Expansion coding and computation in the cerebellum: 50 years after the Marr–Albus codon theory

Sanger, Terence D.; Yamashita, Okito; Kawato, Mitsuo

doi:10.1113/jp278745

Cited by 19 publications

(27 citation statements)

References 88 publications

(234 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The cerebellar circuit is characterized by two specific features (Figure 1B): (1) the mostly feedforward connectivity from MFs as an input to DN as an output; and (2) the expansion from MFs (input to the cerebellum) to granule cells (GCs; input to the cerebellar cortex) and the compression from MFs to DNCs (output from the cerebellum). A possible computational role of the expansion coding in GCs is reviewed in Sanger et al (2020). Given the multitude of computational functions of the internal forward model, it is not surprising that an impairment of the cerebellum leads to a plethora of motor deficiencies collectively known as cerebellar ataxia (Holmes, 1917).…”

Section: Previous Evidence For Cerebellar Forward Model In the Cerebellummentioning

confidence: 99%

“…One may then wonder what the functional advantage of the compact representation is. There appears no consensus on the functional role of the compact representation (Sanger et al, 2020). Given the fact that the cerebellum contributes to fast, trained and automated motor control with reduced effort and attention, the compressed representation may be beneficial or even necessary to extract relevant information from numerous and redundant cerebellar inputs and to assign more attention to the task currently in the focus.…”

Section: Compressed Prediction Of the Cerebellar Internal Modelmentioning

confidence: 99%

See 1 more Smart Citation

The Cerebro-Cerebellum as a Locus of Forward Model: A Review

Tanaka

Ishikawa

Lee

et al. 2020

Front. Syst. Neurosci.

109

113

View full text Add to dashboard Cite

This review surveys physiological, behavioral, and morphological evidence converging to the view of the cerebro-cerebellum as loci of internal forward models. The cerebro-cerebellum, the phylogenetically newest expansion in the cerebellum, receives convergent inputs from cortical, subcortical, and spinal sources, and is thought to perform the predictive computation for both motor control, motor learning, and cognitive functions. This predictive computation is known as an internal forward model. First, we elucidate the theoretical foundations of an internal forward model and its role in motor control and motor learning within the framework of the optimal feedback control model. Then, we discuss a neural mechanism that generates various patterns of outputs from the cerebro-cerebellum. Three lines of supporting evidence for the internal-forwardmodel hypothesis are presented in detail. First, we provide physiological evidence that the cerebellar outputs (activities of dentate nucleus cells) are predictive for the cerebellar inputs [activities of mossy fibers (MFs)]. Second, we provide behavioral evidence that a component of movement kinematics is predictive for target motion in control subjects but lags behind a target motion in patients with cerebellar ataxia. Third, we provide morphological evidence that the cerebellar cortex and the dentate nucleus receive separate MF projections, a prerequisite for optimal estimation. Finally, we speculate that the predictive computation in the cerebro-cerebellum could be deployed to not only motor control but also to non-motor, cognitive functions. This review concludes that the predictive computation of the internal forward model is the unifying algorithmic principle for understanding diverse functions played by the cerebro-cerebellum.

show abstract

Section: Previous Evidence For Cerebellar Forward Model In the Cerebellummentioning

confidence: 99%

Section: Compressed Prediction Of the Cerebellar Internal Modelmentioning

confidence: 99%

The Cerebro-Cerebellum as a Locus of Forward Model: A Review

Tanaka

Ishikawa

Lee

et al. 2020

Front. Syst. Neurosci.

109

113

View full text Add to dashboard Cite

show abstract

“…The Purkinje cells receive excitatory inputs from the axons of granule cells (the parallel fibers) that relay the mossy fibers. In humans, there is a striking expansion of the information, as approximately 250 million mossy fibers contact 50 billion granule cells, followed by compression through 15 million Purkinje cells (Sanger et al, 2020). The Purkinje cells also receive excitatory inputs from the climbing fibers that originate in the inferior olivary.…”

Section: Functional Neuroanatomy Of the Cerebellum Relevant To Forward The Model Theorymentioning

confidence: 99%

The Forward Model: A Unifying Theory for the Role of the Cerebellum in Motor Control and Sense of Agency

Welniarz

Worbe

Gallea

2021

Front. Syst. Neurosci.

View full text Add to dashboard Cite

For more than two decades, there has been converging evidence for an essential role of the cerebellum in non-motor functions. The cerebellum is not only important in learning and sensorimotor processes, some growing evidences show its implication in conditional learning and reward, which allows building our expectations about behavioral outcomes. More recent work has demonstrated that the cerebellum is also required for the sense of agency, a cognitive process that allows recognizing an action as our own, suggesting that the cerebellum might serve as an interface between sensorimotor function and cognition. A unifying model that would explain the role of the cerebellum across these processes has not been fully established. Nonetheless, an important heritage was given by the field of motor control: the forward model theory. This theory stipulates that movements are controlled based on the constant interactions between our organism and its environment through feedforward and feedback loops. Feedforward loops predict what is going to happen, while feedback loops confront the prediction with what happened so that we can react accordingly. From an anatomical point of view, the cerebellum is at an ideal location at the interface between the motor and sensory systems, as it is connected to cerebral, striatal, and spinal entities via parallel loops, so that it can link sensory and motor systems with cognitive processes. Recent findings showing that the cerebellum participates in building the sense of agency as a predictive and comparator system will be reviewed together with past work on motor control within the context of the forward model theory.

show abstract

“…Assuming equal weights, the presynaptic sum of GrC activations is expected to be the same across patterns. To differentiate patterns, it is theorized that PCs can manipulate their synaptic weights to affect (increase or decrease) the postsynaptic linear sum of specific GrC activation 7,8,13,17,18,33,40,46 . For Fig.…”

Section: Numerical Analysesmentioning

confidence: 99%

“…Clustering by the similarity of their inputs generated spatially coherent PC groups, each having higher similarity scores within groups than across groups. These groups of PCs may provide a basis for ensemble learning 21,22,46 .…”

mentioning

confidence: 99%

Structured connectivity in the cerebellum enables noise-resilient pattern separation

Nguyen

Thomas

Rhoades

et al. 2021

Preprint

View full text Add to dashboard Cite

SummaryThe cerebellum is thought to detect and correct errors between intended and executed commands1–3 and is critical for social behaviors, cognition and emotion4–6. Computations for motor control must be performed quickly to correct errors in real time and should be sensitive to small differences between patterns for fine error correction while being resilient to noise7. Influential theories of cerebellar information processing have largely assumed random network connectivity, which increases the encoding capacity of the network’s first layer8–13. However, maximizing encoding capacity reduces resiliency to noise7. To understand how neuronal circuits address this fundamental tradeoff, we mapped the feedforward connectivity in the mouse cerebellar cortex using automated large-scale transmission electron microscopy (EM) and convolutional neural network-based image segmentation. We found that both the input and output layers of the circuit exhibit redundant and selective connectivity motifs, which contrast with prevailing models. Numerical simulations suggest these redundant, non-random connectivity motifs increase discriminability of similar input patterns at a minimal cost to the network’s overall encoding capacity. This work reveals how neuronal network structure can balance encoding capacity and redundancy, unveiling principles of biological network architecture with implications for artificial neural network design.

show abstract

Expansion coding and computation in the cerebellum: 50 years after the Marr–Albus codon theory

Cited by 19 publications

References 88 publications

The Cerebro-Cerebellum as a Locus of Forward Model: A Review

The Cerebro-Cerebellum as a Locus of Forward Model: A Review

The Forward Model: A Unifying Theory for the Role of the Cerebellum in Motor Control and Sense of Agency

Structured connectivity in the cerebellum enables noise-resilient pattern separation

Contact Info

Product

Resources

About