ZenBrain

• arXiv preprint (cs.AI): arxiv.org/abs/2604.23878

• Open-access archive (Zenodo / CERN): doi.org/10.5281/zenodo.19353663

• ORCID: 0009-0001-1793-012X

• License: CC BY 4.0 (paper) · Apache-2.0 (code)

@misc{bering2026zenbrain,
  title         = {ZenBrain: A Neuroscience-Inspired 7-Layer Memory Architecture for Autonomous AI Systems},
  author        = {Bering, Alexander},
  year          = {2026},
  eprint        = {2604.23878},
  archivePrefix = {arXiv},
  primaryClass  = {cs.AI},
  doi           = {10.5281/zenodo.19353663},
  url           = {https://arxiv.org/abs/2604.23878}
}

Feedback, replications, and counter-results are explicitly welcome — please open an issue or reach out via research@zensation.ai.

Your AI forgets everything after every conversation. ZenBrain fixes that — with the same mechanisms your brain uses: spaced repetition, emotional consolidation, Hebbian strengthening, and exponential forgetting curves. Not a vector database with a wrapper. Actual neuroscience.

Architecture vs. this package. ZenBrain's architecture is 15 neuroscience-inspired mechanisms — 9 foundational algorithms + 6 Predictive Memory Architecture (PMA) components (paper). The 6 PMA components are proprietary and run in the production system. This open-source package ships the algorithm library: 10 core algorithms + 10 advanced research modules (20 modules), zero-dependency.

Benchmark: LongMemEval-500

On LongMemEval-500, three of nine head-to-head answer-quality comparisons hold against Letta, Mem0 and A-Mem — all three against A-Mem, the remaining six are ties, none lost. Three competitors x three LLM judges, Bonferroni-corrected (alpha = 0.05/18) and version-matched. It reaches 91.3% of a full-context oracle's binary-judge accuracy at 1/109.6 of the per-query token cost (47.7% vs. 52.2%).

Works out of the box without an embedding provider — lexical ranking, zero dependencies. With nomic-embed-text as the embedding provider you get the configuration those figures were measured in.

The paper prints where ZenBrain loses as well: on LoCoMo, substring-based aggregate F1 favours lexical retrieval (BM25) by metric design, and we do not contest that. The advantage is most pronounced on judge-graded answer quality and cross-session reasoning.

The mechanism comparison further down re-runs from this repository in under a minute — bash scripts/compare-mechanisms.sh, no API keys and nothing to install. It prints a positive and a negative control before the result, so the instrument can be checked before its output is trusted. The method, the effect sizes and the ablations behind the numbers above are in the paper; this repository ships no runner for them. The archived packages below run the significance tests and effect sizes in full, and the mechanism ablation behind paper Tables 7–9.

• Method, effect sizes and ablations: arXiv:2604.23878

• Open-access archive: 10.5281/zenodo.19353663

Related MCP server: brainlayer

Reproduction packages

The raw material behind the numbers above is deposited on Zenodo, open access and citable. Both links are concept DOIs and resolve to the latest version, the same convention this README uses for the paper archive; each description was measured against the version named after it.

• Mechanism ablation, paper Tables 7–9 — 10.5281/zenodo.22162063 (described here: v1.0.0). Four experiment suites (95 tests), the reference JSON the paper's tables were generated from, and verify-against-reference.mjs, which diffs a fresh run against that reference and exits non-zero on drift. npm install && npm run experiments; the run itself needs no API keys and no network, and finishes in under a minute on a laptop. Two of the paper's other ablation tables need data this package does not carry: Table 11 the LoCoMo corpus, Table 13 a different pipeline. The package says so itself.

• Measurement package, LongMemEval-500 and the real-pipeline flag ablation — 10.5281/zenodo.22161977 (described here: v1). Per-(system, judge, seed) judged outputs, the flag manifests as recorded at run time, a SHA256SUMS.txt covering every file in the package, and the analysis scripts. Three of those scripts are stdlib-only and self-checking — the oracle comparison behind the 91.3% figure, the judge-agreement figures, and the real-pipeline flag-ablation table: each prints every re-derived value next to the reference it has to match, and exits non-zero on mismatch. The significance tests behind the nine head-to-head comparisons against Letta, Mem0 and A-Mem sit in a separate script that needs numpy and scipy; it recomputes all eighteen pairwise tests and rewrites the deposited significance JSON byte-identically, so what catches a mismatch there is the checksum, not an exit code.

Both packages name what they do not cover. Replications and counter-results are welcome: research@zensation.ai.

How ZenBrain differs from Mem0, Letta and Zep

ZenBrain implements fifteen mechanisms taken from human memory research. No system among those surveyed in the paper integrates more than two of them. The table below records which of the mechanisms appear in the public source of three widely used memory systems, at pinned versions, on a fixed date.

How this was measured, 27 August 2026. Full-text search over the checked-out public source of mem0ai/mem0 (npm mem0ai 3.1.7, PyPI mem0ai 2.0.19), letta-ai/letta-code (npm @letta-ai/letta-code 0.31.2) and getzep/zep, lockfiles excluded. A dash means the term does not occur in that snapshot — not that the system cannot do something comparable under another name. Dependency counts are declared direct dependencies: @zensation/core resolves to two packages, both our own; mem0ai declares four, @letta-ai/letta-code eighteen. Re-run the whole check yourself with scripts/compare-mechanisms.sh; it prints its own positive and negative controls so you can see the instrument works before you trust the result.

Human memory does not work like a key-value store. The brain keeps specialised systems for different kinds of memory, forgets actively, modulates by emotion and retrieves by context. ZenBrain brings those mechanisms to AI agents.

Advanced algorithms (since v0.3.0, May 2026)

On top of the 10 core algorithms above, @zensation/algorithms ships 10 advanced algorithms grounded in recent neuroscience and ML research. Each is exposed as its own sub-path (@zensation/algorithms/<name>) and remains zero-dependency:

• fsrs-vmPFC — Prediction-Error coupled FSRS

• hebbian-two-factor — Two-Factor synaptic consolidation

• sleep-simulation-selection — RL-based replay selection

• spectral-health — Fiedler-value KG health monitor

• ib-budget — Information-Bottleneck retention budget

• dopamine-routing · hopfield-stm · personalized-pagerank · surprise-gradient-memory · temporal-multi-route

See CHANGELOG.md for details.

Quick Start

Requires Node.js 22 or newer (since 0.4.0). On Node 20 or older, npm silently installs the last compatible release (@zensation/algorithms@0.3.4, @zensation/core@0.2.2) instead of the current one — which looks like a broken package but is a platform mismatch. See CHANGELOG.

npm install @zensation/algorithms

import {
  initFromDecayClass,
  getRetrievability,
  updateAfterRecall,
  tagEmotion,
  computeEmotionalWeight,
  computeHebbianStrengthening,
  propagateForRelation,
} from '@zensation/algorithms';

// 1. Schedule a memory with FSRS
const memory = initFromDecayClass('normal_decay');

// 2. A week later, check recall probability (Ebbinghaus curve)
const aWeekLater = new Date(Date.now() + 7 * 24 * 60 * 60 * 1000);
const retention = getRetrievability(memory, aWeekLater);
console.log(`Recall probability: ${(retention * 100).toFixed(1)}%`);
// ~36.8% — retrievability has decayed over the week

// 3. User recalled it anyway — update scheduling
const updated = updateAfterRecall(memory, 4, retention, aWeekLater);
// stability 7 -> 8.19: recalling at low retrievability gives a bigger boost

// 4. Tag emotional significance
const emotion = tagEmotion('I am absolutely thrilled — I got the promotion!');
const weight = computeEmotionalWeight(emotion);
console.log(`Decay multiplier: ${weight.decayMultiplier}x`);
// 2.7x — emotional memories decay nearly 3x slower

// 5. Strengthen knowledge connections (Hebbian)
const stronger = computeHebbianStrengthening(1.0);
// 1.09 — "neurons that fire together wire together"

// 6. Propagate confidence through your knowledge graph
const confidence = propagateForRelation(0.5, 0.8, 1.0, 'supports');
// 0.9 — supporting evidence increases confidence

Want the advanced algorithms?

import {
  computeKGPredictionError,
  computeAdaptiveFSRSInterval,
} from '@zensation/algorithms/fsrs-vmPFC';

// Couple FSRS scheduling with the prediction-error signal from your
// knowledge graph: when the embedding has shifted a lot since the last
// review (high cosine distance), shrink the next interval; otherwise push
// it out. Both arrays must have the same length.
const lastEmbedding = [0.1, 0.2, 0.3, 0.4];
const currentEmbedding = [0.5, 0.4, 0.1, 0.2];
const pe = computeKGPredictionError(lastEmbedding, currentEmbedding);
const nextInterval = computeAdaptiveFSRSInterval(14, pe);

Each advanced algorithm has its own sub-path (@zensation/algorithms/spectral-health, @zensation/algorithms/ib-budget, …). All zero dependencies.

Runnable examples

Seven self-contained examples live in examples/:

npx tsx examples/basic-chatbot.ts

The integration examples additionally need their respective SDK installed; examples/README.md lists which. Open good first issues are the best way in if you want to contribute.

The Science Behind It

7-Layer Memory Architecture

Layer 7: Cross-Context Memory    ← Shared knowledge across domains
Layer 6: Core Memory             ← Pinned facts (Letta-style)
Layer 5: Procedural Memory       ← "How to do X" (skills & workflows)
Layer 4: Long-Term Semantic      ← Facts with FSRS scheduling
Layer 3: Episodic Memory         ← Concrete experiences & events
Layer 2: Short-Term / Session    ← Current conversation context
Layer 1: Working Memory          ← Active task focus (7±2 items)

Each layer has different retention characteristics, consolidation rules, and retrieval mechanisms — just like the human brain.

FSRS Spaced Repetition

FSRS (Free Spaced Repetition Scheduler) outperforms SM-2 by 30%. It uses the desirable difficulty principle: reviewing when retention is low gives a bigger stability boost. Your AI reviews important facts at optimal intervals — never too early (wasteful), never too late (forgotten).

Emotional Memory

The amygdala modulates memory consolidation — emotional events are remembered more vividly (flashbulb memory). ZenBrain's emotional tagger assigns arousal, valence, and significance scores using a 400+ keyword lexicon (English & German). Emotional memories get up to 3x longer decay half-lives.

Hebbian Learning

"Neurons that fire together wire together" (Hebb, 1949). Knowledge graph edges that are frequently co-activated grow stronger. Unused edges decay and eventually get pruned. The result: a self-organizing knowledge structure that reflects actual usage patterns, with homeostatic normalization to prevent runaway growth.

Ebbinghaus Forgetting Curves

Ebbinghaus (1885) showed that memory decays exponentially: R = e^(-t/S). ZenBrain implements personalized decay profiles that adapt to individual learning patterns, with SM-2 compatibility for existing spaced repetition systems.

Context-Dependent Retrieval

Tulving's Encoding Specificity Principle (1973): memories are recalled better when the retrieval context matches the encoding context. ZenBrain captures temporal context (time of day, day of week) and task type at encoding time, providing up to a 30% retrieval boost when contexts match.

Bayesian Confidence Propagation

Knowledge isn't isolated — facts support or contradict each other. ZenBrain propagates confidence through your knowledge graph using Bayesian belief updates: supporting evidence increases confidence, contradictions decrease it, with damping for numerical stability.

Sleep Consolidation

During sleep, the hippocampus replays recent experiences, strengthening important memories and pruning weak connections (Stickgold & Walker, 2013). ZenBrain simulates this process: selectForReplay() prioritizes emotional and recently-accessed memories, simulateReplay() boosts their stability by 50%, and pruneWeakConnections() removes weak Hebbian edges — implementing the Synaptic Homeostasis Hypothesis (Tononi & Cirelli, 2006).

import { selectForReplay, simulateReplay } from '@zensation/algorithms/sleep-consolidation';

// Select memories for overnight consolidation
const toReplay = selectForReplay(allMemories);
// Simulate sleep replay — stability ↑, weak edges pruned
const result = simulateReplay(toReplay);
console.log(`Replayed ${result.summary.totalReplayed} memories, avg stability +${result.summary.avgStabilityIncrease.toFixed(1)} days`);

Memory Coordinator

The MemoryCoordinator orchestrates all 7 layers into a single cohesive system — inspired by Global Workspace Theory (Baars, 1988):

import { MemoryCoordinator } from '@zensation/core';

const memory = new MemoryCoordinator({ storage: adapter, embedding: embedder });

// Auto-routes to the right layer (semantic, episodic, procedural, or core)
await memory.store('User prefers TypeScript', { type: 'auto' });

// Cross-layer search with ranked, deduplicated results
const results = await memory.recall('programming preferences');

// Consolidate: promote episodic → semantic, apply decay
await memory.consolidate();

// FSRS review queue across all layers
const dueItems = await memory.getReviewQueue();

Packages

Tree-Shakeable Imports

Every algorithm is available as a subpath export:

// Import everything
import { tagEmotion, updateAfterRecall } from '@zensation/algorithms';

// Or just what you need (better tree-shaking)
import { updateAfterRecall } from '@zensation/algorithms/fsrs';
import { tagEmotion } from '@zensation/algorithms/emotional';
import { computeHebbianStrengthening } from '@zensation/algorithms/hebbian';
import { propagateForRelation } from '@zensation/algorithms/bayesian';
import { selectForReplay } from '@zensation/algorithms/sleep-consolidation';
import { getRetrievabilityWithCI } from '@zensation/algorithms/intervals';
import { generateRetentionCurve } from '@zensation/algorithms/visualization';

Use Cases

AI Chatbots with Long-Term Memory

import { updateAfterRecall, getRetrievability, scheduleNextReview } from '@zensation/algorithms/fsrs';
import { tagEmotion, computeEmotionalWeight } from '@zensation/algorithms/emotional';

// When your AI learns a fact about the user:
function rememberFact(fact: string) {
  const memory = initFromDecayClass('normal_decay');
  const emotion = tagEmotion(fact);
  const weight = computeEmotionalWeight(emotion);

  // Emotional facts get longer retention
  return {
    ...memory,
    emotionalWeight: weight.consolidationWeight,
    decayMultiplier: weight.decayMultiplier,
  };
}

// Before each conversation, check what needs reinforcement:
function getFactsDueForReview(facts: MemoryState[]) {
  return facts.filter(f => getRetrievability(f) < 0.7);
}

Knowledge Graph with Self-Organizing Edges

import { computeHebbianStrengthening, computeHebbianDecay } from '@zensation/algorithms/hebbian';
import { propagateForRelation } from '@zensation/algorithms/bayesian';

// When two concepts are mentioned together:
function coActivate(edge: { weight: number }) {
  edge.weight = computeHebbianStrengthening(edge.weight);
}

// Periodic maintenance — decay unused edges:
function decayEdges(edges: { weight: number; lastUsed: Date }[]) {
  for (const edge of edges) {
    edge.weight = computeHebbianDecay(edge.weight);
    // Edges below MIN_WEIGHT (0.1) can be pruned
  }
}

RAG with Confidence Scoring

import { propagateForRelation, isSignificantChange } from '@zensation/algorithms/bayesian';

// After retrieval, propagate confidence through related facts:
function updateConfidenceGraph(facts: Fact[], relations: Relation[]) {
  for (const rel of relations) {
    const newConf = propagateForRelation(
      rel.target.confidence,
      rel.source.confidence,
      rel.weight,
      rel.type // 'supports' | 'contradicts' | 'related_to'
    );
    if (isSignificantChange(newConf, rel.target.confidence)) {
      rel.target.confidence = newConf;
    }
  }
}

Extracted From Production

These aren't toy implementations — ZenBrain's algorithms are extracted from ZenAI, a production AI platform. Everything claimed here is verifiable in this repository:

• 801 tests (489 algorithms · 137 core · 54 adapter-sqlite · 48 adapter-postgres · 55 mcp · 18 ai-sdk), all passing

• Zero runtime dependencies — pure TypeScript, dual ESM + CJS, tree-shakeable subpath exports

• Reproducible — building from this source produces the same 153-file @zensation/algorithms@0.5.0 tarball published on npm

• 7-layer memory architecture grounded in published neuroscience

Contributing

We welcome contributions! See CONTRIBUTING.md for guidelines. Issues and pull requests are answered; replications and counter-results are especially welcome.

Resources: API Reference | Recipes | Architecture | Benchmarks | FAQ | Roadmap

# Clone the repo
git clone https://github.com/zensation-ai/zenbrain.git
cd zenbrain

# Install dependencies
npm install

# Run tests
npm test

# Build all packages
npm run build

Research

ZenBrain's architecture and algorithms are documented in an open-access technical disclosure:

• arXiv preprint (cs.AI): arxiv.org/abs/2604.23878

• Open-access archive: ZenBrain: A Neuroscience-Inspired 7-Layer Memory Architecture for Autonomous AI Systems (Zenodo, DOI: 10.5281/zenodo.19353663 — resolves to the latest version)

• TDCommons: Technical Disclosure (CC BY 4.0)

• HuggingFace: Model Card & Benchmarks

If you use ZenBrain in academic work, please cite:

@misc{bering2026zenbrain,
  title         = {ZenBrain: A Neuroscience-Inspired 7-Layer Memory Architecture for Autonomous AI Systems},
  author        = {Bering, Alexander},
  year          = {2026},
  eprint        = {2604.23878},
  archivePrefix = {arXiv},
  primaryClass  = {cs.AI},
  doi           = {10.5281/zenodo.19353663},
  url           = {https://arxiv.org/abs/2604.23878}
}

Cited by

Five works cite ZenBrain (as of 28 September 2026): a survey by the Sico team at Microsoft Research, MindMemOS by Huawei's Noah's Ark Lab, the EVD preprint, the HindsightTag manuscript and the Lamperouge manifesto.

• Agentic Evolution: From Self-Improving Agents to Co-Evolving Human-AI Systems, a survey by the Sico team at Microsoft Research (2026), calls forgetting "critically understudied": very few of the roughly 100 Memory & Sense papers it surveys implement explicit forgetting. It lists ZenBrain as one of four works in its forgetting/lifecycle group and as its only source for sleep-based consolidation.

• MindMemOS (Huawei's Noah's Ark Lab, arXiv:2608.12428, August 2026) relies on exactly three references for the foundational claim of its introduction: two field surveys and a single system paper — ZenBrain.

• HindsightTag (Vivek Govindbhai Dudhat, manuscript, July 2026) discusses ZenBrain as its closest prior system and uses its MemoryCoordinator interface as the integration example.

• EVD: An Emotional Valence Dimension for Persistent Agent Memory (R. J. Vandelinder and I. Vandelinder, Exile Research, Zenodo, June 2026) places ZenBrain in its related work.

• The field measures the wrong thing (Lamperouge, manifesto, Zenodo, August 2026) selects ZenBrain among six works on persistent internal state and notes that its "days" are simulation steps.

Each work with the sentence that cites ZenBrain: zensation.ai/en/publikationen.

Citing ZenBrain or building on it? Open an issue and we will add you here.

Community

• GitHub Discussions: Ask a question, show what you built — help, show-and-tell, feature requests

• GitHub Issues: Bug reports & feature requests

• Email: open-source@zensation.ai

License

Apache 2.0 — use it in production, modify it, distribute it. Just keep the attribution.