Encoding & salience
Not everything that happens is worth keeping. Something has to decide.
a new day starts — moments incoming…
A field guide to memory — human & artificial
A language model is brilliant, and it forgets everything the moment you close the tab.
Its context window is working memory — a fixed buffer where thinking happens. When the conversation ends, it’s gone. In human terms that’s severe anterograde amnesia: intact intelligence, no ability to form a lasting memory. Solving it means borrowing from the one system that already did — the brain. This is a guide to how memory actually works, how AI is trying to copy it, and where the ground is still open.
Explore the mechanismsThe word for a memory
An engram is the physical trace a memory leaves in the brain — a term coined by Richard Semon in 1904Semon 1904, when it was pure theory. A century later, Susumu Tonegawa’s lab found the cells: they tagged the neurons active during an experience, then reactivated them with light and watched the memory returnTonegawa 2012–2015. Engrams are real, locatable, and switchable. This guide is organized around the mechanisms those traces are built from — one specimen at a time. For a plain explanation of a term, with the experiments behind it, see How memory works.
The spine
Eleven mechanisms
Each specimen reads the same four ways: how the brain does it, the model that formalizes it, how AI implements or fails it, and the map of who has claimed it. Where the map runs thin, there’s open ground.
Showing 11 of 11 mechanisms
Emotional modulation
The moments that mattered are the ones you keep. Feeling is the encoder's thumb on the scale.
lunch with Dana
- eventshe tells you about her new job
- feltan ordinary catch-up
the next six months
- starting strength
- fades to half in
- 5 weeks
small talk ▸ starts weak, fades fast · six months later: barely a trace
Decay & forgetting
Forgetting is not the failure of memory. It is one of its functions.
Contested — Whether forgetting is true decay or retrieval interference is a century-old, still-unsettled question.
a book you read once — what’s left
- that perfect line you underlined100%
- the argument of chapter one100%
- the author's name100%
- how it made you feel100%
- the gist — it changed how you think100%
day 0 · curve at 100% · everything is still fresh
Interference
New memories crowd out old ones. Old ones distort new ones. They compete.
monparked: level 3, by the north stairs0.92
tuedentist, 2:00 pm0.94
two unrelated memories · both recall cleanly
Retrieval & strengthening
Every time you remember something, you change it — and usually make it stronger.
every recall strengthens · every neglect fades — click a memory to keep it alive
Temporal context & association
Recalling one thing pulls its neighbors with it — everything you learned in the same hour.
morning
midday
evening
recalled that song on the radio ▸ surfaced with it: 12:30p (0.74) · 1:05p (0.67) — the morning and evening stayed dark
Prospective memory
Remembering to do something later — the memory that fires itself at the right moment.
○ armedpick up milk on the way home
an intention is armed — no timer, no alarm; it’s waiting for a moment that matches
Consolidation
Memories aren't saved when they're made. They're rebuilt, offline, while you sleep.
Contested — How consolidation works long-term is actively debated — standard systems consolidation vs. multiple-trace theory.
- 9:10acoffee with Sarah — she's stressed about the launch
- 11:30ahallway: Sarah again, still the launch
- 12:40pthe taqueria was packed
- 3:15pSarah's launch slipped a week
- 4:00pemailed the contractor about the fence
5 traces in today’s buffer · 3 are about the same thing
Reconsolidation
Recall makes a memory briefly editable. Then it's re-stored, sometimes rewritten.
Contested — The boundary conditions for reconsolidation — when a recalled trace becomes labile — are still debated.
We caught a twelve-inch trout at the lake, and a few people clapped.
one story, stored · each recall opens the copy, and the re-save keeps whatever changed
Episodic ↔ semantic
"I met her on Tuesday" slowly becomes "I know her" — the event fades, the meaning stays.
Contested — Whether gist is compressed episodic detail or an independent multi-trace abstraction is unresolved.
the episode
- whenTuesday, 3:14 pm
- detailshe ordered an oat-milk flat white
- whereCafé Luna, 2nd Ave
- topictalked about the Montana trip
- eventcoffee with Sarah
the gist
- Sarah — a friend.
- You get coffee together.
- She's easy to talk to.
1 day after the episode · 0 of 5 details gone · the gist is forming
Schema & assimilation
New facts don't land on blank ground. They're folded into what you already believe.
what you know about Sam
- Sam's vegetarian
- not a morning person
- bikes everywhere
a standing picture of Sam — every new fact about them lands on this ground first
The map
The landscape
Every patent, paper, and open-source system found across five rounds of adversarial prior-art search. Sorted by priority date, the shape is the finding: the field is occupied, not crowded — a convergence in 2023–2026, with Microsoft, the ex-Cerego team behind Alai Vault, DeWix, Snap, Meta, Samsung, and a fast-moving research frontier all arriving at once.
47 of 47 references
| Coverage | Cite | |||
|---|---|---|---|---|
| CeregoLearning-effectiveness / memory-retention modelingUS6652283B1 | patent | 1999Granted 2003 | Foundational predictive modeling of learning effectiveness and memory retention — the root of the Cerego lineage. | |
| Howard, KahanaTemporal Context Modeltcm | paper | ~2002Howard & Kahana, 2002 | The temporal-contiguity effect and the drifting-context account underlying session-based association. | |
| Anderson, Laird, DancyACT-R / Soar / ACT-RΦ — cognitive architecturesact-r-soar | paper | ~2013Cognitive architectures, 1990s–2013 | Decades of cognitive-architecture work: prospective memory, affect-modulated declarative memory, base-level activation decay. | |
| DeepMindPrioritized experience replay; salience-weighted transitions (RL)US9679258B2 | patent | ~2015Granted 2017 (~2015 priority) | Prioritized experience replay weights transitions by salience — reinforcement learning, not LLM memory, but foundational salience-weighting art. | |
| Cerego (Smith Lewis)Personalized learning / learner-adaptive enginesUS20170075881A1 | patent | 2015Application | Personalized learning with learner-adaptive scheduling engines. | |
| Cerego (Smith Lewis, Harlow)Personalized learning, automated generationUS11721230B2 | patent | 2017Granted | Personalized learning with automated content generation. Family also includes US11081018B2. | |
| Cerego (Smith Lewis, Harlow)Predictive memory-retention / forgetting model with thresholds + schedulingUS11776417B2 | patent | 2017Granted 2023 | The most on-point Cerego patent: predictive retention/forgetting model with retention thresholds and review scheduling. | |
| CeregoAuto-generating related conceptsUS11487804B2 | patent | 2017Granted | Automatically generating related concepts from learning material. Family also includes US11238085B2. | |
| MicrosoftKnowledge graph that "models the human brain memory system"US20180114111A1 | patent | ~2018Application (~2018) | Knowledge-graph memory explicitly framed as modeling the human brain memory system. | |
| ServiceNowFocused conversation context management; overlay context from stored episodesUS20190294675A1 | patent | 2018Application | Manages focused conversation context, overlaying context retrieved from stored episodes. | |
| MetaMemory-graph traversal; "reactive/proactive memory" for assistantsUS11657094B2 | patent | ~2019Granted (~2019 priority) | Traverses a memory graph; distinguishes reactive and proactive memory for assistant systems. | |
| MetaProactive reminders for assistant systemsUS11567788B1 | patent | 2019Granted | Generates proactive reminders — a prospective-memory analog for assistants. | |
| EMC / Dell"Relevance decay" — linear / exponential / step functionsUS10885464B1 | patent | ~2020Granted 2021 (~2020 priority) | Applies relevance decay via linear, exponential, or step functions. | |
| Human AI Labs (Personal.ai)"Memory retention system" — automatic capture of searchable human memoryUS20210182020A1 | patent | 2020Application | Automatic capture of a searchable store of human memory — the user-facing half of a dual-system memory design. | |
| Cerego (Harlow)Personalized learning systemUS11545042B2 | patent | 2020Granted 2023 | Later Cerego personalized-learning patent extending the retention-modeling lineage. | |
| MicrosoftFoundational RAG (feature-vector similarity → augmented prompt)US20240346256A1 | patent | ~2023Application (~2023) | Foundational retrieval-augmented generation: feature-vector similarity retrieval feeding an augmented prompt. | |
| Modulus AIPreserves salient entities/facts under high-compression summarizationUS12008332B1 | patent | ~2023Granted 2024 (~2023 priority) | Preserves salient entities and facts when summarizing under high compression — pre-compression preservation. | |
| Alai Vault LLC (Smith Lewis, Harlow)Biologically-inspired memory: power decay of salience, episodic/semantic tiers, agent memory toolUS20240289863A1 | patent | 2023Applications pending | Strongest single §103 obviousness reference. Specification teaches parameterized age-dependent decay of salience, power decay of episodic memories, episodic/semantic tiers, vector-DB long-term memory, a conversational memory tool — though the 20 claims cover brand-marketing agent personalization only. Inventors are ex-Cerego. Family: WO2024178435A1, CA3284082A1. | |
| Shinn, et al.Reflexion: Language Agents with Verbal Reinforcement Learningreflexion | paper | 2023arXiv:2303.11366 | First-person self-reflective episodic buffer — anticipates first-person memory encoding. | |
| Park, et al.Generative Agents: Interactive Simulacra of Human Behaviorgenerative-agents | paper | ~2023Stanford, Apr 2023 | Retrieval-time scoring by recency + importance + relevance — §103 art for dynamic strength and context-dependent retrieval. | |
| MicrosoftDynamic LLM prompt construction from summary + scoring modelUS12462095B2 | patent | 2023Granted 2025-11-04 | Constructs LLM prompts dynamically from a running summary plus a memory-scoring model. | |
| Zhong, et al.MemoryBank: Enhancing LLMs with Long-Term Memorymemorybank | paper | ~2023Zhong et al., 2023 | Applies the Ebbinghaus forgetting curve to LLM memory. | |
| MicrosoftPrompt construction by dynamically selecting context by semantic relevanceUS20240394477A1 | patent | 2023Application | Selects context for prompt construction by semantic relevance. | |
| SnapChatbot long-term memory: summarize → store → semantic-search → augment promptUS20240414108A1 | patent | 2023Application | End-to-end chatbot long-term memory: summarize, store, semantic-search, augment the prompt. | |
| MicrosoftExtract memories via generative model → DBSCAN/HDBSCAN cluster into memory banksUS20250021474A1 | patent | 2023Application | Extract-then-cluster consolidation (DBSCAN/HDBSCAN) into memory banks — anticipates two-pass consolidation. | |
| Packer, et al.MemGPT: Towards LLMs as Operating Systemsmemgpt | paper | 2023Packer et al., 2023 | OS-inspired tiered memory (main context vs. archival) with paging between tiers. | |
| Navinfo EuropeCatastrophic forgetting + episodic/semantic (neural-net training)US20240119280A1 | patent | ~2024Application (~2024) | Addresses catastrophic forgetting with an episodic/semantic split — for neural-network training, not a retrievable memory store. | |
| Escendant LtdHierarchical topic manager; session boundaries before context window fillsUS12093846B1 | patent | 2024Granted 2024-09-17 | Hierarchical topic management that draws session boundaries before the context window fills — granted pre-compression preservation art. | |
| MicrosoftLayered working/episodic/collective memory; embedding-compressed historyUS20250371317A1 | patent | 2024Application | Layered working / episodic / collective memory with embedding-compressed history. | |
| MicrosoftSemantic memory vector-DB consolidation; DBSCAN clustering; age/effectiveness selectionWO2025147358A1 | patent | ~2024PCT published 2025-07-10 | Consolidates a semantic-memory vector DB via DBSCAN clustering with age/effectiveness-based selection — anticipates two-pass consolidation. | |
| Beijing Lingxin Intelligent TechLLM memory retrieval; BERT-based "Memory Conflict Resolution"CN119903125B | patent | 2024Granted 2025-08-22 | Encoding-time memory-conflict resolution: a BERT model detects conflict and resolves by elimination or integration — overlaps part of the interference mechanism (a §103 reference, not anticipation). | |
| SamsungKV-cache compression; frequency + recency importance scoringUS20250348732A1 | patent | ~2025Application (~2025) | KV-cache compression driven by tensor-level frequency + recency importance scoring. | |
| Rasmussen, et al.Zep / Graphiti — temporal knowledge-graph memoryzep-graphiti | oss | 2025arXiv:2501.13956 (OSS) | Bounded edge deduplication for complexity reduction in a temporal knowledge graph. | |
| Xu, et al.A-MEM: Agentic Memory for LLM Agentsa-mem | paper | 2025arXiv:2502.12110 | A top-k similarity neighbor set drives memory evolution (Zettelkasten-style linking). | |
| DeWix Co., Ltd.Memory-curve context maintenanceKR102809164B1 | patent | 2025Granted 2025-05-21 | The most directly relevant reference — claims (not just spec) recite the machinery: exponential decay R=e^(-a·x), top-20% summarization, timestamped storage, delete-low-retention (claim 1); importance from parts of speech with adjectives weighted for emotional context (claim 2); ML-derived decay rate (claim 4); semantic-similarity clustering and pruning (claim 5). Claim analysis via machine translation. | |
| Taranjeet, et al.Mem0 — scalable long-term memory for AI agentsmem0 | oss | 2025arXiv:2504.19413 (OSS) | ADD/UPDATE/DELETE/NOOP operations with a bounded top-s dedup context. The UPDATE-with-history operation is exactly the in-place-update ('reconsolidation') mechanism — §102 anticipation for it. | |
| Wang, SunPI-LLM ("Unable to Forget") — proactive interference in LLMspi-llm | paper | 2025arXiv:2506.08184 | A benchmark diagnosing proactive interference as an LLM problem — establishes the term/problem pre-priority, but discloses no mechanism to solve it. | |
| Dynamic Affective Memory Managementdamm | paper | 2025arXiv:2510.27418 | A multi-dimensional numerical salience profile updated by a non-LLM arithmetic rule — §103 art for a salience vector with an affective axis. | |
| Wei, Peng, Dong, Xie, WangFadeMem: Biologically-Inspired Forgetting for Efficient Agent Memoryfademem | paper | 2026arXiv:2601.18642 (Alibaba; Peking University) | The closest research reference. Does contradiction-triggered, encoding-time, graded arithmetic attenuation of memory strength in an LLM agent — most of what a proactive-interference mechanism does. But conflict is LLM-classified (GPT-4o-mini), it attenuates a scalar not a multi-dimensional vector, strength is a stored field batch-rewritten (not computed on read), and it uses one model tier. Predates engram's priority by ~5 weeks. | |
| PAM — retrieval through temporal co-occurrencepam | paper | ~2026arXiv:2602.11322 | Retrieval through temporal co-occurrence — §103 art for temporal association. | |
| Gentleman-ProgrammingGentleman-Programming/engramgentleman-engram | oss | ~2026Open source | Conventional CRUD memory store (SQLite + FTS). | |
| foramorentforamorent/engram-ai-memoryforamorent-engram | oss | 2026Open source | Basic Ebbinghaus decay for AI memory. | |
| SleepGate — proactive interference + sleep consolidationsleepgate | paperpost-priority | 2026arXiv:2603.14517 | Field context only — post-dates engram's priority, so not prior art. Combines proactive interference with sleep-style consolidation. | |
| SCM — multi-dimensional importance + dual-phase consolidationscm | paperpost-priority | ~2026arXiv:2604.20943 | Field context only — post-priority. Multi-dimensional importance scoring with dual-phase consolidation. | |
| HeLa-Mem — Hebbian episodic→semantic distillationhela-mem | paperpost-priority | ~2026arXiv:2604.16839 | Field context only — post-priority. Hebbian distillation of episodic memories into semantic ones. | |
| YourMemoryyourmemory | osspost-priority | ~2026Open source | Field context only — post-priority OSS memory system. | |
| Mike LaPeterCounterpartscounterparts | osspost-priority | 2026Open source (MIT); on npm since Sep 2026 | Interpretation over transcription: the AI writes what it learned in its own words, memories fade unless used, beliefs about people and things are revised only under accumulated pressure and keep their history, and an identity the AI reads at the start of every session. Local: one SQLite file. |
The uncharted region
Where the ground is still open
If the obvious mechanisms are occupied, where is the real work? Read as a map, the search points fairly clearly.
Evaluation
Almost everyone is building memory systems; almost no one agrees how to measure them. Most results lean on synthetic benchmarks that don’t capture real interaction — the gap behind the RECALL conversation-grounded benchmark.
Interference, solved not just diagnosed
LLMs demonstrably can’t suppress outdated information. FadeMem and Counterparts both attempt mechanisms — but “the model retrieves a value it should have forgotten” is far better characterized than it is solved.
Long-horizon memory
Nearly every system operates on days or weeks. What a memory should do over months and years — what genuinely consolidates, what becomes irrecoverable versus merely hard to find — is close to unexplored.
Forgetting vs. retrieval failure
Human memory distinguishes “the trace is gone” from “the trace exists but the path is lost.” Most AI systems only delete. Graded, partially reversible forgetting has very little prior art and real conceptual room.
How this was built
Method & caveats
The landscape was compiled from five escalating rounds of adversarial prior-art search conducted during the engram project (May 2026), Counterparts’ predecessor — broad academic and OSS sweeps, then patents across the US, Europe, Korea, and China, then focused deep dives on the closest references. The goal was deliberately hostile: not to confirm an idea was new, but to find everything that would prove it wasn’t.
- This is a research record and a field reference — not legal advice, and not a freedom-to-operate or patentability opinion.
- Patent numbers were verified against live Google Patents pages during the searches; the Korean patent’s claim analysis used machine translation.
- Contested points in the neuroscience are marked ⚖; they reflect active scientific debate, not settled fact.
- engram, Counterparts’ predecessor, is open source (github.com/mlapeter/claude-engram); the RECALL benchmark is at github.com/mlapeter/recall-bench.
References
Sources
- (1904). Die Mneme (introducing the term "engram"). Wilhelm Engelmann, Leipzig.
- (1885). Über das Gedächtnis (Memory: A Contribution to Experimental Psychology). Duncker & Humblot.
- (1974). Single-trace fragility theory of memory dynamics. Memory & Cognition, 2(4).
- (2012–2015). Optogenetic identification and manipulation of memory engram cells. Nature; Science.
- (1972). Episodic and semantic memory. Organization of Memory (Academic Press).
- (2002). Fuzzy-Trace Theory and false memory. Current Directions in Psychological Science, 11(5).
- (2000). Fear memories require protein synthesis in the amygdala for reconsolidation after retrieval. Nature, 406.
- (2000). Memory — a century of consolidation (amygdala modulation of emotional memory). Science, 287.
- (2002). A distributed representation of temporal context. Journal of Mathematical Psychology, 46.
- (1995). Why there are complementary learning systems in the hippocampus and neocortex. Psychological Review, 102(3).
- (1997). Memory consolidation, retrograde amnesia and the hippocampal complex (multiple-trace theory). Current Opinion in Neurobiology, 7(2).
- (2010). The memory function of sleep (sharp-wave ripples, replay). Nature Reviews Neuroscience, 11.
- (2006). Test-enhanced learning: the testing effect and retrieval-induced strengthening. Psychological Science, 17(3).
- (1932). Remembering: A Study in Experimental and Social Psychology (schema theory). Cambridge University Press.
- (1990). Normal aging and prospective memory. Journal of Experimental Psychology: LMC, 16(4).
- (1957). Interference and forgetting (proactive interference). Psychological Review, 64(1).
42 prior-art references · 11 mechanisms · a defensive publication, not a patent. Take the map and go further.