updates as the person develops. The goal: keep the person working in the zone where genuine development is happening. Too easy = no growth. Too hard = breakdown and disengagement. In education: calibrated to produce approximately 85% accuracy on growth-edge problems. In fitness: calibrated to the readiness state and current adaptation floor. A well-calibrated system feels challenging but achievable — which is also what makes it sustainable.

The growth of new small blood vessels (capillaries) in muscle tissue. Node 8 in the fitness adaptation DAG — a foundational node with no prerequisites. Capillarization improves the delivery of oxygen and nutrients to muscle and the removal of waste products. It is a prerequisite for hypertrophy (Node 2) because muscle growth requires adequate blood supply, and for metabolic adaptation (Node 5) because substrate utilization efficiency depends on capillary density. Trained primarily through sustained aerobic and sub-maximal work.

The diagnostic step that continues one level above the identified floor to check for the wrong-schema condition — the case where a person appears to have higher-level capability resting on a corrupted lower-level foundation. In education: if schema is found partial or incorrect at Tier 2, the diagnostic continues to Tier 3 to check whether apparent Tier 3 knowledge is genuine or is actually built on the wrong Tier 2 structure. In fitness: if an adaptation floor is found deficient, checking whether higher-load adaptations that depend on it are being performed on an unstable base. The ceiling check is mandatory — skipping it misses the most dangerous diagnostic finding.

The amount of mental effort a task places on working memory, which has a limited capacity. When a task exceeds working memory's capacity, neither the task itself nor anything else requiring thought gets adequate attention. This is why foundational fluency matters: a student who must consciously calculate basic arithmetic during an algebra problem is using working memory for the arithmetic that could be used for the algebra. The result looks like an algebra error but is actually a resource allocation problem. Cognitive load theory (John Sweller, 1988) is one of the primary theoretical foundations of the education framework.

The process of replacing an incorrect understanding with a correct one. Distinguished from gap-filling (adding knowledge that was absent) because it requires the student to first become dissatisfied with their existing belief before they will replace it. Research (Strike and Posner, 1992) shows that incorrect beliefs don't simply get erased by correct instruction — they get suppressed, and the old belief can reassert itself under stress or after a period without use. The most effective approach is to build the correct understanding alongside the existing incorrect one, then demonstrate the contradiction. Conceptual change is harder and slower than teaching from scratch, which is why wrong-schema is more dangerous than an empty schema.

The speed of the shortening phase of a muscle contraction — the "lifting" part of a movement, as opposed to the lowering phase (eccentric). Measured in meters per second. Used as a proxy for neuromuscular fatigue and form integrity. When concentric velocity drops more than 15% outside expected fatigue curves, the system flags Biomechanical Form Drift — a signal that movement quality has degraded beyond normal fatigue and structural breakdown is occurring.

The design constraint that all concepts introduced in a session must appear at every difficulty tier in the assignment. Not just the easy concepts on the easy problems and the hard concepts on the hard problems — every concept at every level. This produces the gradient assignment's self-calibrating effect: a student who can only do foundational problems still encounters harder problems on the same concepts, which provides diagnostic signal and productive exposure. A student who can do all levels gets the full picture. Violating this rule — putting certain concepts only on harder problems — breaks the self-calibrating effect and turns the assignment into a tiered test rather than a gradient diagnostic.

A framework from educational psychology (Reinhard Pekrun, 2006) explaining which emotions students experience in academic settings. Two dimensions predict the emotional outcome: perceived control (how much the student believes their effort affects their results) and perceived value (how much they care about the activity). High control + high value = enjoyment, engagement, hope. Low control = anxiety, helplessness. Low value = boredom. Low control + low value = passive resignation. Most students in most conventional classrooms are in the low-control, low-value zone most of the time because the curriculum is assigned and the pacing is set externally.

A complete cycle of: attempt → evaluate → correct → move on. In education: the student attempts to recall a concept, checks it against the correct answer, corrects any error on the spot, then moves to the next concept. In fitness: the system detects a form or load error, provides correction, and updates the session plan accordingly. The key word is "closed" — a loop that closes means no error survives uncorrected before the next attempt. An open loop — attempting many things and only checking at the end — allows errors to consolidate before correction happens, which is less effective and can be actively harmful (consolidating wrong patterns).

The point in the intake process where the student confirms their chosen subject of study. Treated as a commitment point — once set, changing it requires starting the full diagnostic over because the tier map, fluency battery, and reading baseline all reload for the new subject. The two-minute rule applies: if the student cannot settle on a course title within approximately two minutes of the selection conversation, the session ends and they return the next day. This protects against impulsive selection that wastes setup resources and produces a poorly calibrated starting profile.

A mathematical structure used to represent dependencies. "Directed" means the relationships have a direction (A requires B, not the other way around). "Acyclic" means there are no loops (you can't follow the arrows and end up back where you started). In the fitness engine, the DAG maps the nine primary physiological adaptations and their prerequisite relationships — aerobic capacity must be established before anaerobic capacity can be meaningfully trained; elasticity must be established before bone density training is safe; and so on. The DAG is the fitness system's equivalent of the education system's subject tier map.

Knowledge of facts and concepts — knowing that something is true or what something is. Distinguished from procedural knowledge (knowing how to do something). "The Pythagorean theorem states that a² + b² = c² in a right triangle" is declarative. "Being able to apply the Pythagorean theorem to find a missing side" is procedural. Both are necessary. Neither replaces the other. The education framework's emphasis on understanding why a procedure works (not just that it works) is an emphasis on integrating declarative and procedural knowledge rather than having one without the other.

Conditions that slow immediate, in-session learning but produce substantially stronger long-term retention and the ability to apply knowledge in new contexts. Coined by psychologist Robert Bjork. Examples: retrieving information from memory rather than rereading it (harder now, stronger later); spacing practice across time rather than cramming (lower immediate performance, much better retention at a week or month); interleaving different problem types rather than blocking by type (more confusing in the session, better transfer to novel problems). The framework's 85/15 calibration, retrieval practice protocol, and spaced practice schedule all implement desirable difficulties deliberately.

The assessment process that maps the current state of a person's knowledge or physical capacity. Not a test with a score — a structured process that produces a detailed profile of what is solid, what is partial, what is absent, and what is corrupted. In education: the intake diagnostic covers schema adequacy, foundational fluency, and reading comprehension. In fitness: the intake includes readiness score assessment, adaptation floor mapping, and physiological history. The diagnostic is the input that makes all subsequent calibration possible. Without it, the system is guessing.

The use of deliberately constructed wrong-answer options to identify which specific misconception a student holds, rather than only recording that they got the question wrong. In this framework, wrong answer options in multiple-choice items are not arbitrary — each option corresponds to the specific answer a student would produce if they applied a known misconception. When a student selects that option, the system logs not just "wrong" but "wrong due to misconception X." This makes wrong answers diagnostic rather than merely negative. A consistent pattern of selecting the same misconception-driven distractor across multiple items triggers a wrong-schema flag.

The central data structure of the fitness engine. The direct analog of the Dynamic Learning Profile in the education system. Records the current state of each physiological adaptation node, tracks wrong-adaptation patterns (the fitness equivalent of wrong-schema flags), logs recovery bottlenecks, and maintains the predictive fatigue trend. Updated after every session. Queried by the calibration engine before every workout to determine what can be trained, at what load, and with what constraints. It is not a fitness score — it is a detailed map of the body's current adaptive state.

The central data structure of the education system. Not a grade, score, or level — a detailed map of what a student currently knows, where their knowledge structure is sound, where it has gaps, where it is built on incorrect foundations, and what the calibration should be for their next session. Contains: schema floor per subject branch, wrong-schema flags, fluency bottlenecks with implicit/explicit distinction, language ceiling, calibration state, and session log. Updated after every session. Every instructional decision the system makes reads from the DLP. Equivalent to the fitness system's Dynamic Adaptation Profile.

A standardized number that measures how large a difference is, so that results from different studies using different scales can be compared. Cohen's d is the most common: d = 0.2 is small, d = 0.5 is medium, d = 0.8 is large. When a paper says retrieval practice has an effect size of d = 0.62, that means students who use retrieval practice perform 0.62 standard deviations better than students who don't — which translates roughly to moving from the 50th percentile to the 73rd percentile, for the same amount of study time. Effect sizes allow you to compare "how much does this help" across completely different types of interventions.

The development of connective tissue (tendons, ligaments, fascia) flexibility and load tolerance. Node 6 in the fitness DAG — a foundational node with no prerequisites. Elasticity is a prerequisite for neuromuscular performance (Node 1) and bone density adaptation (Node 7) because connective tissue must be able to handle the stress of high-intensity work before those adaptations can be safely loaded. Elasticity adaptation is slower than muscular adaptation, which is a common source of overuse injury — muscles get stronger faster than tendons can adapt to handle the new loads.

The philosophical position that knowledge comes from observation and experience — from what we can see, measure, and test in the physical world. In the philosophy of mathematics: the view that mathematical truths are true because we have observed them to hold in many cases (J.S. Mill's position). The framework rejects this view of mathematics: 2 + 2 = 4 is not true because we have counted many times and it keeps working out. It is true because of what 2, +, =, and 4 mean. Mathematical truth is necessary, not contingent on what the physical world happens to do.

Relating to knowledge — how we know things, what counts as knowledge, and what the limits of knowledge are. "Epistemic humility" means acknowledging the limits of what you know. "Epistemic emotions" are emotions that arise specifically in relation to knowing and understanding — curiosity, confusion, and the feeling of sudden insight (the "aha" moment) are epistemic emotions. The framework addresses epistemic emotions through its definition-first, derivation-over-memorization approach: when students understand why a relationship holds (not just that it holds), they experience insight rather than mere recognition, which is both more durable and more intrinsically rewarding.

The ability to deliberately retrieve, name, and apply a foundational operation on demand — to produce it consciously, out of context, and explain it when asked. Distinct from implicit fluency (using something correctly in context without being able to name it). A student who can use grammatical agreement correctly in sentences they write (implicit) may not be able to state the rule when asked (not explicit). Both are assessed separately in the diagnostic because the instructional response differs: a student with implicit-only fluency needs help making the tacit knowledge explicit; a student with neither needs the operation built from scratch.

The level at which feedback is directed. Hattie and Timperley (2007) identify four: task level (this answer is right or wrong), process level (this reasoning is sound or flawed and here is why), self-regulation level (here is how you might monitor your own thinking), and self level (you are smart/struggling/not a math person). The research is clear: task-level feedback is weakest for long-term learning; process-level feedback is strongest. Self-level feedback — whether positive or negative — produces near-zero learning effects and can actively damage the student's relationship to the subject. The framework's protocol requires process-level evaluation as the default.

How completely and precisely a method is actually implemented, as opposed to how it is theoretically supposed to work. High fidelity: every principle of the framework is being applied at the precision the evidence calls for. Low fidelity: some principles present, some absent, precision degraded by time pressure, class size, or resource constraints. Research consistently shows fidelity is the primary moderator of educational and fitness intervention outcomes — the same method at high fidelity produces substantially larger effects than at low fidelity. One key reason one-on-one tutoring outperforms group instruction by 2 standard deviations (Bloom, 1984) is that it enables higher fidelity implementation of good pedagogical principles. The adaptive platform is specifically designed to deliver high fidelity at scale.

The belief that one's performance in a subject reflects a fixed, unchangeable capacity — "I'm just not a math person." Research by Carol Dweck shows this is a learned belief, not an accurate description of fixed biology. It is produced by specific experiences: public failure in evaluative contexts, feedback at the self level ("you're just not getting this"), and incorrect correction of sound reasoning. Once established, fixed-ability attribution produces a self-fulfilling pattern: the student stops engaging in ways that would produce improvement, the lack of improvement confirms the belief, and the cycle reinforces itself. It has nothing to do with actual capacity.

In philosophy of mathematics: the view that mathematics is a formal game played with symbols according to rules, and that mathematical statements have no meaning beyond their positions in a formal system. On this view, mathematics is neither about the physical world nor about abstract objects — it's about valid manipulation of symbols within a specified system. The framework rejects strict formalism because it cannot explain why mathematics is so effective at describing physical reality, and because it makes the relationship between mathematical proof and mathematical truth puzzling. If truth is just "follows from the axioms," Gödel's incompleteness theorems become paradoxical.

The ability to perform prerequisite operations automatically, with minimal working memory cost, so that cognitive resources are available for higher-level thinking. In mathematics: instant recall of multiplication facts, fraction operations, order of operations. In language: instant recognition of grammatical structures. Fluency is not the goal — understanding is — but fluency frees the mental resources that understanding requires. A student who has to laboriously calculate 7×8 while working through algebra is using working memory for arithmetic that should be available for the algebraic reasoning. Foundational fluency is a prerequisite for reliable performance at higher levels, not a lower-order alternative to it.

Creating problems on demand using an AI language model to a precise specification, rather than pulling from a pre-built bank of stored problems. The specification is derived from the student's Dynamic Learning Profile and Student Context Profile: target concept, difficulty tier, misconception-driven distractors to include, whether round numbers should be used (when arithmetic is not the learning target), and the student's interest context for framing. Generative creation eliminates staleness (problems can't be memorized or shared), enables infinite variation, and allows the distractor set to be specifically calibrated to this student's known misconception history. The tradeoff: generated problems must be validated before delivery to ensure correctness.

An assignment structured as a deliberate difficulty slope — starting with foundational problems the student should get right 95-100% of the time, moving through growth-edge problems targeting approximately 85% accuracy, and ending with ceiling problems where the expected accuracy descends from 70% down to approximately 50%. The overall expected accuracy on a well-calibrated gradient assignment is 80-85%. When a student gets 100%, it signals that understanding has recently consolidated — a valuable diagnostic event. No student is expected to finish the assignment; the ceiling extends beyond any student's current capability so there is always productive work available.

A lecture structured as a deliberate difficulty slope — beginning at a foundational entry point accessible to all students in the room and increasing in complexity until only the most advanced students can follow. Distinguished from a standard lecture by its intentional design across the full ability range rather than aimed at a class average. The lecture has a floor (accessible to all), a middle (challenging the majority), and an upper register (extending beyond most students' current ability). Each zone serves a different function: the floor builds shared foundation, the middle develops the target concepts, the upper register provides the functions described under "upper-register exposure."

The variation in time between successive heartbeats, measured in milliseconds. Not the same as heart rate — HRV measures the irregularity of the intervals between beats, not how fast the heart is beating. Higher HRV generally indicates a well-recovered, parasympathetically dominant state (rest-and-digest, good for performance). Lower HRV indicates stress, fatigue, or sympathetic dominance (fight-or-flight, signals the need for recovery). HRV is one of the primary inputs to the fitness engine's readiness score and predictive fatigue trend. Measured most accurately first thing in the morning, before getting out of bed.

A teacher's awareness of how mathematical or subject-matter topics relate to each other across the full span of the curriculum — not just at the level being taught, but above and below it. Identified by Ball, Thames, and Phelps (2008) as a distinct type of knowledge required for teaching beyond simply knowing the subject. A teacher without horizon content knowledge cannot reliably evaluate whether a student's novel method is mathematically valid, because evaluation requires knowing how the method relates to mathematics above the current level. It is also required for building valid subject tier maps — identifying which concepts are genuinely load-bearing requires knowing what is built on top of them.

An increase in the size of muscle cells, producing larger muscles. Node 2 in the fitness adaptation DAG. Prerequisites: aerobic capacity (Node 4) and capillarization (Node 8). Without adequate aerobic base and blood supply to muscle tissue, hypertrophy training produces limited results and higher injury risk. Hypertrophy adaptation requires a meaningful training stimulus (sufficient volume and intensity), adequate protein intake, and sufficient recovery time between sessions targeting the same muscle groups.

See: Fidelity.

The ability to use a foundational operation correctly in context without being able to explicitly name, define, or articulate it. Most native speakers of a language have implicit fluency in grammatical rules they cannot state. Most people with adequate arithmetic experience have implicit fluency in order of operations — they apply it correctly in familiar contexts without being able to explain why. Implicit fluency is sufficient for many purposes and is often present when explicit fluency is absent. The diagnostic tests both layers separately because the instructional response differs: a student with implicit-but-not-explicit fluency needs help articulating what they already do; a student with neither needs the skill built from the ground up.

Two results proved by Kurt Gödel in 1931 that permanently changed the philosophy of mathematics. First theorem: any consistent formal system powerful enough to express basic arithmetic contains true statements that cannot be proved within that system — there are mathematical truths that no set of axioms can capture. Second theorem: no such system can prove its own consistency. These results show that mathematical truth is richer than any formal system we construct to describe it — there are always true mathematical facts that lie beyond what any given set of rules can reach. This is one of the strongest arguments for Mathematical Platonism: if truth were just "follows from the axioms," these results would be impossible.

Telling a student they are wrong when their reasoning is actually sound. The most damaging feedback error in the taxonomy, because it attacks the student's trust in their own reasoning process — the cognitive resource most necessary for independent mathematical thinking. Distinct from appropriate correction (telling a student their reasoning is wrong when it actually is wrong). The framework identifies six mechanisms of incorrect correction, of which the most severe is penalizing a valid novel method — when a student derives a correct answer through an unfamiliar approach and is marked wrong because the teacher doesn't recognize the method.

Mixing different types of problems or concepts within a practice session, rather than completing all problems of one type before moving to the next. Feels harder and produces lower immediate performance than blocked practice (doing all of one type first, then all of another). Produces substantially stronger long-term retention and transfer to novel problems. The reason: interleaving forces the student to identify which approach applies to each problem, rather than being able to pattern-match from the previous problem. This identification skill — recognizing problem type — is exactly what transfers to real-world situations where problem types aren't pre-labeled.

The dominant approach to student modeling in educational software (Corbett and Anderson, 1994, and its successors). Models the probability that a student has "learned" a specific skill, updated after each problem attempt. Predicts whether the student will get the next problem of that type right. Effective for tracking performance probability. Does not model the structure of the student's understanding — whether their internal model of the concept is correct, partial, or built on corrupted foundations. A student who memorizes a procedure that works on standard problem formats will show high knowledge tracing scores while having no genuine understanding of why the procedure works. Schema tracing (this framework) addresses what knowledge tracing misses.

The estimated point at which imprecise reading becomes the binding constraint on a student's applied performance, independent of their reasoning or computational ability. Stored in the Dynamic Learning Profile as a separate parameter from mathematical capability. A student can have a higher mathematical ceiling than their language ceiling allows them to demonstrate — their reading is the bottleneck, not their math. The diagnostic measures both constraints separately, and the system adjusts how problems are presented (complexity of language, lexical demands) in response to the language ceiling, not only how difficult the mathematics is.

The time between when a stimulus appears and when the person responds. In education: the time between a problem being displayed and the student's first committed response. Used as a proxy for cognitive effort — a fast correct answer suggests automatic retrieval; a slow correct answer suggests effortful processing even though the answer was right. The working memory cost of the deliberation is real regardless of the correct outcome. In fitness: response latency in movement execution, particularly concentric velocity, used to detect fatigue and form degradation. The 2-second threshold in the education system distinguishes automatic from effortful retrieval as a working heuristic subject to individual calibration.

The interference caused by words that carry both everyday meanings and precise domain-specific meanings, where the everyday meaning is more practiced and activates automatically before the technical meaning can override it. In mathematics: "product" (multiplication result vs. thing made), "mean" (average vs. unkind), "prime" (most important vs. divisible only by 1 and itself), "rational" (sensible vs. expressible as a ratio of integers), "domain" (territory vs. set of valid inputs for a function). A student who encounters these terms in a mathematical context and activates the everyday meaning first has experienced lexical ambiguity failure — a reading error that looks like a mathematical error. Distinguished from general prose comprehension failure by the mechanism and by the appropriate remediation.

A relationship is logically necessary if it could not possibly be otherwise given the definitions of the terms involved — it is true in every possible situation where those definitions apply, not just in the actual world. Mathematical relationships are logically necessary: 2 + 2 = 4 is not contingently true (true in this world but possibly false in another) — it is necessarily true because given what 2, +, =, and 4 mean, the relationship cannot be otherwise. This is why mathematical errors are always logical contradictions rather than empirical surprises, and why every mathematical error can in principle be traced back to the step where the reasoning departed from what the definitions require.

The deliberate inclusion of material the student has already mastered in every practice session. Approximately 15% of session time — enough to serve its purpose without dominating the session. Not filler: it serves two functions. First, it provides concrete evidence to the student that they know things, sustaining confidence through the harder material that follows. Second, it maintains fluency on foundational operations that would otherwise decay without use. The 15% figure is a time ratio, not a problem count ratio — mastery reinforcement problems are faster, so the time split and the problem count split are different.

The philosophical position that mathematical structures and their relationships exist independently of the physical world, human minds, and cultural convention — that they are discovered, not invented. Mathematical truths are necessarily true: they would hold in any possible universe with any physical laws, because they do not describe the physical world at all. They describe relationships between abstract objects that exist in what Penrose calls the Platonic mathematical world. Defended rigorously by Frege (1884), Gödel (1964), and Penrose (1989, 2004). The instructional implication: every mathematical error is a logical contradiction that can always be traced to its source and resolved, because the necessary relationships that were violated are always available in the definitions.

When the effect of one variable on another is explained (at least in part) by a third variable that sits between them in the causal chain. If A affects B, and it does so because A affects C and C affects B, then C mediates the relationship between A and B. In the framework: language comprehension affects word-problem performance, and part of this effect is mediated by mathematics vocabulary — language comprehension helps word-problem performance partly because it helps the student understand the specific vocabulary of the problem. The mediation relationship was identified in Lin (2021).

A statistical method that combines data from many separate studies (meta-analysis) and then examines the relationships between multiple variables simultaneously (structural equation modeling). This allows researchers to identify which variables are unique predictors of an outcome — which variables still matter after controlling for everything else. Lin (2021) used this method on 98 studies with 111,346 students to identify language comprehension as a unique predictor of word-problem performance after controlling for working memory, attention, nonverbal reasoning, processing speed, arithmetic, and mathematics vocabulary simultaneously. The combination of large sample size and rigorous statistical control makes this one of the most credible findings in the reading-mathematics relationship literature.

A skill about how to engage with the learning or training process itself, rather than a specific content skill. In education: knowing how to study effectively, how to distinguish genuine understanding from surface familiarity, how to use error as diagnostic information rather than as a verdict. In fitness: recognizing the difference between productive fatigue and overreaching, knowing when to follow data over feeling, self-debugging form and recovery decisions. Meta-skills are explicitly diagnosable in both systems — they have their own misconception IDs, detection methods, and system remediations. A student or athlete who has correct meta-skills learns more efficiently from every session than one who doesn't, independent of their current content or fitness level.

A specific incorrect belief about a concept that is not random but systematic — it produces predictable wrong answers because the person is consistently applying an incorrect rule or framework. In education: "the equals sign means 'the answer goes here' rather than expressing a balance relationship" is a misconception — it produces predictable errors in equation-solving. In fitness: "any soreness means I'm overreaching and should stop completely" is a misconception — it produces predictable training avoidance and undertraining. Misconceptions are diagnostically valuable because they are fingerprinted: the wrong answers they produce are specific and identifiable, allowing the system to detect them through distractor analysis rather than just recording incorrect responses.

A variable that changes the size or direction of an effect between two other variables. Implementation fidelity is a moderator of educational intervention outcomes: it doesn't cause outcomes directly, but it changes how large the effect of a given method is. Higher fidelity → larger effect. Lower fidelity → smaller effect. Same method, different fidelity, different result. Understanding which variables moderate an effect is essential for predicting whether a method that worked in a research study will work in a different real-world implementation.

A five-stage taxonomy of the types of errors students make on word problems, developed by M.A. Newman (1977). Stage 1: Reading error (cannot decode the problem text). Stage 2: Comprehension error (reads the words but misunderstands what is being asked). Stage 3: Transformation error (understands the problem but selects the wrong mathematical operation). Stage 4: Process skills error (selects the correct operation but executes it incorrectly). Stage 5: Encoding error (performs the correct calculation but records the result incorrectly). The first two stages are linguistic errors that occur before any mathematics begins. The framework uses this taxonomy to distinguish reading bottlenecks from mathematical ones in the diagnostic and feedback systems.

Improvements in the nervous system's ability to recruit and coordinate muscle fibers — becoming stronger and faster not by growing more muscle but by using existing muscle more efficiently. Node 1 in the fitness adaptation DAG. Prerequisites: elasticity (Node 6) and autonomic stability (Node 9). Neuromuscular adaptation is responsible for the rapid strength gains seen in the first weeks of training — before muscle growth has had time to occur. It is also how athletes maintain performance after detraining: neural efficiency is more durable than muscle size.

To translate an abstract concept into specific, measurable, implementable actions or criteria. "Productive struggle" is an abstract concept. Operationalizing it means specifying: what accuracy rate constitutes productive struggle (approximately 85%), what accuracy rate signals that difficulty is too high (below 60%), how difficulty is adjusted when the boundaries are crossed, and how often the calibration updates. Without operationalization, a principle can be agreed upon in theory but impossible to implement consistently. Much of the technical specification work in this framework is operationalization — turning pedagogical principles into precise engineering requirements.

The specific knowledge required to teach a subject effectively — distinct from simply knowing the subject. Identified by Lee Shulman (1986) and formalized by Ball, Thames, and Phelps (2008). Includes knowing the common misconceptions students hold about a topic, knowing multiple ways to represent and explain a concept, knowing which representations work best for different students, and knowing which concepts students find difficult and why. A mathematician who knows calculus deeply does not automatically have the pedagogical content knowledge to teach calculus effectively. A teacher who has only taken calculus once as an undergraduate likely lacks the depth of subject-matter knowledge required for the pedagogical content knowledge to develop fully.

See: Mathematical Platonism.

A weighted slope calculated from the last five days of HRV and acute-to-chronic workload ratio (ACWR) data, used to predict whether fatigue is building toward a problematic level before it becomes symptomatic. Not just current readiness — the direction and rate of change in readiness. A student who is currently at acceptable readiness but trending steeply downward should be treated differently from a student at the same current readiness but trending flat or upward. The predictive fatigue trend allows the system to intervene before overreaching occurs rather than after.

Something that must be in place before something else can be built on it. In education: equality and the balance principle must be understood before equation-solving can be reliably learned. In fitness: aerobic capacity must be established before meaningful anaerobic training can occur. Prerequisites are the load-bearing concepts and adaptations — their absence or corruption doesn't just leave a gap, it destabilizes everything built on top of them. Identifying true prerequisites (things that genuinely must come first) versus preferred sequences (things that are convenient to teach first but not structurally required) is the central challenge of building a subject tier map or fitness DAG.

Knowledge of how to do something — step-by-step procedures for performing a task. Distinguished from declarative knowledge (knowing that something is true). Being able to apply the quadratic formula is procedural. Knowing what a quadratic equation is and why the formula works is declarative. The framework argues that both are necessary and that understanding why a procedure works (declarative) produces more durable and transferable knowledge than memorizing the procedure alone (procedural without declarative). A procedure understood from its derivation can be reconstructed when forgotten; a memorized procedure that is forgotten is simply gone.

An instructional design approach (Manu Kapur) where students attempt to solve a problem before receiving instruction on how to do it. The "failure" is that students don't solve it correctly — but the attempt produces deeper learning when the correct method is subsequently taught, compared to receiving instruction first. Meta-analysis (Sinha and Kapur, 2021; N > 12,000) confirms the effect: problem-solving before instruction produces substantially stronger conceptual understanding and transfer than instruction before problem-solving. The mechanism: struggling with the problem before instruction activates prior knowledge, surfaces misconceptions, and primes the student to learn from the explanation in a way that passive reception of instruction doesn't achieve.

The experience of working on something genuinely difficult — at the edge of current capability — and making real progress. Distinguished from destructive frustration, where the difficulty so far exceeds capability that engagement collapses. In the productive struggle zone, the student is confused but engaging, pushing but not shutting down, close enough to the solution that the effort feels meaningful. This is where durable learning happens. The 85/15 calibration target is designed to keep students in this zone: approximately 85% accuracy on growth-edge problems means they are succeeding most of the time but genuinely working for it.

A composite score calculated from multiple physiological signals that represents the body's current capacity for training load. In the fitness engine: a weighted combination of HRV change, resting heart rate change, sleep efficiency, self-reported soreness, and hydration status. Scores above 80 indicate optimal readiness (all nodes unlocked). Scores 60-79 indicate level 1 restriction (lower-load nodes preferred). Scores 40-59 indicate level 2 restriction (minimal load, concentric-only work). Scores below 40 trigger a full recovery response (mandatory autonomic recovery work only). The weights are starting heuristics subject to empirical calibration — not derived constants.

The act of deliberately recalling information from memory, rather than re-reading or reviewing it. One of the most replicated findings in cognitive psychology (Abbott, 1909; Karpicke and Roediger, 2008; Yang et al., 2021). Students who practice retrieving material retain substantially more at one week and one month than students who spend the same time reviewing the material. The mechanism: attempting to retrieve something — even imperfectly, even with some forgetting — strengthens the memory trace in a way that passive review does not. In this framework, retrieval practice targets conceptual schemas and strategy-selection pathways, not rote procedural sequences — the goal is durable understanding, not memorized steps.

Temporary support that enables a learner to engage with material they couldn't yet handle independently, withdrawn as the learner develops the capacity to handle the task without it. Comes from Vygotsky's framework. The expertise reversal effect (Kalyuga et al., 2003) establishes an important constraint: scaffolding that helps novices can actively interfere with more advanced learners by introducing unnecessary steps into processes they can already execute efficiently. Good scaffolding is always designed to be removed — it is a means to independence, not a permanent accommodation.

An organized mental structure that groups related knowledge into a pattern, allowing it to be recognized and applied without effortful step-by-step reconstruction. By definition, schema is load-bearing structure — the relationships that hold the rest of the knowledge together, not peripheral facts or surface features. Experts organize their knowledge in schemas; novices organize it item by item. Learning, in cognitive terms, is largely the process of building and refining schemas. A partial or incorrect schema is more dangerous than an absent one because it feels like knowledge and resists correction — this is the wrong-schema problem.

The tier at which a student's schema is first found to be partial, incorrect, or absent. Instruction begins at or below this tier. Stored per subject branch in the Dynamic Learning Profile. Distinguished from a performance level or grade level — a student may perform at a 7th-grade level on standardized tests and have a schema floor at Tier 1 in a specific branch because they have been pattern-matching to familiar problem formats without genuine conceptual understanding. The schema floor identifies where genuine understanding begins, not where performance-level matching begins.

Tracking the health and architecture of a student's mental model — whether the internal structure of their understanding is correct, partial, or corrupted — as distinct from knowledge tracing, which tracks the probability that they will get the next problem right. Schema tracing asks: is the student's understanding built on correct foundations? Knowledge tracing asks: will they perform correctly on the next item? A student who memorizes procedures that work on standard formats will score high on knowledge tracing while having low schema health. Schema tracing catches what knowledge tracing misses — which is precisely the case that most needs intervention: confident performance built on wrong foundations.

A person's belief in their own ability to succeed at a specific type of task. Not general confidence — domain-specific belief in capability. Research (Bandura, 1997; Honicke and Broadbent, 2016) shows self-efficacy is one of the strongest predictors of academic achievement. Crucially, it is not a fixed trait — it is produced by specific conditions, primarily by mastery experiences: accumulated evidence that effort in this domain produces results. The system's calibration is designed to produce these mastery experiences continuously: problems calibrated so that effort does produce correct answers most of the time, with the evidence visible to the student in real time.

The mental representation a reader constructs of the scenario being described in a text. In mathematics word problems, building the situation model — understanding what real-world situation the problem describes — is a comprehension task that happens before any mathematical operation. The student must first understand what is being described, what is being asked, and what the relationships between quantities are, before they can begin setting up the mathematics. Errors that occur at the situation model stage produce wrong equations even when the student can solve equations correctly. Identified by Kintsch and Greeno (1985) as the foundational mechanism of word-problem solving.

Mathematical knowledge unique to the work of teaching — distinct from the mathematical knowledge shared with anyone who knows the subject. Includes: knowing how to analyze the validity of alternative solution methods, identifying the mathematical source of student errors, and explaining why a procedure works rather than just that it works. Identified by Ball, Thames, and Phelps (2008) as a distinct domain of mathematical knowledge for teaching. A mathematician with deep subject knowledge but no teaching experience may lack specialized content knowledge. A teacher who has only taken one course beyond what they teach almost certainly lacks it. Required for avoiding the incorrect correction mechanisms of penalizing valid novel methods and teaching arbitrary memorization where derivation is available.

Distributing practice across time rather than concentrating it in one session. A student who practices material across three sessions spread over a week retains substantially more at one month than a student who spends the same total time in one massed session. The mechanism: memories consolidate and strengthen when retrieved after a delay — when some forgetting has begun. Immediate re-study before forgetting starts produces less consolidation than studying the same material after an interval. The framework implements spaced practice at two levels: the between-session recall protocol spaces retrieval events within each day, and the program-level cycling of topics spaces content across the full program duration.

A measure of how spread out values are in a dataset. In educational research: a way of expressing how far a student's performance is from the average. If the average test score is 70 and the standard deviation is 10, then a student one standard deviation above the mean scored 80, and a student two standard deviations above scored 90. Bloom's 2-sigma problem (1984) found that one-on-one tutoring produces a two-standard-deviation advantage over conventional instruction — meaning the average tutored student would outperform 98% of conventionally instructed students. Standard deviations allow comparison across different tests and different populations.

A structured record of who the student is beyond their academic performance — their interests, hobbies, work history, aspirational goals, media preferences, life context, and available materials. Distinct from the Dynamic Learning Profile, which records academic state. The SCP drives problem framing and example selection — a student who works in construction should see measurement problems framed around construction contexts; a student who plays poker should see probability problems framed around card games. The SCP makes the curriculum feel built for this specific person rather than for an anonymous average student. Collected through the intake interview. Updated when new context is learned.

A hybrid mental model mixing correct and incorrect elements, formed when a student receives detailed instruction they only partially understand and fills the gaps with plausible-seeming incorrect assumptions. Identified by Vosniadou (1994) as a primary failure mode of conventional instruction. Synthetic models feel like knowledge — the student is not aware they are wrong — and resist correction precisely because they are internally consistent enough to produce correct answers on familiar problem types. More dangerous than a simple wrong answer or an empty gap, because the student's confidence in the incorrect framework makes them resistant to correction and because the incorrect framework can contaminate subsequent learning built on top of it.

Data automatically collected by sensors during or around a fitness session. In the fitness engine: HRV from a wearable device, resting heart rate, skin temperature, blood oxygen saturation (SpO2), and signal quality indicators (sensor contact quality, motion artifact index). Telemetry is the primary input to the readiness score and predictive fatigue trend. Signal sanitization is critical: sensor readings contaminated by poor contact or excessive motion are discarded rather than averaged in, because corrupted signals produce worse decisions than no signal.

See: Retrieval practice.

A structured diagram of a subject's dependency structure: which concepts are genuinely load-bearing (cannot be understood without them), which build on those, and what the logical sequence is from foundational to advanced. Like a blueprint that identifies which walls are load-bearing before construction begins. Built by domain experts with horizon content knowledge — someone who understands not just the current level but what is built above it. Used by the diagnostic system to assess where a student's schema is solid, where it has gaps, and where it is built on incorrect foundations. Not a curriculum outline: peripheral concepts that are useful but not structurally necessary are excluded.

The ability to apply knowledge learned in one context to a new, different context — especially one that looks different on the surface but has the same underlying structure. Transfer is the ultimate goal of education and the primary thing that memorized procedures without understanding fail to produce. A student who understands why a method works can apply it in novel situations; a student who memorized the method without understanding can only apply it to situations that match the template they memorized. The framework's emphasis on schema retrieval over procedural recall, and derivation over memorization, is specifically aimed at producing transfer-capable understanding.

The fitness system equivalent of the education system's Student Context Profile. Records who the person is beyond their physiological data: their training goal (general health, sport performance, body composition, rehabilitation), life context (occupation, stress load, sleep constraints), injury history, available equipment, and anything else that affects how training decisions should be framed and explained. Without the TCP, the system can make physiologically correct decisions but explain them in language that doesn't connect to what this specific person cares about. With the TCP, the same decision is explained in terms that make sense for a 45-year-old construction worker versus a 22-year-old competitive athlete.

Whether a measurement or assessment actually measures what it claims to measure. A valid reading assessment measures reading comprehension, not general intelligence or prior knowledge of the topic. A valid schema diagnostic measures the health of a student's conceptual understanding, not their ability to perform on familiar problem formats. Many standard assessments have questionable validity for the purposes the framework cares about: a student can score at grade level on a standardized mathematics test while having a Tier 1 schema floor, if they have pattern-matched to standard problem formats without genuine understanding.

See: Zone of Proximal Development. Lev Vygotsky (1896–1934) was a Soviet psychologist whose work on the social and cultural nature of cognitive development remains foundational in educational psychology. His most cited concept in this framework is the ZPD, but his broader insight — that learning is fundamentally social and that cognitive development happens through interaction and language — also underlies the framework's emphasis on Socratic dialogue as a distinct instructional element separate from retrieval practice.

The cognitive system that holds and manipulates information during active thinking. Has a strictly limited capacity — roughly 4-7 items for most people. When a task exceeds this capacity, neither the task nor anything else requiring thought gets adequate attention. This is why foundational fluency is essential: a student who must consciously compute basic arithmetic during algebra is using working memory for arithmetic that should be available for the algebraic reasoning. The result looks like an algebra error but is a resource allocation problem. Many apparent failures of understanding in education are actually failures of working memory management — the task demanded more simultaneous mental holding than the student's working memory could support.

A diagnostic marker indicating that a person appears to have higher-level capability resting on a corrupted lower-level foundation. In education: a student who demonstrates apparent Tier 3 knowledge that is actually built on an incorrect Tier 2 schema — they have the language and surface familiarity of a higher level, but the internal model is wrong. In fitness: an athlete who appears to be performing higher-load training adequately while underlying prerequisite adaptations are deficient — the performance is a compensatory pattern, not genuine capability. The wrong-schema flag changes the instructional or training sequence from gap-filling (building forward) to conceptual change or foundation remediation (fixing what is wrong before continuing).

Vygotsky's term for the range of tasks a learner cannot yet complete independently but can engage with meaningfully given appropriate challenge or support. Below the ZPD: too easy, no development happens. Above it: too overwhelming, no productive engagement. Within it: genuine learning. One way to understand the framework's calibration system is as a continuous attempt to keep each student working within their ZPD — which is specific to that student and changes as they develop. The framework's productive struggle zone (approximately 85% accuracy on growth-edge problems) is the operational specification of the ZPD for this system.