AchiralAchiral

Concepts · Humans and machines

Concepts

Foundational concepts for studying memory as a cognitive, behavioral, organizational, and computational phenomenon.

Summary Capsule

Explore ACT-R, retrieval, agent memory, and organizational context as related but distinct ideas.

This collection explains what cognitive architectures can contribute to AI-system design without treating engineering patterns as models of the human mind.

Watercolor illustration of one person naming a seedling for another, with blank tags and a flowing ribbon of speech marks.
1.9k reads

Canonical Memory Glossary

An up-to-date glossary of memory terms across science, computing, AI, and technology.

Watercolor illustration of chocolate chips being added to cookie dough when the current recipe step calls for them.
840 reads

What Is ACT-R Memory Architecture?

ACT-R is Carnegie Mellon’s cognitive architecture for memory and action, using chunks, buffers, activation, and production rules.

Watercolor illustration of a traveler choosing between paths toward Mount Fuji, with one sunlit path selected.
0 reads

What Is SOAR Cognitive Architecture?

SOAR is a cognitive architecture for goal-directed problem solving. It selects operators over working memory and creates substates when selection stalls.

Watercolor illustration of a pilot using a flight log, checklist, map, and current runway conditions before flying a small plane.
610 reads

ACT-R as Agent Memory

ACT-R-style agent memory uses goals, active context, retrieved chunks, and production rules to help AI agents choose what to do next.

Watercolor illustration of an air traffic controller using flight strips and an airfield map to coordinate planes on the ground and in the air.
640 reads

What Is a Memory-Native Organization?

A memory-native organization has a shared way to remember what happened, why it mattered, who can use it, and when it should affect the next action.

Watercolor illustration of a stone wine cellar with one selected bottle on a tasting table, bottles aging on shelves, a thermometer, and an open notebook.
720 reads

RAG vs AI Memory

RAG retrieves context for a response; AI memory governs what information persists, changes, and is reused over time.

ACT-R Architecture Chapters

Read the ACT-R memory sequence

Start with ACT-R
  1. 1What Is ACT-R Memory Architecture?A plain introduction to ACT-R memory architecture: goals, buffers, chunks, production rules, activation, and action selection.
  2. 2Ebbinghaus and ForgettingHow nonsense syllables, savings scores, and the forgetting curve made memory measurable.
  3. 3Chunks and Declarative MemoryHow ACT-R represents facts and experiences as retrievable chunks with graded activation.
  4. 4Production RulesHow procedural memory selects actions through condition-action rules and utility estimates.
  5. 5ACT-R and AI MemoryHow cognitive architecture can inspire AI systems without claiming that software is conscious.

SOAR Architecture Chapters

Read the SOAR problem-solving sequence

Start with SOAR
  1. 1What Is SOAR Cognitive Architecture?A simple guide to SOAR cognitive architecture: states, operators, goals, preferences, impasses, and learning.
  2. 2Working Memory in SOARHow SOAR keeps the current problem state readable through working memory elements.
  3. 3Operators in SOARHow SOAR represents possible next steps as operators that change the current state.
  4. 4Preferences in SOARHow preferences help SOAR choose between matching operators.
  5. 5Impasses in SOARHow SOAR handles stuck points by creating subgoals instead of guessing.
  6. 6Chunking in SOARHow SOAR learns new production rules from solved subproblems.
  7. 7Semantic and Episodic Memory in SOARHow SOAR separates general knowledge from remembered experience.
  8. 8SOAR and AI AgentsWhat agent memory can borrow from SOAR without pretending software has a human mind.

Concept Article Index

Latest concepts and chapters

23 published resources

  1. SOAR and AI AgentsWhat AI agents can borrow from SOAR: explicit state, operator selection, preferences, impasses, subgoals, learning, and separate memory systems.SOAR chapter
  2. What Is SOAR Chunking?SOAR chunking creates a production rule from a subgoal result so similar situations can be handled faster without repeating the same problem solving.SOAR chapter
  3. Impasses and Subgoals in SOARSOAR creates substates when it cannot choose or apply an operator. These impasses turn confusion into a smaller problem to solve.SOAR chapter
  4. Operators in SOARSOAR solves problems by proposing, selecting, and applying operators to the current state.SOAR chapter
  5. Preferences in SOARSOAR uses preferences and a decision procedure to choose one operator from the candidates available in working memory.SOAR chapter
  6. SOAR Semantic and Episodic MemorySOAR semantic memory stores general knowledge, while episodic memory records experience over time. See how each store supports retrieval and behavior.SOAR chapter
  7. What Is SOAR Working Memory?SOAR working memory represents the current situation as WMEs: states, operators, objects, and other context used during the decision cycle.SOAR chapter
  8. What Is SOAR Cognitive Architecture?SOAR is a cognitive architecture for goal-directed problem solving. It selects operators over working memory and creates substates when selection stalls.SOAR chapter
  9. What Is Retrieval Augmented Generation and How It WorksRAG helps a model check outside sources before it answers. Learn how retrieval augmented generation works and why it is not memory.Concept article
  10. ACT-R and AI MemoryWhat AI memory systems can borrow from ACT-R: separate facts from actions, use buffers for current context, retrieve by activation, and keep human review around learned behavior.ACT-R chapter
  11. Declarative vs Procedural Memory for AI AgentsDeclarative memory stores what an AI agent knows. Procedural memory stores how it acts. Reliable agents need both facts and learned execution rules.Concept article
  12. How Emergent Memory Systems Remember and ForgetAn ACT-R-inspired emergent memory system decides what to remember by judging usefulness, scope, trust, permissions, freshness, and conflict.Concept article
  13. Production Rules and Procedural MemoryHow ACT-R uses production rules to choose actions from the current goal, buffers, and retrieved memory.ACT-R chapter
  14. Chunks and Declarative MemoryHow ACT-R encodes facts and experiences as chunks, retrieves them with cues, and uses activation to decide what comes back.ACT-R chapter
  15. Ebbinghaus Forgetting Curve | Definition & SavingsThe Ebbinghaus forgetting curve shows how memory fades over time, why relearning can be faster, and why AI memory needs decay.ACT-R chapter
  16. Canonical Memory GlossaryAn up-to-date glossary of memory terms across science, computing, AI, and technology.Core concept
  17. Memory vs StorageStorage preserves information; memory preserves the influence of experience on future behavior.Concept article
  18. Organic AI MemoryOrganic AI memory forms from activity over time, strengthening useful context and allowing irrelevant memories to weaken.Concept article
  19. What Is Memory in AI?AI memory lets experience change future behavior: form, encode, activate, reinforce, and decay context across interactions, not just store text for one prompt.Concept article
  20. ACT-R as Agent MemoryACT-R-style agent memory uses goals, active context, retrieved chunks, and production rules to help AI agents choose what to do next.Core concept
  21. What Is ACT-R Memory Architecture?ACT-R is Carnegie Mellon’s cognitive architecture for memory and action, using chunks, buffers, activation, and production rules.ACT-R chapter
  22. What Is a Memory-Native Organization?A memory-native organization has a shared way to remember what happened, why it mattered, who can use it, and when it should affect the next action.Core concept
  23. RAG vs AI MemoryRAG retrieves context for a response; AI memory governs what information persists, changes, and is reused over time.Core concept

Memory and continuity in AI systems

Memory is not a single technical object. In cognitive science it concerns mechanisms of learning and recall; in AI engineering it often refers to persistent context, retrieval, or state management. This collection separates those meanings before asking how they can inform practical systems.

Achiral applies these ideas to a shared operational-context layer for AI systems. That layer is an engineering design, not a model of human memory or cognition.

PerspectiveCentral question
Cognitive architectureWhat must be represented, retrieved, activated, and acted on for a system to behave with continuity rather than isolated response?
Behavioral adaptationHow does repeated experience change future choices, priorities, habits, and interventions without requiring every rule to be restated?
Organizational memoryHow do decisions, context, exceptions, and handoffs become durable shared state instead of disappearing across tools and conversations?
Computational retrievalWhen should a system search, rank, reinforce, forget, or withhold context so memory remains useful instead of becoming noise?
Governance and safetyHow can compounding memory stay permissioned, inspectable, reversible, and aligned with human review as it begins to shape action?