Skip to content
zensation

Research overview

Three tracks, architecture and agenda

Publications

Preprints, software, identifiers

Methodology

Operational standards and validation

Research ethics

Fundamental rights and compliance

Resources

Code, data, citation, open science

Adopt AI

Neutral roadmap & interactive simulation

Public sector & funding

Collaborations in the public sector

Protecting public spaces

Track B — rights-preserving early warning

Technology

The architecture behind every application

Playground

Run the open memory core live

Work with usAboutOpen SourceBlog
Contact
zensation
Research overviewPublicationsMethodologyResearch ethicsResources
Adopt AIPublic sector & fundingProtecting public spacesTechnologyPlayground
Work with usAboutOpen SourceBlogContact
  1. Home/
  2. Technology

The science behind ZenBrain

15 neuroscience-inspired mechanisms across 7 memory layers. Open-access preprint on arXiv. All documented.

Open-source release: March 25, 2026Alexander Bering, Principal Investigator

“Memory is not storage — it is a living process of forgetting, consolidation, and rediscovery. We translated this process into software.”

— Alexander Bering
TECHNICAL PAPER

Published research

Our technical paper is publicly available as an open-access preprint on Zenodo and arXiv, with a defensive disclosure on Elsevier TDCommons.

PreprintOpen Access

ZenBrain: A Neuroscience-Inspired 7-Layer Memory Architecture for Autonomous AI Systems

Alexander Bering — ZenSation AI, Kiel, Germany

We present ZenBrain, a neuroscience-inspired 7-layer memory architecture integrating 15 mechanisms grounded in peer-reviewed neuroscience: 9 foundational algorithms plus a Predictive Memory Architecture (PMA) with NeuromodulatorEngine, ReconsolidationEngine, TripleCopyMemory, PriorityMap, StabilityProtector, and MetacognitiveMonitor. Evaluated across ten experiments on LoCoMo, MemoryAgentBench, MemoryArena, and the LongMemEval-S Full-500 replication. On LongMemEval-500, three of nine head-to-head judge comparisons hold for ZenBrain, all three against A-Mem (Bonferroni-corrected p ≤ 6.2e-31); the remaining six, against Letta and Mem0, are ties at the paper's own criterion, judged at equal evidence depth and under version-matched judges. No comparison is lost. Under the official binary judge, ZenBrain reaches 91.3 % of long-context-oracle accuracy at 1/109.6 of the per-query token budget — the oracle beats ZenBrain by only 4.5 pp while using ~109.6× more tokens and no memory architecture. Sleep consolidation: +37 % stability, −47.4 % storage. TripleCopyMemory retains 91.2 % strength at 30 days. The full 15-mechanism ablation reveals a cooperative survival network where 9 of 15 mechanisms become individually critical under stress (decay=0.25, 60 days). All experiments are reproducible with seeded PRNG; the implementation is released as open-source packages under the @zensation npm scope.

DOI10.5281/zenodo.19353663
Download PDF
ZenodoPublishedarXiv (2604.23878)LiveHuggingFaceModel CardElsevier TDCommonsPublishedOpenAIREPublished
8 INTEGRATED SYSTEMS

Eight innovations, shipped as one system

Eight technologies shipped together in production in this combination.

Sleep Consolidation Engine

March 2026

Idle-time memory consolidation in production — inspired by hippocampal replay (Stickgold & Walker 2013). Weak connections are pruned, stable ones strengthened.

Among the systems compared here — Mem0, Letta, Zep — none ships sleep-time consolidation; the published designs for it remain proposals.

Source: zenbrain → (opens in new tab)

7-Layer Memory Coordinator

March 2026

Unified orchestrator for 7 memory layers in production — from working memory to cross-context memory. Based on Global Workspace Theory (Baars 1988).

Mem0 ships 2 layers, Letta 3, Zep 2. ZenBrain ships 7 — among the deepest memory architectures in open source today.

Source: zenbrain → (opens in new tab)

A-RAG (Autonomous Retrieval Agent)

March 2026

Meta-agent that creates retrieval plans before any search is executed. Heuristic-first with LLM fallback, max 4 dependent steps.

A dedicated planning layer ahead of retrieval, with a heuristic gate: simple queries never reach the planner and cost no LLM call.

Source: ZenAI snapshot (8 May 2026) → (opens in new tab)

Multi-Agent Debate Protocol

March 2026

Structured 3-round debate protocol when agents disagree. Challenge → Response → Resolution with automatic escalation.

After three rounds without consensus the protocol escalates to a human instead of computing a majority.

Source: ZenAI snapshot (8 May 2026) → (opens in new tab)

Curiosity Engine

March 2026

Automatic knowledge gap detection with quantified gap score. Analyzes query history, fact density, and confidence — recommends targeted actions.

Gap detection runs on its own and proposes actions rather than waiting for a question. Inspired by Loewenstein's information gap theory (1994).

Source: ZenAI snapshot (8 May 2026) → (opens in new tab)

Prediction Engine

March 2026

User intent prediction from temporal and sequential patterns. Learns from prediction errors — the more often wrong, the better the next prediction.

Among the systems compared here — Mem0, Letta, Zep — none predicts user intent from behavioral patterns within the memory layer itself.

Source: ZenAI snapshot (8 May 2026) → (opens in new tab)

HyperAgent L0–L2

3-level recursive self-improvement with formal safety bounds. Level 0 optimizes knowledge, Level 1 optimizes Level-0 strategies, Level 2 optimizes Level-1 parameters.

3-level recursion with immutable core properties, daily budgets, and automatic rollback on quality regression: self-improvement runs only inside those bounds.

Source: ZenAI snapshot (8 May 2026) → (opens in new tab)

Cross-Context Entity Merging

March 2026

Detection and merging of entities across 4 isolated contexts (Operations, Finance, People, Strategy). Bayesian confidence updates on conflicts.

Among the systems compared here — Mem0, Letta, Zep — none manages entity identity across isolated contexts.

Source: zenbrain → (opens in new tab)
EXPLORE INTERACTIVELY

Explore the system interactively

An accessible, interactive depiction of the memory-based system — from the 7-layer memory and its neuroscience inspiration to how the agents work together. With a toggle between plain-language and technical explanations.

The idea in one sentence: collective intelligence on a cognitive architecture — humans and machines together, on a structure that keeps knowledge. The explorer below makes exactly that architecture tangible.

Open in full screen ↗

These mechanisms run for real — run the open library live →

Interactive depiction · best viewed on desktop

INTERACTIVE · THE NARROW BAND

The principle behind the quality imperative

A working system sits on a narrow edge. Change the survival rule step by step and watch the grid collapse, carry structure, or clog — live. Most rules fail; only a narrow band holds.

Why quality is not optional for us: the very efficiency that makes a system valuable amplifies every rule you give it — for better and for worse.

Open full screen ↗

The full story — why quality is not optional →

Interactive · click the neighbour counts on or off · best on desktop

MEMORY SYSTEM

Seven layers of memory, grounded in neuroscience

The HiMeS architecture, inspired by the Atkinson-Shiffrin model (1968) and modern cognitive science.

↑↓1Working Memory2Short-Term Memory3Episodic4Semantic5Procedural6Core Memory7Cross-Context
Seven differentiated memory layers operating as a coherent system — modeled on the Atkinson-Shiffrin framework and contemporary cognitive neuroscience.
1

Working Memory

Active focus — 7±2 items per Miller's Magical Number. Fastest access, shortest lifespan.

2026
2

Short-Term Memory

Session context and conversation continuity. Survives the current session.

2026
3

Episodic

Concrete experiences with emotional tagging. 400+ keyword lexicon (DE+EN) for arousal/valence scoring.

2026
4

Semantic

Factual knowledge with FSRS scheduling. Spaced repetition optimizes recall timing — 30% better than SM-2.

2026
5

Procedural

Workflows and skills. Tool chains are analyzed and optimized.

2026
6

Core Memory

Immutable foundations following the Letta pattern. Pinned facts that are never forgotten.

2026
7

Cross-Context

Knowledge and entities linked across isolated contexts — with Bayesian confidence updates on conflict.

2026

15 Mechanisms Inspired by Neuroscience

FSRS Spaced Repetition

open-spaced-repetition/fsrs4anki

Optimal review timing at ~90% target retention

Ebbinghaus Forgetting Curve

Ebbinghaus (1885)

Exponential decay with configurable half-life

Hebbian Learning

Hebb (1949)

Co-activated facts strengthen connections (×1.09/activation)

Homeostatic Normalization

Turrigiano (2004)

Prevents runaway growth of edge weights

Sleep Consolidation

Stickgold & Walker (2013)

Hippocampal replay with +50% stability boost

Synaptic Homeostasis

Tononi & Cirelli (2006)

Weak connections pruned during sleep

Emotional Modulation

LeDoux (1996)

Emotional memories decay 2.7× slower

Bayesian Propagation

Pearl (1988)

Confidence updates across the entire knowledge graph

Global Workspace Theory

Baars (1988)

Conscious access through competitive context assembly

Information Gain Scoring

Shannon (1948)

Entropy-based prioritization of new facts

Knowledge Gap Theory

Loewenstein (1994)

Systematic detection of missing knowledge

Working Memory Capacity

Miller (1956)

7±2 active items in working memory

Neuromodulation (PMA)

Schultz (1997) · Aston-Jones (2005)

Dopamine, NE, 5-HT, ACh — four channels with tonic + phasic dynamics

Reconsolidation (PMA)

Nader (2000) · Schiller (2010)

Memory destabilizes on retrieval — four PE-gated update modes with rollback

Triple-Copy Memory (PMA)

Squire & Bayley (2007)

Three traces with divergent dynamics: fast (4h), medium (14d), deep (logarithmic)

RETRIEVAL & RAG

Intelligent knowledge retrieval

6 strategies, dynamically selected per query. Not one pipeline — an adaptive system.

End-to-end RAG pipeline

Self-RAG Critiqueloops back if confidence < 0.5?01Query🧭02A-RAG Plan💭03HyDE🔍04Vector Search⚖️05Cross-Encoder📊06Confidence✓07Answer
Query
↓
A-RAG Plan
↓
HyDE
↓
Vector Search
↓
Cross-Encoder
↓
Confidence
↓
Answer

↻ Self-RAG Critique: loops back if confidence < 0.5

A-RAG Planning

2026

Meta-agent plans retrieval steps before execution. Heuristic-first, LLM fallback.

GraphRAG 3-Layer

2026

Event subgraph + semantic graph + community summaries. 5 parallel strategies.

Self-RAG Critique

2026

Automatic reformulation at confidence < 0.5. 4-component scoring.

HyDE Retrieval

2026

Hypothetical answer → embedding → search. Auto-detection with 5s timeout.

Contextual Retrieval

2026

Chunk enrichment per Anthropic method. +67% retrieval accuracy.

Embedding Drift Detection

2026

BullMQ worker monitors drift >10%. Automatic cache invalidation.

AGENT SYSTEM

Autonomous agents with safety bounds

Multi-agent orchestration with structured debate, dynamic team composition, and recursive self-improvement.

Multi-agent team architecture

HyperAgent L0–L2recursive self-improvementDebate Protocol3 rounds: challenge → response → resolve🧠OrchestratorPersistent Loopspause · resume · cancel🔬Researcher✍️Writer🔍Reviewer💻Coder
Orchestratorrecursive self-improvement · pause · resume · cancel
Researcher
Writer
Reviewer
Coder

3 rounds: challenge → response → resolve

Debate Protocol

2026

3-round debate on disagreement. Challenge → Response → Resolution.

Dynamic Team Builder

2026

5 specialist agents, automatically composed by task type.

HyperAgent L0–L2

2026

Recursive self-improvement with daily budgets, sandbox tests, and auto-rollback.

Persistent Agent Loops

2026

Pause/resume with state checkpointing. Long-running tasks over days.

A2A Protocol

2026

Agent-to-agent communication per Google standard. /.well-known/agent.json discovery.

Implicit Feedback

2026

Automatic behavior detection without explicit labeling.

COGNITIVE ARCHITECTURE

AI that thinks about thinking

Curiosity, prediction, metacognition — three pillars of cognitive intelligence.

Curiosity Engine

2026

Quantified gap score: query frequency × fact density × confidence × RAG quality.

Prediction Engine

2026

Temporal + sequential pattern recognition. Learns from prediction errors.

Metacognition

2026

Confidence calibration, confusion detection, capability profiling.

Adaptive Thinking

2026

4-tier thinking budgets: 1K→16K→64K→128K tokens. Auto-detection. 60-80% cost savings.

PERFORMANCE

91.3% accuracy at 1/109.6 of the tokens

On LongMemEval-500, ZenBrain reaches 91.3% of long-context-oracle accuracy at a per-query token budget of 1/109.6.

In our own measurements on LongMemEval-500, three of nine pairwise comparisons hold, all three against A-Mem, Bonferroni-corrected p ≤ 6.2 × 10⁻³¹ across three independent LLM judges. The six against Letta and Mem0 are ties at the paper's own criterion, judged at equal evidence depth and under version-matched judges; none is lost.

Read the full analysis →

Open and reproducible

ZenBrain is open source. All algorithms, all tests, all documentation — openly available.

View on GitHubDiscover ZenAIGet in Touch

Frequently asked questions about the architecture

What makes ZenBrain different from a vector store?

A vector store searches for similarity in a flat index. ZenBrain models memory as a process: what is used repeatedly consolidates, what is left alone fades. Seven layers separate working memory, episodic, semantic and procedural knowledge, and a consolidation phase at rest condenses experience into knowledge. The difference shows where context has to hold for weeks rather than for one conversation.

Why seven layers instead of one?

Because different kinds of memory need different rules. Working memory has to be fast and volatile, semantic knowledge durable and condensed, procedural memory retrievable without deliberate search. Each layer carries its own decay and consolidation logic; fifteen mechanisms in total, nine foundational plus six in a predictive architecture.

Are the algorithms actually grounded in neuroscience?

Yes, and each one with its source in the paper: FSRS for spacing intervals, Hebbian learning, the Ebbinghaus forgetting curve, Bayesian confidence propagation, a two-factor synaptic model, a vmPFC-coupled prediction-error update, and sleep consolidation. Grounded here means in peer-reviewed neuroscience, not in an analogy.

How does ZenBrain compare to other open memory systems?

On LongMemEval-500, at the same token budget and judged by three independent LLM judges, three of nine head-to-head answer-quality comparisons hold for ZenBrain, all three against A-Mem. The remaining six, against Letta and Mem0, are ties at the paper's own criterion, with none lost. All nine contrasts are Bonferroni-corrected and judged under version-matched judges. It reaches 91.3 percent of the accuracy of a full-context oracle at 1/109.6 of the per-query token budget.

Can I verify the results myself?

Yes, and that is the point. The memory core is open under Apache 2.0, the publications carry DOIs on Zenodo, and the benchmark page gives every figure its method, source and verification path. Anyone who doubts a number should be able to recompute it without asking us.

Is the paper peer-reviewed?

The preprint itself (arXiv:2604.23878) is not peer-reviewed. The neuroscience it builds on is. We state that plainly, because the distinction matters — and because open evidence has to carry a claim, not a reference to a process.

More from this research

Related pages

Core and application fields, on a shared ethics foundation.

  • Research →Three tracks: cognitive architectures, safety in public spaces, applied AI.
  • Adopt AI →A neutral roadmap for AI adoption — prioritisation, stage-gates, KPIs and EU AI Act / GDPR governance, with an interactive simulation.
  • Methodology →Pre-registration, reproducibility, external validation, data minimisation — operational standards in detail.
  • Publications →arXiv, Zenodo (DOI), software releases, open-access principles.
  • Benchmarks →ZenBrain by the numbers: every figure with its method, source and verification link.
  • Resources →Code, replication material, BibTeX citation, licences, identifiers.
  • Research ethics →GDPR Art. 89, AI Act Art. 5, Brokdorf line. Eight mandatory corrections.
  • Public sector & funding →Research offerings for BMBF, BBK, universities and research consortia.
  • Principal Investigator →Profile, background, identifiers, contact paths.
zensationIndependent AI research lab · Kiel
ResearchPublicationsMethodologyResearch ethicsResourcesAdopt AIPublic sectorProtecting public spacesTechnologySystem explorerPlaygroundWork with usOpen SourceAboutBlogChangelog
GitHubLinkedInarXivZenodoORCIDScholarSemantic ScholarHuggingFacenpmDiscussions

© 2026 Alexander Bering / ZenSation Enterprise Solutions

RSSPrivacy PolicyLegal NoticeAccessibility