A model of flexible motor sequencing through thalamic control of cortical dynamics

Logiaco, Laureline; Abbott, L. F.; Escola, Sean

doi:10.1101/2019.12.17.880153

Cited by 16 publications

(32 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ance during preparation (by construction; Figure 7B, to activate those neurons that are responsible for the 750 next loop in the sequence (Logiaco et al, 2019). In-751 terestingly, their cortical network must still be properly 752 initialized prior to each movement chunk, as it must in 753 our model.…”

mentioning

confidence: 99%

“…This is consistent with the causal role of thalamus in the preparation of directed licking in mice (Guo et al, 2017). Moreover, we posit that the basal ganglia operate an on/off switch on the thalamocortical loop (Jin and Costa, 2010; Cui et al, 2013; Halassa and Acsády, 2016; Logiaco et al, 2019), thereby flexibly controlling the timing of both movement planning and initiation.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Optimal anticipatory control as a theory of motor preparation: a thalamo-cortical circuit model

Kao

Sadabadi

Hennequin

2020

Preprint

View full text Add to dashboard Cite

Across a range of motor and cognitive tasks, cortical activity can be accurately described by low-dimensional dynamics unfolding from specific initial conditions on every trial. These "preparatory states" largely determine the subsequent evolution of both neural activity and behaviour, and their importance raises questions regarding how they are -or ought to be -set. Here, we formulate motor preparation as optimal prospective control of future movements. The solution is a form of internal control of cortical circuit dynamics, which can be implemented as a thalamo-cortical loop gated by the basal ganglia. Critically, optimal control predicts selective quenching of variability in components of preparatory population activity that have future motor consequences, but not in others. This is consistent with recent perturbation experiments performed in mice, and with our novel analysis of monkey motor cortex activity during reaching. Together, these results suggest optimal anticipatory control of movement.Fast ballistic movements (e.g. throwing) require spa-1 lamocortical loop during motor preparation, with tha-62 lamic afferents providing the desired optimal control in-63 puts. This is consistent with the causal role of thalamus 64 in the preparation of directed licking in mice (Guo et al., 65 2017). Moreover, we posit that the basal ganglia oper-66 ate an on/off switch on the thalamocortical loop (Jin 67 and Costa, 2010; Cui et al., 2013; Halassa and Acsády, 68 2016; Logiaco et al., 2019), thereby flexibly controlling 69 the timing of both movement planning and initiation.70We further analyze the model, and formulate predic-71 tions which we have successfully tested in data. At the 72 most abstract level, our core prediction is that the "op-73 timal subspace" is likely high dimensional, with many 74 different initial conditions giving rise to the same cor-75 rect movement. This has an important consequence for 76 preparatory control: at the population level, only a few 77 components of preparatory activity impact future mo-78 tor outputs, and it is these components only that need 79 active controlling. In contrast, one expects substantial 80 pre-movement variability in other, inconsequential com-81 ponents. Concretely, we predict that following a pertur-82 bation, preparatory activity should recover only in state 83 space directions that matter for subsequent movement, 84 but not (necessarily) in others. We find that this pre-85 diction agrees with the effects of optogenetic perturba-86 tions reported by Svoboda and colleagues, in a directed 87 licking task in mice (Li et al., 2016). Furthermore, the 88 existence of a preparatory nullspace predicts selective 89 variability quenching at preparation onset: trial-by-trial 90 variability should drop predominantly in components 91 that have motor consequences. We perform novel analy-92 ses of monkey M1 and dorsal premotor cortex (PMd) ac-93 tivity recorded during reaching, and find that the struc-94 ture of variability quenching supports our main predic-95 tion. Finally...

show abstract

mentioning

confidence: 99%

mentioning

confidence: 99%

Optimal anticipatory control as a theory of motor preparation: a thalamo-cortical circuit model

Kao

Sadabadi

Hennequin

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Modelling studies so far have either focused on the study of sequential dynamics (Chenkov et al 2017;Billeh and Schaub 2018;Setareh et al 2018;Spreizer et al 2019) or on motif acquisition (Stroud et al 2018;Logiaco et al 2019;Maes et al 2020). This paper introduces an explicitly hierarchical model as a fundamental building block for the learning and replay of sequential dynamics of a compositional nature.…”

Section: From Serial To Hierarchical Modellingmentioning

confidence: 99%

“…Here, we present a model for learning temporal sequences on multiple scales implemented through a hierarchical network of bio-realistic spiking neurons and synapses. In contrast to current models, which focus on acquiring the motifs and speculate on the mechanisms to learn a syntax (Stroud et al 2018;Logiaco et al 2019;Maes et al 2020), our spiking network model learns motifs and syntax independently from a target sequence presented repeatedly. Furthermore, the plasticity of the synapses is entirely local, and does not rely on a global optimisation such as FORCE-training (Nicola and Clopath 2017;Hardy and Buonomano 2018;Nicola and Clopath 2019) or backpropagation through time (Werbos 1990).…”

Section: Introductionmentioning

confidence: 99%

Learning compositional sequences with multiple time scales through a hierarchical network of spiking neurons

Maes

Barahona

Clopath

2020

Preprint

View full text Add to dashboard Cite

Sequential behaviour is often compositional and organised across multiple time scales: a set of individual elements developing on short time scales (motifs) are combined to form longer functional sequences (syntax). Such organisation leads to a natural hierarchy that can be used advantageously for learning, since the motifs and the syntax can be acquired independently. Despite mounting experimental evidence for hierarchical structures in neuroscience, models for temporal learning based on neuronal networks have mostly focused on serial methods. Here, we introduce a network model of spiking neurons with a hierarchical organisation aimed at sequence learning on multiple time scales. Using biophysically motivated neuron dynamics and local plasticity rules, the model can learn motifs and syntax independently. Furthermore, the model can relearn sequences efficiently and store multiple sequences. Compared to serial learning, the hierarchical model displays faster learning, more flexible relearning, increased capacity, and higher robustness to perturbations. The hierarchical model redistributes the variability: it achieves high motif fidelity at the cost of higher variability in the between-motif timings.

show abstract

“…Previous work mostly considered randomly connected neural networks [23][24][25][26]. More recent techniques deal with networks mixing randomness and structure focusing on motor cortex, linear networks [27] focusing on continuum movement generation and role of thalamus and basal ganglia, and nonlinear networks [28] focusing on optical control and role of thalamus modulated by basal ganglia. Work on low rank structure [29][30][31][32] exists but does not emphasize how it is related to the thalamus.…”

Section: Introductionmentioning

confidence: 99%

Stable thalamocortical learning between medial-dorsal thalamus and cortical attractor networks captures cognitive flexibility

Qiu

2021

Preprint

View full text Add to dashboard Cite

Primates and rodents are able to continually acquire, adapt, and transfer knowledge and skill, and lead to goal-directed behavior during their lifespan. For the case when context switches slowly, animals learn via slow processes. For the case when context switches rapidly, animals learn via fast processes. We build a biologically realistic model with modules similar to a distributed computing system. Specifically, we are emphasizing the role of thalamocortical learning on a slow time scale between the prefrontal cortex (PFC) and medial dorsal thalamus (MD). Previous work [1] has already shown experimental evidence supporting classification of cell ensembles in the medial dorsal thalamus, where each class encodes a different context. However, the mechanism by which such classification is learned is not clear. In this work, we show that such learning can be self-organizing in the manner of an automaton (a distributed computing system), via a combination of Hebbian learning and homeostatic synaptic scaling. We show that in the simple case of two contexts, the network with hierarchical structure can do context-based decision making and smooth switching between different contexts. Our learning rule creates synaptic competition [2] between the thalamic cells to create winner-take-all activity. Our theory shows that the capacity of such a learning process depends on the total number of task-related hidden variables, and such a capacity is limited by system size N. We also theoretically derived the effective functional connectivity as a function of an order parameter dependent on the thalamo-cortical coupling structure.Significance StatementAnimals need to adapt to dynamically changing environments and make decisions based on changing contexts. Here we propose a combination of neural circuit structure with learning mechanisms to account for such behaviors. Specifically, we built a reservoir computing network improved by a Hebbian learning rule together with a synaptic scaling learning mechanism between the prefrontal cortex and the medial-dorsal (MD) thalamus. This model shows that MD thalamus is crucial in such context-based decision making. I also make use of dynamical mean field theory to predict the effective neural circuit. Furthermore, theoretical analysis provides a prediction that the capacity of such a network increases with the network size and the total number of tasks-related latent variables.

show abstract

A model of flexible motor sequencing through thalamic control of cortical dynamics

Cited by 16 publications

References 44 publications

Optimal anticipatory control as a theory of motor preparation: a thalamo-cortical circuit model

Optimal anticipatory control as a theory of motor preparation: a thalamo-cortical circuit model

Learning compositional sequences with multiple time scales through a hierarchical network of spiking neurons

Stable thalamocortical learning between medial-dorsal thalamus and cortical attractor networks captures cognitive flexibility

Contact Info

Product

Resources

About