Short-Term Synaptic Plasticity Stabilizes Goal-Conditioned Dynamics in a PFC-Inspired Reservoir Model for Multistep Goal-Directed Action Planning
Authors: Jin Nakamura, Yuichi Katori
Organizations: Graduate School of Information Science and Technology, The University of Osaka, 1-5 Yamadaoka, Suita, 565-0871, Osaka, Japan · School of Systems Information Science, Future University Hakodate, 116-2 Kamedanakano-cho, Hakodate, 041-8655, Hokkaido, Japan
Abstract
The prefrontal cortex (PFC) maintains goal information for action planning, but how recurrent circuits preserve it in an action-usable form over behavioral timescales remains unclear. Here we ask whether short-term synaptic plasticity (STP) can stabilize goal information as action-usable, goal-conditioned dynamics. We incorporated STP into a PFC-inspired reservoir computing model with basal-ganglia-inspired temporal-difference readout learning, and evaluated paired models with and without STP across 100 independently generated networks in a multistep goal-directed action-selection task with delayed execution. Goal identity was highly decodable during the delay even without STP, so STP was not required to form a linearly readable goal representation. Under state noise, however, success without STP fell from 75.8% to 49.5%, whereas the model with STP remained essentially unchanged (91.8% without noise versus 89.2% under noise; paired Cohen's dz=1.31). Time-resolved decoding, state-space separability, and action-value-difference analyses showed that STP preserved goal information as action-relevant goal-conditioned dynamics available at later action opportunities. Gain-matched and STP-state perturbation controls argued against a simple fixed recurrent-scaling explanation and supported online, history-dependent synaptic modulation. Effective-connectivity analyses showed delay-period goal-specific patterning that increased toward the later part of the trial with STP, where it should be read as goal- and task-state-conditioned patterning; effective connectivity without STP was time-invariant. A grid search identified a facilitation-dominant range of STP time constants associated with high success rates. These results suggest that STP supports robust goal-conditioned dynamics through dynamic modulation of goal-dependent effective recurrent connectivity.
Short-term plasticity (STP) is fundamental to temporal information processing in biological neural systems but remains difficult to realize efficiently in neuromorphic hardware. Memristive electrochemical random-access memory (ECRAM) devices naturally exhibit non-equilibrium ionic dynamics that produce transient conductance modulation; however, these behaviors are typically treated as undesirable variability or tolerated as side effects in memory-centric computing paradigms. In this work, we instead transform these volatile dynamics from a tolerated device artifact into a computational resource through a cross-layer device-circuit-system co-design framework. We introduce a delay-feedback leaky integrate-and-fire (LIF) neuron architecture co-designed with ECRAM synapses that exploits activity-dependent conductance modulation with negligible additional circuit overhead. The architecture integrates ECRAM-based synapses with a tunable delay-feedback spike-generation path, enabling transient device dynamics to directly modulate neuron excitability and synaptic efficacy. We used experimentally characterized ECRAM devices exhibiting transient conductance modulation (1.5 KOhms per spike) to develop a compact behavioral model suitable for circuit-level simulation. Circuit simulations demonstrate two key STP behaviors -- synaptic facilitation and intrinsic excitability modulation -- while consuming 2 pJ per spike, and the same device-driven mechanisms extend across multiple neuron topologies. Network-level analysis further demonstrates frequency-selective spike processing, allowing individual synapses to act as tunable temporal filters within spiking neural networks. This work demonstrates that non-equilibrium ECRAM dynamics can serve as a native hardware substrate for STP and temporal computation in neuromorphic circuits.
Alex Currie, Sean Borkholder, Nithil Harris Manimaran +4
Adaptive behavior requires the brain to transition between distinct contexts while maintaining representations of prior experience. The ability to reconfigure neural representations without erasing previously acquired knowledge is central to learning in dynamic environments, yet the neural mechanisms that support this balance remain unclear. Understanding these mechanisms is also critical for addressing catastrophic forgetting in artificial systems designed for lifelong learning. Here, we identify joint sparse coding and temporal dynamics in both the mouse medial prefrontal cortex (mPFC) and computational networks as mechanisms that help preserve prior representations during context transitions. Specifically, sparsity in context-dependent representations reduces cross-context interference, whereas temporal dynamics within the network activity further enhance context separability across time. Strikingly, networks endowed with both properties, such as spiking neural networks, exhibit improved retention during lifelong learning without auxiliary heuristics. These findings establish joint sparse coding and temporal dynamics as a core mechanism supporting flexible context reconfiguration in lifelong learning and, through their activity constraining nature, as an energy-efficient architectural principle for stable adaptation. Together, they provide a mechanistic framework for understanding how the brain preserves prior knowledge while flexibly adapting to new contexts.
A hallmark of intelligence is the ability to adapt in non-stationary environments, yet deep Reinforcement Learning (RL) agents often struggle in such settings. Prior studies introduce non-stationarity through abrupt shifts in features or dynamics, whereas real-world environments often evolve gradually through continual drift. This distinction has important implications for the "stability-plasticity dilemma" in RL, as abrupt task changes may demand more plasticity than naturalistic settings. To address this, we modify existing 3D Miniworld and MuJoCo environments to incorporate naturalistic, continual non-stationarity, and use them to examine how stability and adaptation affect performance under continuous environmental change. We find that methods favoring stability, such as synaptic consolidation, outperform approaches focused on plasticity, such as parameters resetting. Motivated by this result, and prior evidence that Successor Features (SFs) reduce interference, we investigate whether SFs are better consolidation targets than Q-values. Across both environments, applying neuro-inspired synaptic consolidation to SFs yields superior performance on continually changing settings. Moreover, consolidation is most effective when SFs are stabilized across multiple timescales, which capture complementary aspects of gradual environmental change. Together, these results suggest that stability is more critical in continual learning when changes are gradual, and that multi-timescale consolidation of predictive representations is an effective approach.