The irreducible epistemic atoms underlying the curriculum. 4,828 atoms across 8 types and 2 molecules
Transitive relationship: a relationship type where if entity A relates to entity B and entity B relates to entity C, then entity A also relates to entity C, with the relationship type being consistent across all links in the chain
Transitive inference: the automatic cognitive operation of computing implied relationships from chains of explicit relationships without conscious effort, where the relationship type must support transitivity for the inference to be valid
Redundancy: the presence of multiple independent paths between critical nodes in a system, where each path does not share a failure mode with the others, such that the system remains connected and functional even when some paths fail.
Bottleneck: a structural feature of a system — a node or connection through which disproportionate flow must pass, creating a constraint that governs the throughput of the entire system
Articulation point: a node whose removal disconnects the graph — meaning parts of the network that were previously connected can no longer reach each other
Epistemic action: a type of action that changes the world in order to simplify a cognitive task, thereby enabling more efficient thinking rather than simply moving physically closer to a goal
Cognitive artifact: a tool or representation that reorganizes cognitive processes and actively participates in thinking rather than passively storing information
Nesting: the universal property of meaning organization where any concept contains sub-concepts and belongs to super-concepts, forming a recursive hierarchical structure that repeats at every scale and enables complex systems to organize themselves through stable intermediate forms
Recursive structure: a nested system defined in terms of itself where a single rule generates infinite depth, enabling the same structural pattern to operate at every level of organization without requiring different rules for each scale
Lattice: a directed acyclic graph where every node can have multiple parents and every pair of elements has both a unique least upper bound (join) and a unique greatest lower bound (meet), representing the multidimensional nature of knowledge where concepts genuinely belong to multiple categories simultaneously
Polyhierarchy: a design principle in knowledge organization where concepts genuinely belong to multiple parent categories simultaneously, as opposed to tree structures that force concepts into single parent relationships, preserving retrieval paths from multiple access dimensions
Override: an explicit, deliberate annotation or modification that modifies or replaces a specific inherited property from a parent node, indicating that while the child item inherits from the parent, it does not conform to that specific inherited characteristic due to its unique context or requirements.
Progressive disclosure: the information architecture principle that structures content in layers where each level provides sufficient context to decide whether to drill deeper, ensuring that overview is presented first and detail is available on demand rather than forcing users to consume everything at once
Containment: the structural relationship where an item exists as part of a parent entity, shares the parent's lifecycle, and cannot be accessed independently.
Hierarchical thinking: the fundamental cognitive capability to organize things into nested levels and move between those levels with purpose, operating as a core architectural feature of human cognition and manifesting in neural hierarchy, linguistic structure, developmental milestones, and system design principles
Personal experiment: a specific, actionable test designed to validate or invalidate a schema through controlled observation and measurement
Schema test: a structured action designed to evaluate a schema by making a falsifiable prediction, taking defined action, observing results, and specifying revision triggers, where the test must have genuine risk of falsification to be valid
Indirectly validatable schema: a schema that cannot be confirmed or disconfirmed through a single direct observation but can be validated through convergence of multiple independent indicators from different domains
Epistemic confidence: the level of confidence that tracks the evidential basis for a belief, as opposed to psychological confidence which tracks how central the belief feels to one's identity and worldview
Schema validation: the systematic practice of testing beliefs against evidence, welcoming disconfirmation, and refusing to protect comfortable models from uncomfortable data
Emotional tolerance for schema change: the capacity to endure the physiological discomfort and emotional resistance that arises when confronting contradictory evidence about deeply held beliefs, without immediately resolving it through dismissal or rationalization
Incremental schema revision: the practice of making small, frequent adjustments to mental models rather than large, rare overhauls, which preserves functional structure while correcting dysfunctional elements and operates within working memory constraints
Schema debt: the compounding liability that accumulates when a mental model is known to be inaccurate but is not updated, with interest payments manifesting as suboptimal decisions, wasted effort, and degraded performance in the domain the schema describes
Trigger conditions: specific, observable signals or thresholds that automatically prompt schema review regardless of subjective judgment, defined prospectively before problems are suspected and designed to detect schema degradation early