Biologically-inspired memory architectures for LLM agents go beyond flat vector stores by modeling the dynamic, associative nature of human memory. SYNAPSE introduces a unified episodic-semantic graph with spreading activation, while E-mem uses multi-agent episodic context reconstruction to preserve reasoning integrity over long horizons.

Standard RAG-based agent memory uses embedding similarity to retrieve relevant context. This approach suffers from:

• Static retrieval: Cannot capture dynamic relevance that emerges from the query's relationship to the full memory graph
• Weak-signal blindness: Misses distantly related but important memories that have low direct similarity
• Context fragmentation: Retrieved chunks lose their temporal and causal relationships
• Flat structure: Cannot distinguish between raw episodic memories and abstract semantic knowledge

Biologically-inspired approaches address these limitations by modeling memory as cognitive science suggests humans do – through interconnected networks where activation spreads along associative pathways.

SYNAPSE (arXiv:2601.02744) introduces a brain-inspired memory architecture that unifies episodic and semantic memories in a directed graph with spreading activation dynamics.

The memory graph contains two types of nodes:

• Episodic nodes: Raw interaction turns with text content, embeddings, and timestamps – the “what happened” memory
• Semantic nodes: Abstract concepts, entities, and preferences extracted periodically by the LLM – the “what I know” memory

Three types of edges connect the graph:

• Temporal edges: Chronological links between episodic nodes (recency)
• Abstraction edges: Bidirectional links between episodic and semantic nodes (generalization/instantiation)
• Association edges: Correlations between semantic nodes (conceptual relatedness)

When a query arrives, the retrieval process works as follows:

1. Embedding injection: The query embedding activates the most similar nodes in the graph
2. Energy propagation: Activation “energy” spreads outward through edges:
   • Temporal edges propagate based on recency
   • Abstraction edges propagate between specific memories and general concepts
   • Association edges propagate between related concepts
3. Convergence: After multiple propagation steps, the activation pattern stabilizes
4. Context assembly: The highest-activated nodes (both episodic and semantic) are assembled into context for the LLM

This emulates the cognitive science model of spreading activation in human memory: when you think of “dog,” activation spreads to “pet,” “bark,” “walk,” and eventually reaches more distant concepts like “veterinarian” or “loyalty.”

• Weak-signal retrieval: Discovers relevant memories that have low direct similarity but strong associative connections
• Dynamic relevance: The same memory graph produces different retrieval results for different queries, based on the activation pattern
• Noise filtering: Low-relevance nodes receive insufficient activation to cross the retrieval threshold
• Emergent context: Combines episodic and semantic information naturally through the propagation dynamics

• Outperforms 10 baselines across system-level, graph-based, retrieval, and agentic categories
• Hybrid episodic-semantic design achieves Pareto frontier for efficiency and consistency
• Removing spreading activation drops performance to static graph levels (average score 30.5 vs. full system)
• Pure vector retrieval scores only 25.2 on average – activation is essential for long-horizon reasoning

E-mem (Wang et al., 2026, arXiv:2601.21714) shifts from memory preprocessing to episodic context reconstruction, addressing the problem of “destructive de-contextualization” in traditional memory systems.

Traditional memory preprocessing methods (embeddings, graphs, summaries) compress complex sequential dependencies into pre-defined structures. This severs the contextual integrity essential for System 2 (deliberative) reasoning:

• Embeddings lose sequential structure
• Graphs lose nuance and narrative flow
• Summaries lose details needed for precise reasoning

Inspired by biological engrams (the physical traces of memories in neural tissue):

• Master agent: Orchestrates global planning and coordinates retrieval across the full memory
• Assistant agents: Each maintains an uncompressed segment of the memory context – no lossy preprocessing
• Active reasoning: Unlike passive retrieval, assistants locally reason within their activated memory segments, extracting context-aware evidence before aggregation

1. Master agent receives a query and formulates a retrieval plan
2. Relevant assistant agents are activated (analogous to engram activation in neuroscience)
3. Each activated assistant reasons over its uncompressed memory segment
4. Assistants extract context-aware evidence (not just raw text chunks)
5. Master agent aggregates evidence from all assistants into a coherent response

• LoCoMo benchmark: Over 54% F1 score
• Surpasses state-of-the-art GAM by 7.75%
• Reduces token cost by over 70% compared to full-context approaches
• Preserves reasoning integrity that preprocessing-based methods destroy

import numpy as np
from collections import defaultdict
 
class SpreadingActivationMemory:
    """SYNAPSE-style episodic-semantic memory with spreading activation."""
 
    def __init__(self, decay=0.85, threshold=0.1, max_steps=5):
        self.nodes = {}        # id -> {type, content, embedding}
        self.edges = defaultdict(list)  # id -> [(target_id, edge_type, weight)]
        self.decay = decay
        self.threshold = threshold
        self.max_steps = max_steps
 
    def add_episodic(self, node_id, content, embedding, timestamp):
        self.nodes[node_id] = {
            "type": "episodic", "content": content,
            "embedding": embedding, "timestamp": timestamp
        }
 
    def add_semantic(self, node_id, concept, embedding):
        self.nodes[node_id] = {
            "type": "semantic", "content": concept,
            "embedding": embedding
        }
 
    def add_edge(self, source, target, edge_type, weight=1.0):
        self.edges[source].append((target, edge_type, weight))
 
    def retrieve(self, query_embedding, top_k=10):
        """Spreading activation retrieval."""
        # Step 1: Initial activation from query similarity
        activation = {}
        for nid, node in self.nodes.items():
            sim = np.dot(query_embedding, node["embedding"])
            if sim > self.threshold:
                activation[nid] = sim
 
        # Step 2: Spread activation through edges
        for step in range(self.max_steps):
            new_activation = dict(activation)
            for nid, energy in activation.items():
                if energy < self.threshold:
                    continue
                for target, etype, weight in self.edges.get(nid, []):
                    spread = energy * self.decay * weight
                    new_activation[target] = max(
                        new_activation.get(target, 0), spread
                    )
            activation = new_activation
 
        # Step 3: Return top-k activated nodes
        ranked = sorted(activation.items(), key=lambda x: -x[1])
        return [(self.nodes[nid], score) for nid, score in ranked[:top_k]]

Both SYNAPSE and E-mem draw on established models from cognitive science and neuroscience:

• Spreading activation (Collins & Loftus, 1975): Retrieval from human semantic memory works by activation spreading through associative networks, not by direct lookup
• Episodic-semantic distinction (Tulving, 1972): Human memory maintains separate but interconnected systems for specific experiences and general knowledge
• Engrams (Tonegawa et al., 2015): Physical memory traces in neural tissue are reactivated during recall – E-mem's assistant agents model this reactivation process
• Consolidation: Over time, episodic memories are abstracted into semantic knowledge – SYNAPSE's periodic LLM extraction of semantic nodes mirrors this process

• SYNAPSE: Brain-Inspired Episodic-Semantic Memory with Spreading Activation (arXiv:2601.02744)
• E-mem: Multi-agent based Episodic Context Reconstruction for LLM Agent Memory (arXiv:2601.21714)

• Agentic Uncertainty – how memory quality affects downstream confidence
• KV Cache Compression – complementary approach to managing long contexts
• Persona Simulation – memory systems for maintaining persona consistency