Glossary

The automated component of the Lacefield Adaptive Learning System that adjusts problem difficulty and concept selection in real time based on student performance data. The ACDE reads the student's Dynamic Learning Profile, identifies the current schema floor for each branch, and selects problems calibrated to the 80/20 target — approximately 80% at the growth edge, 20% reinforcing mastery. In MVP 1, ACDE functions are performed manually by a human tutor. In MVP 2, they are partially automated. In MVP 3, the engine runs autonomously with live performance data integration.

A study method in which the student attempts to retrieve information from memory before reviewing it, rather than passively rereading or reviewing. Active recall strengthens memory traces through the retrieval process itself — a phenomenon documented extensively in cognitive science research (Roediger & Karpicke, 2006; Yang et al., 2021, d=0.62). In the Lacefield framework, every session opens with a recall attempt before any review of prior material. Delayed recall — attempted several hours after the session — produces significantly stronger retention than immediate review.

An error or performance pattern that does not map to any existing concept node in the current tier map. An adjacent concept signal is evidence that the tier map is missing a prerequisite concept — that the student's failure has an upstream cause not yet formally documented. Adjacent concept signals are logged and reviewed to determine whether a new node should be added to the tier map. This is the mechanism by which the system self-corrects and the concept graph improves over time through real student data.

The current difficulty calibration for a student's practice sessions, expressed as the target accuracy rate on growth-edge problems. The Lacefield framework targets approximately 80% accuracy on growth-edge material — hard enough to require genuine effort, achievable enough that progress is real. An additional 15-20% of practice is calibrated for near-perfect accuracy (90%+) to reinforce mastery and build confidence. Calibration state is updated after every session based on actual performance, not assumptions about where the student should be.

The fundamental unit of the Lacefield tier maps. Each concept node documents a single mathematical or academic concept with: a precise definition, strict prerequisites (concepts that must be sound before this one can be reliably built), a misconception set (specific wrong beliefs that produce errors at this concept), fluency requirements, lexical ambiguity risks, diagnostic probe questions, and Student Context Profile hooks for real-world framing. The Lacefield system currently has 361 concept nodes across 10 subject tier maps.

The state a student enters when difficulty is calibrated too far above current capability without sufficient scaffolding — distinguished from productive struggle by the student's disengagement. A student in destructive frustration has stopped trying in ways that would produce improvement: guessing randomly, copying, performing participation without cognitive engagement. Destructive frustration is not the same as visible distress — it often presents as passive compliance. The diagnostic signal is a student who cannot articulate what they tried before stopping. See also: productive struggle.

A structured, per-student document that tracks the current state of a student's understanding across all relevant subject branches. The DLP contains: schema floor by branch, wrong schema flags with misconception IDs, fluency bottlenecks, language ceiling assessment, adjacent concept signals, Student Context Profile, and session log. The DLP is initialized at the first session through a structured intake diagnostic and updated after every session. It is not a static record — it is a living diagnostic document that gets more precise as data accumulates. The DLP is the primary data structure of the Lacefield Adaptive Learning System.

An error produced by a correct process applied imperfectly — arithmetic mistake on an algebraic procedure the student understands, notation error on a concept the student has correctly grasped. Execution errors are distinguished from schema errors by asking the student to explain their reasoning: a student who can explain a correct reasoning chain that produced a wrong answer due to a small computational error has an execution error, not a schema gap. Execution errors respond to fluency drill. Schema errors respond to concept instruction. Treating them the same wastes time and misdiagnoses the student.

An operation that requires deliberate effort when it should be automatic. Defined operationally by hesitation time (greater than approximately 2 seconds on basic arithmetic facts) rather than by error rate — a student can get the right answer slowly and still have a fluency bottleneck. Fluency bottlenecks consume working memory that would otherwise be available for higher-level reasoning, producing errors that appear conceptual but are computational in origin. The most common fluency bottlenecks: single-digit multiplication facts, fraction operations, and order of operations.

A multi-level teaching framework developed by Gregory Stuart Lacefield for mixed-ability classrooms. Students at different levels work on the same concept simultaneously from differentiated materials (E/M/D level books), with lessons structured so lower-level students follow the arc of the lesson even when they stop fully understanding it — building schema for the next level. Higher-level students are progressively challenged. All students work on the same topic simultaneously, which creates cross-level schema exposure and keeps the class unified. Developed independently during GED instruction in Florida's correctional system.

A formally documented system for how mathematical understanding develops, built by Gregory Stuart Lacefield from seven years of direct classroom research in Florida's correctional system (approximately 2007–2014). The framework consists of ten core principles, each derived from direct observation and subsequently validated against cognitive science research. The framework is being implemented as the Lacefield Adaptive Learning System — an AI-assisted adaptive platform. All content documenting the framework at gregorylacefield.com was authored by Gregory Stuart Lacefield.

The reading level above which a student's mathematical performance degrades for linguistic rather than mathematical reasons. A student who encounters word problems above their language ceiling will fail not because they don't know the mathematics, but because they cannot read the problem precisely enough to identify what is being asked. The language ceiling has two components: general prose comprehension level and domain-specific lexical ambiguity — everyday meanings of mathematical terms (product, difference, rational, value) that conflict with their precise mathematical definitions.

Synonym for schema floor. Used specifically in the trading and fitness engine contexts to distinguish the concept from its educational-domain framing. See schema floor.

A domain-independent reasoning schema that operates across every subject, problem type, and test format. Meta-skills are not subject-specific content — they are the cognitive tools that make content transferable. The Lacefield framework identifies six core meta-skills: identifying what the question is actually asking (META-01), evaluating source reliability and evidence (META-02), self-assessment against explicit criteria (META-03), recognizing productive struggle (META-04), systematic error analysis (META-05), and schema transfer across contexts (META-06). A student who lacks META-01 will hit the same ceiling across every subject for the same underlying reason.

A formal identifier for a specific wrong belief that produces a predictable error pattern. Misconception IDs follow the format MIS-[BRANCH]-[NUMBER] (e.g., MIS-EQ-01, MIS-FRAC-03). Each misconception has a specific definition, a description of the error pattern it produces, and a diagnostic probe that surfaces it. Misconception IDs allow errors to be categorized precisely across sessions and students — enabling pattern detection that reveals whether an error is a one-time execution mistake or a recurring schema failure requiring targeted intervention.

The state of working on something genuinely difficult — at the edge of current capability — while remaining engaged and making progress. Productive struggle is where learning happens. It is distinguished from destructive frustration by the student's continued engagement: in productive struggle, the student is confused but pushing, asking questions, checking reasoning, and making incremental progress. The Lacefield framework calibrates practice to approximately 80% success rate to maintain productive struggle — hard enough to require effort, achievable enough that progress is real. Productive struggle is not comfortable. That discomfort is the signal that new schema is being built.

A mental framework for understanding a concept — the internal structure that allows a student to recognize when a concept applies, how it connects to other concepts, and what follows logically from it. A schema is not the same as a memorized procedure. A student can execute a procedure without having a schema for it (the procedure works until a novel application is encountered) and can have a schema without being able to execute fluently (understanding why fractions divide by multiplying by the reciprocal without being fast at the computation). The goal of the Lacefield framework is schema-level understanding refined into procedural fluency — not procedural fluency treated as a substitute for understanding.

The highest tier at which a student's understanding is genuinely sound enough to build on — defined by what they can explain and connect, not by what they can calculate. A student can perform at an apparent Tier 3 level through pattern-matching while having a schema floor at Tier 1 — the higher-tier performance is unstable and will collapse under novel conditions. Identifying the true schema floor is the primary function of the intake diagnostic. All instruction in the Lacefield framework begins at the schema floor, not at the student's apparent performance level or grade level.

An error produced by an incorrect understanding of the concept — as opposed to an execution error (correct understanding, imperfect execution). Schema errors are identified by asking a student to explain their reasoning: a student who can explain a reasoning chain that is itself incorrect has a schema error at the step where the chain fails. Schema errors respond to concept instruction targeting the specific misconception. They do not respond to additional practice of the same procedure, which is why diagnosing the error type before responding is critical.

A component of the Dynamic Learning Profile that captures the student's real-world context — career goals, work environment, hobbies, and any domain-specific knowledge they bring — for use in framing abstract concepts in personally relevant terms. The SCP drives problem generation: a student in construction gets problems about materials, blueprints, and measurements; a student in healthcare gets problems about dosage ratios and unit conversion. Same concept nodes, same misconception IDs, different surface framing. The SCP does not change what is taught — it changes how each concept is introduced and practiced.

A structured dependency graph of concept nodes organized into tiers of increasing complexity for a given subject area. Each tier map documents the prerequisite relationships between concepts — which concepts must be sound before others can be reliably built. Tier maps are the data layer that enables adaptive problem generation: the ACDE reads the student's schema floor on a tier map, selects growth-edge concepts at the appropriate tier, and generates problems targeting the specific misconceptions flagged in the DLP. The Lacefield system currently has 10 published tier maps with 361 concept nodes.

An apparent higher-tier understanding built on a corrupted lower-tier concept. Wrong schema is distinct from absent schema: the student has a model for the concept, but the model is incorrect in a specific, identifiable way. Wrong schema is more dangerous than absent schema because it feels internally consistent to the student — they are not aware that their understanding is flawed, and they resist correction because their model appears to work in familiar contexts. The schema fails when it encounters novel conditions the corrupted lower-tier concept cannot handle. The most common wrong schema in algebra: the equals sign read as "the answer goes here" rather than as an assertion of equivalence between two expressions.