Learning complex temporal dependencies via local synaptic plasticity
- 1University of Oxford
Abstract
The ability to extract and exploit temporal structure across diverse tasks is central to human cognition. Neuroscientists have typically relied on recurrent neural networks (RNNs) trained with backpropagation through time (BPTT) when modelling neural and behavioural processes such as decision-making and motor control. However, this algorithm has limited biological plausibility, hence the computational principles underlying efficient learning of temporal dependencies remain unresolved. Here, we investigate temporal predictive coding (tPC), a recently proposed framework that extends predictive coding to the temporal domain while preserving local Hebbian update rules. We analyse and extend tPC to establish its relationship with several influential computational models of learning in RNNs, including BPTT, reservoir computing, and eligibility propagation (e-prop). We first demonstrate structurally equivalent weight update rules between tPC and tBPTT1, a variant of BPTT in which gradients are propagated only one time step into the past. We then show that tPC can leverage reservoir dynamics to encode short-range temporal context, and simultaneously sculpt neural trajectories in state space to support downstream readout. We further demonstrate that hierarchical recurrent dynamics can facilitate learning of more complex temporal dependencies, while additionally conferring robustness to strong distractors. Finally, we show that tPC networks can be augmented with biologically inspired eligibility traces to solve temporally extended context-dependent tasks. Together, these results reveal that relatively simple recurrent networks governed by local plasticity can support temporal learning in more complex settings than previously appreciated.
Related
- Genetic and Molecular Mechanisms of Synaptic Plasticity in Perioperative Delirium: From Neuroinflammation to Circuit Dysfunction and Recovery
- Myelin Maintenance and Addition Regulate Synaptic Plasticity in the Adult Mouse Cortex
- Astrocytes Control Cocaine-Induced Synaptic Plasticity and Reward Through the Matricellular Protein Hevin
- Regulation of miR-219 and its target RalGPS is essential for synaptic plasticity at the Drosophila neuromuscular junction
- Human cortical slices reveal pulse-number-dependent synaptic plasticity during theta-burst magnetic stimulation
BibTeX
@misc{ngkeekwong2026learning,
title = {{Learning complex temporal dependencies via local synaptic plasticity}},
author = {J. Ng-Kee-Kwong and M. Tang and T. Akam and R. Bogacz},
howpublished = {bioRxiv},
year = {2026},
month = jul,
doi = {10.64898/2026.07.09.737423},
url = {https://www.biorxiv.org/content/10.64898/2026.07.09.737423v2},
note = {CC BY},
}
RIS
TY - UNPB
TI - Learning complex temporal dependencies via local synaptic plasticity
AU - J. Ng-Kee-Kwong
AU - M. Tang
AU - T. Akam
AU - R. Bogacz
PY - 2026
DA - 2026/07/09
PB - bioRxiv
DO - 10.64898/2026.07.09.737423
UR - https://www.biorxiv.org/content/10.64898/2026.07.09.737423v2
AB - The ability to extract and exploit temporal structure across diverse tasks is central to human cognition. Neuroscientists have typically relied on recurrent neural networks (RNNs) trained with backpropagation through time (BPTT) when modelling neural and behavioural processes such as decision-making and motor control. However, this algorithm has limited biological plausibility, hence the computational principles underlying efficient learning of temporal dependencies remain unresolved. Here, we investigate temporal predictive coding (tPC), a recently proposed framework that extends predictive coding to the temporal domain while preserving local Hebbian update rules. We analyse and extend tPC to establish its relationship with several influential computational models of learning in RNNs, including BPTT, reservoir computing, and eligibility propagation (e-prop). We first demonstrate structurally equivalent weight update rules between tPC and tBPTT1, a variant of BPTT in which gradients are propagated only one time step into the past. We then show that tPC can leverage reservoir dynamics to encode short-range temporal context, and simultaneously sculpt neural trajectories in state space to support downstream readout. We further demonstrate that hierarchical recurrent dynamics can facilitate learning of more complex temporal dependencies, while additionally conferring robustness to strong distractors. Finally, we show that tPC networks can be augmented with biologically inspired eligibility traces to solve temporally extended context-dependent tasks. Together, these results reveal that relatively simple recurrent networks governed by local plasticity can support temporal learning in more complex settings than previously appreciated.
ER -