Motor cortical influence relies on task-specific activity covariation

Warriner, Claire L.; Fageiry, Samaher; Saxena, Shreya; Costa, Rui M.; Miri, Andrew

doi:10.1016/j.celrep.2022.111427

Cited by 15 publications

(17 citation statements)

References 92 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…MINT is not capable of generalizing to trajectories that traverse regions of state space far from the mesh. This is relevant, because empirical evidence indicates that 'new' behaviors can sometimes be associated with neural trajectories that are very different from all previously observed trajectories -often occupying orthogonal subspaces [63,[91][92][93]. MINT could be modified to allow extrapolation outside the library, but we suspect that this form of generalization will still not typically be available.…”

Section: Generalizationmentioning

confidence: 99%

An emerging view of neural geometry in motor cortex supports high-performance decoding

Perkins

Cunningham

Wang

et al. 2023

Preprint

View full text Add to dashboard Cite

Decoders for brain-computer interfaces (BCIs) assume constraints on neural activity, chosen to reflect scientific beliefs while yielding tractable computations. We document how low tangling – a typical property of motor-cortex neural trajectories – yields unusual neural geometries. We designed a decoder, MINT, to embrace appropriate statistical constraints for these geometries. MINT takes a trajectory-centric approach: a library of neural trajectories (rather than a set of neural dimensions) provides a scaffold approximating the neural manifold. Each neural trajectory has a corresponding behavioral trajectory, allowing straightforward but highly nonlinear decoding. MINT outperformed other interpretable methods, and outperformed expressive machine learning methods in 37 of 42 comparisons. Yet unlike such methods MINT's constraints are known, not the implicit result of optimizing decoder output. MINT performed well across tasks, suggesting its assumptions are generally well-matched to the statistics of neural data. Despite embracing highly nonlinear relationships between behavior and potentially complex neural trajectories, MINT's computations are simple, scalable, and provide interpretable quantities such as data likelihoods. MINT's performance and simplicity suggest it may be an excellent candidate for clinical BCI applications.

show abstract

Section: Generalizationmentioning

confidence: 99%

An emerging view of neural geometry in motor cortex supports high-performance decoding

Perkins

Cunningham

Wang

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…We utilized an M1-dependent head-fixed motor task 32 to screen for cell type engagement at different stages of training. The task consists of forelimb-driven pulls of a small rotary wheel wherein mice had to cross an infrared (IR) beam at the apex of the wheel and make contact with the rungs to commence a trial (Figure 1A).…”

Section: M1 Ct Neurons Are Enriched At Late Stages Of Trainingmentioning

confidence: 99%

Corticothalamic Neurons in Motor Cortex Have a Permissive Role in Motor Execution

Carmona

Tun

et al. 2022

Preprint

View full text Add to dashboard Cite

The primary motor cortex (M1) is a central hub for motor learning and execution, as it receives input from and broadcasts to a multitude of other motor regions. It has become increasingly clear that M1 is heterogeneous in terms of cell types, many of which exhibit varying relationships to movement during learning and execution. Previous studies have mainly used a single candidate approach to identify cell types contributing to motor learning and execution. Here, we employed an unbiased screen to tag active neurons at different stages of performance of a motor task. We characterized the relative cell type composition among the active neurons at different stages of motor training and identified varying patterns across training. One cell type consistently showed enrichment as training and execution progressed: corticothalamic neurons (M1CT). Anatomical characterization of M1CT neurons revealed an extensive projection pattern to thalamic nuclei with marked topographical organization. Using two-photon calcium imaging, we found that M1CT neuron activity is largely suppressed during movement. M1CT neurons display a negative correlation with movement that scales with movement vigor and augments with training. Closed-loop optogenetic manipulation of M1CT activity revealed that their activity rapidly affects movement execution, even after movement is initiated. Taken together, these results identify and reveal a critical permissive role of corticothalamic neurons in movement execution.

show abstract

“…Mice were chronically implanted with EMG electrodes during the surgery in which headplates were attached as described previously ( 45, 61 ). Insertions targeted the biceps (elbow flexor), triceps (elbow extensor), extensor carpi radialis (wrist extensor) and palmaris longus (wrist flexor).…”

Section: Methodsmentioning

confidence: 99%

Hierarchy between forelimb premotor and primary motor cortices and its manifestation in their firing patterns

Saiki-Ishikawa,

Agrios,

Savya

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

Though hierarchy is commonly invoked in descriptions of neural system function, its presence and manifestation in firing patterns remain poorly resolved. Here we use rapid activity perturbations to demonstrate asymmetric influence between premotor and primary motor cortices on fast timescales during reaching in mice, confirming hierarchy among endogenous firing patterns. However, this hierarchy had a surprising manifestation: firing in each region was dominated by temporally-aligned patterns and was equally predictive of future firing in the other region. Network modeling suggests an explanation: functional hierarchy can be determined by the relative dependence of local circuits on across-region input, rather than asymmetry in that input. We propose that motor cortical firing patterns are primarily defined not by across-region input, but by the constraints of patterning motor output.

show abstract

Motor cortical influence relies on task-specific activity covariation

Cited by 15 publications

References 92 publications

An emerging view of neural geometry in motor cortex supports high-performance decoding

An emerging view of neural geometry in motor cortex supports high-performance decoding

Corticothalamic Neurons in Motor Cortex Have a Permissive Role in Motor Execution

Hierarchy between forelimb premotor and primary motor cortices and its manifestation in their firing patterns

Contact Info

Product

Resources

About