Feedback-regulated dual-role memory consolidation for continual learning: a stability–plasticity framework inspired by hippocampo-cortical systems consolidation
Aug 2026· Cognitive Neurodynamics· Vol 20· 0 citations· 51 references
Medicine
TL;DR
The results support DRMCL as a stability-oriented computational framework; they do not establish a circuit-level model of hippocampo-cortical consolidation or a new EEG decoding benchmark.
Together, these results link representational drift to the stability--plasticity trade-off: its magnitude is shaped by the mechanism that protects old knowledge, and suppressing it can restrict future learning.
Simple Summary Animals adapt their decisions by combining current information with previous experience. During learning, they may shift from flexible strategies based on environmental structure toward more stable choices guided by learned values. In this study, pigeons performed a sequential decision-making task while neural activity was recorded from the hippocampus and striatum. Behavioral modeling showed a gradual shift toward value-guided choice, accompanied by changes in communication between these brain regions. In particular, gamma-band information flow from the hippocampus to the striatum decreased as task performance became more stable. These preliminary findings suggest that learning is accompanied by frequency-specific changes in brain communication and provide new insights into the neural basis of adaptive decision-making in birds.
Li-Fang Yang, Ying Ma, Zhi-Hui Li et al.· Animals· 0 citations
Incremental learning (IL) aims to continually acquire new knowledge (plasticity) while retaining previously learned information (stability). However, striking a balance between plasticity and stability remains a significant challenge for intelligent systems. The human brain achieves exceptional balance, owing to various memory units that collaboratively encode and store information. Inspired by human memory mechanisms, this paper introduces an IL framework with knowledge enCoding, Consolidation, and Integration (Co $^{2}$ In). Co $^{2}$ In is designed as a bi-level memory architecture: a working memory for adaptive knowledge acquisition and a long-term memory dedicated to the persistent retention of information. The two memory modules work cooperatively during IL. The working memory first learns from new data to encode knowledge into parameters. Subsequently, Co $^{2}$ In performs a consolidation process to identify underlying patterns in the learned parameters and re-express them into a compact knowledge representation. Next, the knowledge representation and the identified patterns are transformed into separate network layers and integrated into the long-term memory. These designs empower Co $^{2}$ In to accumulate knowledge with high plasticity and stability. We evaluate Co $^{2}$ In on CIFAR-10, CIFAR-100, and Tiny-ImageNet under exemplar-free Class-IL and Task-IL settings. Experimental results show that Co $^{2}$ In achieves state-of-the-art performance with efficient memory consumption.
Wenju Sun, Qingyong Li, Boyang Li et al.· IEEE Transactions on Pattern...· 0 citations
Test-Time Memory Calibration (TTMC), a novel gradient-free analytic framework that introduces a transductive calibration mechanism that seamlessly fuses the second-order statistics of the unlabelled test stream into the accumulated long-term memory via a closed-form solution, allowing for real-time alignment with the test distribution.
Yuyang Han, Zi-Yu Li, Diwei Su et al.· Proceedings of the 32nd ACM...· 0 citations
The results suggest that the empirical success of memory replay goes beyond the mitigation of forgetting, actively reintroducing the benefits of data co-observation into the learning process.
Timm Hess, Abhishek Jha, Gido M. van de Ven et al.· 0 citations
Vision Transformers provide strong visual representations but typically rely on slowly updated parameters, limiting their ability to organize newly acquired information across different memory timescales. This work proposes \textit{Hierarchical Hebbian Memory}, a three-level memory architecture composed of rapid Working Memory, persistent Routed Episodic Memory, and slower Semantic Memory. A learned controller regulates memory contribution, read and write routing, plasticity, retention, and consolidation. A causal read-before-write lifecycle ensures that the current outcome cannot influence the prediction it supervises. The architecture is evaluated on Omniglot 5-way 1-shot recognition and CORe50 continual object recognition. With Swin-Tiny, the hierarchical model reaches 97.39\% accuracy on Omniglot and 95.37\% final accuracy on CORe50 when combined with experience replay. Learned multi-bank retrieval reaches 47.50\% delayed-association accuracy, compared with 24.17\% for a single persistent bank and 25.00\% without memory. After intervening distractors, Episodic Memory retains approximately 0.96 cosine similarity with stored associations, while Working Memory falls to approximately 0.05. These results show that Hebbian association and learned memory routing can jointly organize online visual experience across rapid, persistent, and consolidated memory timescales within Vision Transformers.
Mohammed Yusuf Mujawar, Noorbakhsh Amiri Golilarz· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
