Distributed Cerebellar Motor Learning: A Spike-Timing-Dependent Plasticity Model

Abstract

Deep cerebellar nuclei neurons receive both inhibitory (GABAergic) synaptic currents from Purkinje cells (within the cerebellar cortex) and excitatory (glutamatergic) synaptic currents from mossy fibers. Those two deep cerebellar nucleus inputs are thought to be also adaptive, embedding interesting properties in the framework of accurate movements. We show that distributed spike-timing-dependent plasticity mechanisms (STDP) located at different cerebellar sites (parallel fibers to Purkinje cells, mossy fibers to deep cerebellar nucleus cells, and Purkinje cells to deep cerebellar nucleus cells) in close-loop simulations provide an explanation for the complex learning properties of the cerebellum in motor learning. Concretely, we propose a new mechanistic cerebellar spiking model. In this new model, deep cerebellar nuclei embed a dual functionality: deep cerebellar nuclei acting as a gain adaptation mechanism and as a facilitator for the slow memory consolidation at mossy fibers to deep cerebellar nucleus synapses. Equipping the cerebellum with excitatory (e-STDP) and inhibitory (i-STDP) mechanisms at deep cerebellar nuclei afferents allows the accommodation of synaptic memories that were formed at parallel fibers to Purkinje cells synapses and then transferred to mossy fibers to deep cerebellar nucleus synapses. These adaptive mechanisms also contribute to modulate the deep-cerebellar-nucleus-output firing rate (output gain modulation toward optimizing its working range).

Modeling the Cerebellar Microcircuit: New Strategies for a Long-Standing Issue

“…Plasticity has been reported not just in acute brain slices but also in vivo (Jörntell and Ekerot, 2002 ; Roggeri et al, 2008 ; Diwakar et al, 2011 ; Johansson et al, 2014 ; Ramakrishnan et al, 2016 ), revealing that patterned sensory inputs can determine a complex set of changes encompassing multiple synaptic relays. Importantly several of the cerebellar synapses may show forms of spike-timing-dependent plasticity (STDP), linking intracerebellar oscillations to the ability of generating plasticity (D’Angelo et al, 2015 ; Garrido et al, 2016 ; Luque et al, 2016 ). Understanding the importance of these forms of plasticity may greatly benefit from integrated network modeling.…”

“…The ultimate challenge appears then to run the whole-cerebellum network model in a simulated brain operating in closed-loop. While a radical approach is out of reach at the moment (it would require, in addition to fully developed cerebellum models, also realistic models of large brain sections outside the cerebellum), a first attempt has been done by reducing the complexity of cerebellar models and using simplified versions to run closed-loop robotic simulations (Casellato et al, 2012 , 2014 , 2015 ; Garrido et al, 2013 ; Luque et al, 2014 , 2016 ).…”

“…Clearly, in this case a top-down approach is adopted and the relationship of the simplified model with the real system is a matter of speculation. This approach has been used to generate cerebellar spiking networks (SNN) allowing to reproduce a single basic cerebellar module running with high efficiency in a robotic controller yet maintaining some fundamental features of neurons and connections (Casellato et al, 2012 , 2014 , 2015 ; Garrido et al, 2013 ; Luque et al, 2014 , 2016 ). For example, in these models, neurons were represented by integrate-and-fire single-compartment elements, the local inhibitory interneuron networks were not included and the GCL was not fully implemented resorting to the concept of a non-recurrent states in a liquid-state machine (Yamazaki and Tanaka, 2007 ).…”

Inhibitory Plasticity: From Molecules to Computation and Beyond

“…By using a supervised learning scheme for feedforward and recurrent connections, Gilra and Gestner showed that IPs could efficiently accomplish linear, non-linear, or chaotic dynamics, as well as motor coordination dynamics [138]. Similarly, the implementation of STDP rules at different sites in a cerebellar like structure allowed to implement an efficient adaptive scheme capable of motor learning performance [139]. Finally, the specificity of inhibitory feedback sustaining grid cell organization has been suggested to require IP for the generation of grid cell population (Table 1) [140].…”

Homeostatic cerebellar compensation of age-related changes of vestibulo-ocular reflex adaptation: a computational epidemiology study

“…(Badura et al, 2016, Clopath et al, 2014, Arenz et al, 2008, Lisberger and Fuchs, 1978. MF responses consisted of non-overlapping activations of equally sized neural subpopulations, which maintained a constant overall firing rate (Luque et al, 2016).…”