EMBER: Autonomous Cognitive Behaviour from Learned Spiking Neural Network Dynamics in a Hybrid LLM Architecture

View PDF HTML (experimental) Abstract:We present (Experience-Modulated Biologically-inspired Emergent Reasoning), a hybrid cognitive architecture that reorganises the relationship between large language models (LLMs) and memory: rather than augmenting an LLM with retrieval tools, we place the LLM as a replaceable reasoning engine within a persistent, biologically-grounded associative substrate. The architecture centres on a 220,000-neuron spiking neural network (SNN) with spike-timing-dependent plasticity (STDP), four-layer hierarchical organisation (sensory/concept/category/meta-pattern), inhibitory E/I balance, and reward-modulated learning. Text embeddings are encoded into the SNN via a novel z-score standardised top-k population code that is dimension-independent by construction, achieving 82.2\% discrimination retention across embedding dimensionalities. We show that STDP lateral propagation during idle operation can trigger and shape LLM actions without external prompting or scripted triggers: the SNN determines when to act and what associations to surface, while the LLM selects the action type and generates content. In one instance, the system autonomously initiated contact with a user after learned person-topic associations fired laterally during an 8-hour idle period. From a clean start with zero learned weights, the first SNN-triggered action occurred after only 7 conversational exchanges (14 messages). Comments: Preprint. 9 pages, 2 figures, 3 tables. NeurIPS 2026 format Subjects: Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE) Cite as: arXiv:2604.12167 [cs.AI] (or arXiv:2604.12167v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2604.12167 arXiv-issued DOI via DataCite (pending registration) Submission history From: William Savage [view email] [v1] Tue, 14 Apr 2026 00:51:47 UTC (131 KB)