Motor primitives in space and time via targeted gain modulation in cortical networks

Abstract

Motor cortex (M1) exhibits a rich repertoire of neuronal activities to support the generation of complex movements. Although recent neuronal-network models capture many qualitative aspects of M1 dynamics, they can generate only a few distinct movements. Additionally, it is unclear how M1 efficiently controls movements over a wide range of shapes and speeds. We demonstrate that modulation of neuronal input–output gains in recurrent neuronal-network models with fixed architecture can dramatically reorganize neuronal activity and thus downstream muscle outputs. Consistent with the observation of diffuse neuromodulatory projections to M1, a relatively small number of modulatory control units provide sufficient flexibility to adjust high-dimensional network activity using a simple reward-based learning rule. Furthermore, it is possible to assemble novel movements from previously learned primitives, and one can separately change movement speed while preserving movement shape. Our results provide a new perspective on the role of modulatory systems in controlling recurrent cortical activity.

Interactions of encoding and decoding problems to understand motor control

“…Neural dynamics (position activity maps and velocity neural tuning curves) learned by neural encoding model have a special name in the literature [36][37][38][39][40][41] -Motor primitives. Motor cortex is believed to control movement through flexible combinations of motor primitives, elementary building blocks that can be combined and composed to give rise to complex motor behavior.…”

“…They built movement trajectories through linear combinations of those velocity tuning curves. In related research, Stround et al 41 used gain patterns over neurons or neural groups to predict movement trajectories.…”

High capacity and dynamic accessibility in associative memory networks with context-dependent neuronal and synaptic gating

“…Alternatively, neuronspecific gating could be implemented through excitatory control -e.g., neurons may be in a suppressed state by default, and only participate in recall if they receive extra contextual excitation -supported by recent work showing that baseline shifts modulate free recall 17 . Finally, recent evidence suggests that gain or excitability changes in individual neurons may play a role in memory allocation [63][64][65][66] , and computational work has applied this idea to motor learning 22 and sequence learning 67 . Experimental evidence suggests that around 10 − 30% of neurons are allocated for a given engram in the amygdala and hippocampus 65 , which would correspond to an area of high capacity in our model of neuron-specific gating.…”

“…Recent evidence suggests a role for excitation 17 , but also for different inhibitory cell types in controlling top-down modulation 18,19 . Such modulation may place the network into different states for storage and retrieval of memory 20,21 -e.g., through modulation 22 , or changes in the balance of excitation and inhibition 23,24 . Despite the clear evidence for such context-dependent state changes, and their proposed use in models of other cognitive functions 25,26 , to our best knowledge they have not yet been included in models of (auto-)associative memory.…”

The remarkable robustness of surrogate gradient learning for instilling complex function in spiking neural networks

“…For instance, performance-optimized artificial neural networks bear striking representational similarity with the visual system [5][6][7][8][9][10][11] and serve to formulate hypotheses about their mechanistic underpinnings. Similarly, the activity of artificial recurrent neural networks optimized to solve cognitive tasks resembles cortical activity in prefrontal [12,13], medial frontal [14], and motor areas [15,16] and inspire comparison and lively discussion.…”