Biology subjects

Anwar, H.

Publications and source records attributed to Anwar, H..

6 recordsLinked to original sources

bioRxiv2026

Context-Aware Evidence-Gated Plasticity for Multi-Goal Learning in Spiking Neural Networks

Background / IntroductionBiologically inspired spiking neural networks can model adaptive behavior, but learning multiple goals is difficult because synaptic updates for different targets can interfere. We tested whether multi-timescale plasticity and context-specific credit assignment could improve continual multi-goal learning in a spiking navigation system inspired by entorhinal-hippocampal circuitry. MethodsWe developed a closed-loop spiking model containing grid-like, place-like, target-related, association, and motor-output populations. An agent navigated in a two-dimensional environment with randomized starting locations and learned through reward-modulated spike-timing dependent plasticity (STDP/RL) and a novel evidence-gated plasticity (EGP) framework. EGP accumulates candidate synaptic modifications, evaluates them using reward evidence, and consolidates only changes that improve performance. A target-context variant maintained separate proposal stores and reward evaluation for each target. ResultsSTDP/RL learned and retained a single-target navigation policy, but multi-target training produced substantial interference, including attraction to incorrect targets after learning. Across 10 connectivity seeds, target-context EGP achieved higher late-stage reward than global EGP, improved weakest-target performance, and increased the fraction of targets achieving positive reward. In a longer continual-learning simulation, reward increased for all targets, TEST-phase performance increasingly exceeded TRAIN-phase performance, and proposal magnitudes grew over learning. Dwell-time confusion analyses showed that target-context EGP reduced wrong-target attraction and improved target selectivity relative to multi-target STDP/RL. ConclusionsThese results demonstrate that spiking navigation circuits can learn goal-directed behavior using local plasticity, but robust multi-goal learning benefits from context-specific evidence-based consolidation. Target-context EGP provides a biologically motivated mechanism for reducing interference during continual reinforcement learning in spiking neural networks.

neuroscience↗

bioRxiv2022

Inter-animal variability in activity phase is constrained by synaptic dynamics in an oscillatory network

The levels of voltage-gated and synaptic currents in the same neuron type can vary substantially across individuals. Yet, the phase relationships between neurons in oscillatory circuits are often maintained, even in the face of varying oscillation frequencies. We examined whether synaptic and intrinsic currents are matched to maintain constant activity phases across preparations, using the lateral pyloric (LP) neuron of the stomatogastric ganglion of the crab, Cancer borealis. LP produces stable oscillatory bursts upon release from inhibition, with an onset phase that is independent of oscillation frequency. We quantified the parameters that define the shape of the synaptic current inputs across preparations and found no linear correlations with voltage-gated currents. However, several synaptic parameters were correlated with oscillation period and burst onset phase, suggesting they may play a role in phase maintenance. We used the dynamic clamp to apply artificial synaptic inputs and found that those synaptic parameters correlated with phase and period were ineffective in influencing burst onset. Instead, parameters that showed the least variability across preparations had the greatest influence. Thus, parameters that influence circuit phasing are constrained across individuals, while those that have little effect simply co-vary with phase and frequency.

neuroscience↗

bioRxiv2021

Multi-timescale biological learning algorithms train spiking neuronal network motor control

Despite being biologically unrealistic, artificial neural networks (ANNs) have been successfully trained to perform a wide range of sensory-motor behaviors. In contrast, the performance of more biologically realistic spiking neuronal network (SNN) models trained to perform similar behaviors remains relatively suboptimal. In this work, we aimed at pushing the field of SNNs forward by exploring the potential of different learning mechanisms to achieve optimal performance. Inspired by biological learning mechanisms operating at multiple timescales, we used spike-timing-dependent reinforcement learning (STDP-RL) and evolutionary strategy (EVOL) with SNNs to solve the CartPole reinforcement learning (RL) control problem. Though the role of STDP-RL in biological systems is well established, several other mechanisms, though not fully understood, work in concert during learning in vivo. Recreating accurate models that capture the interaction of STDP-RL with these diverse learning mechanisms is extremely difficult. EVOL is an alternative method, and has been successfully used in many studies to fit model neural responsiveness to electrophysiological recordings and in some cases for classification problems. One advantage of EVOL is that it may not need to capture all interacting components of synaptic plasticity, and thus provides a better alternative to STDP-RL. Here, we compared the performance of each algorithm after training, which revealed EVOL as a powerful method to training SNNs to perform sensory-motor behaviors. Our modeling opens up new capabilities for SNNs in RL and could serve as a testbed for neurobiologists aiming to understand multi-timescale learning mechanisms and dynamics in neuronal circuits.

neuroscience↗

bioRxiv2021

Training a spiking neuronal network model of visual-motor cortex to play a virtual racket-ball game using reinforcement learning

Recent models of spiking neuronal networks have been trained to perform behaviors in static environments using a variety of learning rules, with varying degrees of biological realism. Most of these models have not been tested in dynamic visual environments where models must make predictions on future states and adjust their behavior accordingly. The models using these learning rules are often treated as black boxes, with little analysis on circuit architectures and learning mechanisms supporting optimal performance. Here we developed visual/motor spiking neuronal network models and trained them to play a virtual racket-ball game using several reinforcement learning algorithms inspired by the dopaminergic reward system. We systematically investigated how different architectures and circuit-motifs (feed-forward, recurrent, feedback) contributed to learning and performance. We also developed a new biologically-inspired learning rule that significantly enhanced performance, while reducing training time. Our models included visual areas encoding game inputs and relaying the information to motor areas, which used this information to learn to move the racket to hit the ball. Neurons in the early visual area relayed information encoding object location and motion direction across the network. Neuronal association areas encoded spatial relationships between objects in the visual scene. Motor populations received inputs from visual and association areas representing the dorsal pathway. Two populations of motor neurons generated commands to move the racket up or down. Model-generated actions updated the environment and triggered reward or punishment signals that adjusted synaptic weights so that the models could learn which actions led to reward. Here we demonstrate that our biologically-plausible learning rules were effective in training spiking neuronal network models to solve problems in dynamic environments. We used our models to dissect the circuit architectures and learning rules most effective for learning. Our model shows that learning mechanisms involving different neural circuits produce similar performance in sensory-motor tasks. In biological networks, all learning mechanisms may complement one another, accelerating the learning capabilities of animals. Furthermore, this also highlights the resilience and redundancy in biological systems.

neuroscience↗

bioRxiv2020

Taxonomy of neural oscillation events in primate auditory cortex

Electrophysiological oscillations in the brain have been shown to occur as multi-cycle events, with onset and offset dependent on behavioral and cognitive state. To provide a baseline for state-related and task-related events, we quantified oscillation features in resting-state recordings. We used two invasively-recorded electrophysiology datasets: one from human, and one from non-human primate auditory system. After removing incidentally occuring event related potentials, we used a wavelet transform based method to quantify oscillation features. We identified about 2 million oscillation events, classified within traditional frequency bands: delta, theta, alpha, beta, low gamma, gamma, high gamma. Oscillation events of 1-44 cycles were present in at least one frequency band in 90% of the time in human and non-human primate recordings. Individual oscillation events were characterized by non-constant frequency and amplitude. This result naturally contrasts with prior studies which assumed such constancy, but is consistent with evidence from event-associated oscillations. We measured oscillation event duration, frequency span, and waveform shape. Oscillations tended to exhibit multiple cycles per event, verifiable by comparing filtered to unfiltered waveforms. In addition to the clear intra-event rhythmicity, there was also evidence of inter-event rhythmicity within bands, demonstrated by finding that coefficient of variation of interval distributions and Fano Factor measures differed significantly from a Poisson distribution assumption. Overall, our study demonstrates that rhythmic, multi-cycle oscillation events dominate auditory cortical dynamics.

neuroscience↗

bioRxiv2019

Short-Term Synaptic Dynamics Control the Activity Phase of Neurons in an Oscillatory Network

In oscillatory systems, neuronal activity phase is often independent of network frequency. Such phase maintenance requires adjustment of synaptic input with network frequency, a relationship that we explored using the crab, Cancer borealis, pyloric network. The burst phase of pyloric neurons is relatively constant despite a >2-fold variation in network frequency. We used noise input to characterize how input shape influences burst delay of a pyloric neuron, and then used dynamic clamp to examine how burst phase depends on the period, amplitude, duration, and shape of rhythmic synaptic input. Phase constancy across a range of periods required a proportional increase of synaptic duration with period. However, phase maintenance was also promoted by an increase of amplitude and peak phase of synaptic input with period. Mathematical analysis shows how short-term synaptic plasticity can coordinately change amplitude and peak phase to maximize the range of periods over which phase constancy is achieved.

neuroscience↗