Cerebellar Internal Models Research Articles

The Evolution of the Optimization of Cognitive and Social Functions in the Cerebellum and Thereby the Rise of Homo sapiens Through Cumulative Culture.

The evolution of the prominent role of the cerebellum in the development of composite tools, and cumulative culture, leading to the rise of Homo sapiens is examined. Following Stout and Hecht's (2017) detailed description of stone-tool making, eight key repetitive involvements of the cerebellum are highlighted. These key cerebellar learning involvements include the following: (1) optimization of cognitive-social control, (2) prediction (3) focus of attention, (4) automaticity of smoothness, appropriateness, and speed of movement and cognition, (5) refined movement and social cognition, (6) learns models of extended practice, (7) learns models of Theory of Mind (ToM) of teachers, (8) is predominant in acquisition of novel behavior and cognition that accrues from the blending of cerebellar models sent to conscious working memory in the cerebral cortex. Within this context, the evolution of generalization and blending of cerebellar internal models toward optimization of social-cognitive learning is described. It is concluded that (1) repetition of movement and social cognition involving the optimization of internal models in the cerebellum during stone-tool making was the key selection factor toward social-cognitive and technological advancement, (2) observational learning during stone-tool making was the basis for both technological and social-cognitive evolution and, through an optimizing positive feedback loop between the cerebellum and cerebral cortex, the development of cumulative culture occurred, and (3) the generalization and blending of cerebellar internal models related to the unconscious forward control of the optimization of imagined future states in working memory was the most important brain adaptation leading to intertwined advances in stone-tool technology, cognitive-social processes behind cumulative culture (including the emergence of language and art) and, thereby, with the rise of Homo sapiens.

Conserved Population Dynamics in the Cerebro-Cerebellar System between Waking and Sleep.

Despite the importance of the cerebellum for motor learning, and the recognized role of sleep in motor memory consolidation, surprisingly little is known about neural activity in the sleeping cerebro-cerebellar system. Here, we used wireless recording from primary motor cortex (M1) and the cerebellum in three female monkeys to examine the relationship between patterns of single-unit spiking activity observed during waking behavior and in natural sleep. Across the population of recorded units, we observed similarities in the timing of firing relative to local field potential features associated with both movements during waking and up state during sleep. We also observed a consistent pattern of asymmetry in pairwise cross-correlograms, indicative of preserved sequential firing in both wake and sleep at low frequencies. Despite the overall similarity in population dynamics between wake and sleep, there was a global change in the timing of cerebellar activity relative to motor cortex, from contemporaneous in the awake state to motor cortex preceding the cerebellum in sleep. We speculate that similar population dynamics in waking and sleep may imply that cerebellar internal models are activated in both states, despite the absence of movement when asleep. Moreover, spindle frequency coherence between the cerebellum and motor cortex may provide a mechanism for cerebellar computations to influence sleep-dependent learning processes in the motor cortex.SIGNIFICANCE STATEMENT It is well known that sleep can lead to improved motor performance. One possibility is that off-line learning results from neural activity during sleep in brain areas responsible for the control of movement. In this study we show for the first time that neuronal patterns in the cerebro-cerebellar system are conserved during both movements and sleep up-states, albeit with a shift in the relative timing between areas. Additionally, we show the presence of simultaneous M1-cerebellar spike coherence at spindle frequencies associated with up-state replay and postulate that this is a mechanism whereby a cerebellar internal model can shape plasticity in neocortical circuits during sleep.

The posterior cerebellum and inconsistent trait implications when learning the sequence of actions.

It has been proposed that the cerebellum contributes to social cognition. Based on the view that cerebellar internal models create predictions on motions and actions, we hypothesize that the posterior cerebellum supports identifying temporal sequences of persons’ actions as well as detecting inconsistent actions that violate the implied trait. Participants were required to memorize the temporal order of a set of sentences that implied a personality trait. Importantly, the sentence sets were designed in such a way that the first half of each set involved actions that were consistent with the same trait, while the other half was either consistent or inconsistent with that trait. As expected, we found robust posterior cerebellar activation when memorizing the order of the actions, irrespective of trait consistency, but more crucially also for actions implying an inconsistent trait in comparison to consistent trait actions. We also found that the medial prefrontal cortex and posterior cerebellum were associated with confidence level in retrieving the sequences. This study supports the hypothesis that the posterior cerebellum identifies and predicts the low-level temporal order of actions and demonstrates for the first time that this area is also involved in the high-level prediction of trait implications of those actions.

Cerebellar Contributions to Proactive and Reactive Control in the Stop Signal Task: A Systematic Review and Meta-Analysis of Functional Magnetic Resonance Imaging Studies.

The cerebellum facilitates and modulates cognitive functions using forward and inverse internal models to predict and control behavior, respectively. Despite neuroimaging evidence that regions of the cerebellum are active during executive function (EF) tasks in general, little is known about the cerebellum's role in specific EFs and their underlying neural networks. Inhibitory control specifically may be facilitated by cerebellar internal models predicting responses during proactive control (withholding), and controlling responses during reactive control (inhibiting). The stop signal task (SST) is an inhibitory control task often used in neuroimaging studies to measure neural responses to both proactive and reactive control. Thus, in this review, we examine evidence for the cerebellum's role in inhibitory control by reviewing studies of healthy adults that utilized the SST in event-related functional magnetic resonance imaging (fMRI) experiments. Twenty-one studies that demonstrated cerebellar results were eligible for review, including 749 participants, 28 contrasts, and 38 cerebellar clusters. We also performed activation likelihood estimation (ALE) meta-analysis of contrasts derived from reviewed studies. This review illustrates evidence for the cerebellum participating in inhibitory control independent of motor control. Most significant cerebellar clusters were located in the left posterior cerebellum, suggesting that it communicates with the established cortical right-lateralized inhibitory control network. Cerebellar activity was most consistently observed for contrasts that measured proactive control, and ALE analysis confirmed that left Crus I is most likely to be activated in studies of proactive control measuring monitoring and anticipation. Results suggest that the left posterior cerebellum may communicate with right frontal and parietal cortices, using forward models to predict appropriate responses. Reactive control contrasts indicated a possible role for cerebellar regions in enhancing inhibition efficiency through inverse models, but ALE meta-analysis did not confirm this hypothesis. Limitations in the current literature, clinical implications, and directions for future research are discussed.

The prominent role of the cerebellum in the social learning of the phonological loop in working memory: How language was adaptively built from cerebellar inner speech required during stone-tool making.

Based on advances in cerebellum research as to its cognitive, social, and language contributions to working memory, the purpose of this article is to describe new support for the prominent involvement of cerebellar internal models in the adaptive selection of language. Within this context it has been proposed that (1) cerebellar internal models of inner speech during stone-tool making accelerated the adaptive evolution of new cause-and-effect sequences of precision stone-tool knapping requirements, and (2) that these evolving cerebellar internal models coded (i.e., learned in corticonuclear microcomplexes) such cause-and-effect sequences as phonological counterparts and, these, when sent to the cerebral cortex, became new phonological working memory. This article describes newer supportive research findings on (1) the cerebellum's role in silent speech in working memory, and (2) recent findings on genetic aspects (FOXP2) of the role of silent speech in language evolution. It is concluded that within overall cerebro-cerebellar evolution, without the evolution of cerebellar coding of stone-tool making sequences of primitive working memory (beginning approximately 1.7 million years ago) language would not have evolved in the subsequent evolution of Homo sapiens.

Functional Connectivity between the Cerebellum and Somatosensory Areas Implements the Attenuation of Self-Generated Touch.

Since the early 1970s, numerous behavioral studies have shown that self-generated touch feels less intense and less ticklish than the same touch applied externally. Computational motor control theories have suggested that cerebellar internal models predict the somatosensory consequences of our movements and that these predictions attenuate the perception of the actual touch. Despite this influential theoretical framework, little is known about the neural basis of this predictive attenuation. This is due to the limited number of neuroimaging studies, the presence of conflicting results about the role and the location of cerebellar activity, and the lack of behavioral measures accompanying the neural findings. Here, we combined psychophysics with fMRI to detect the neural processes underlying somatosensory attenuation in male and female healthy human participants. Activity in bilateral secondary somatosensory areas was attenuated when the touch was presented during a self-generated movement (self-generated touch) than in the absence of movement (external touch). An additional attenuation effect was observed in the cerebellum that is ipsilateral to the passive limb receiving the touch. Importantly, we further found that the degree of functional connectivity between the ipsilateral cerebellum and the contralateral primary and bilateral secondary somatosensory areas was linearly and positively related to the degree of behaviorally assessed attenuation; that is, the more participants perceptually attenuated their self-generated touches, the stronger this corticocerebellar coupling. Collectively, these results suggest that the ipsilateral cerebellum is fundamental in predicting self-generated touch and that this structure implements somatosensory attenuation via its functional connectivity with somatosensory areas.SIGNIFICANCE STATEMENT When we touch our hand with the other, the resulting sensation feels less intense than when another person or a machine touches our hand with the same intensity. Early computational motor control theories have proposed that the cerebellum predicts and cancels the sensory consequences of our movements; however, the neural correlates of this cancelation remain unknown. By means of fMRI, we show that the more participants attenuate the perception of their self-generated touch, the stronger the functional connectivity between the cerebellum and the somatosensory cortical areas. This provides conclusive evidence about the role of the cerebellum in predicting the sensory feedback of our movements and in attenuating the associated percepts via its connections to early somatosensory areas.

A Cerebellar Internal Models Control Architecture for Online Sensorimotor Adaptation of a Humanoid Robot Acting in a Dynamic Environment

Humanoid robots are often supposed to operate in non-deterministic human environments, and as a consequence, the robust and gentle rejection of the external perturbations is extremely crucial. In this scenario, stable and accurate behavior is mostly solved through adaptive control mechanisms that learn an internal model to predict the consequences of the outgoing control signals. Evidences show that brain-based biological systems resolve this control issue by updating an appropriate internal model that is then used to direct the muscles activities. Inspired by the biological cerebellar internal models theory, that couples forward and inverse internal models into the biological motor control scheme, we propose a novel methodology to artificially replicate these learning and adaptive principles into a robotic feedback controller. The proposed cerebellar-like network combines machine learning, artificial neural network, and computational neuroscience techniques to deal with all the nonlinearities and complexities that modern robotic systems could present. Although the architecture is tested on the simulated humanoid iCub, it can be applied to different robotic systems without excessive customization, thanks to its neural network-based nature. During the experiments, the robot is requested to follow repeatedly a movement while it is interacting with two external systems. Four different internal model architectures are compared and tested under different conditions. The comparison of the performances confirmed the theories about internal models combinatory action. The combination of models together with the structural and learning features of the network, resulted in a benefit to the adaptation mechanism, but also the system response to nonlinearities, noise and external forces.

The evolution of theory of mind (ToM) within the evolution of cerebellar sequence detection in stone-tool making and language: implications for studies of higher-level cognitive functions in degenerative cerebellar atrophy

IntroductionWithin the context of Clausi, Olivito, Lupo, Siciliano, Bozzali and Leggio’s (Cell Neurosci 12:510, 2019) insightful study of how prediction of theory of mind (ToM) is compromised in degenerative cerebellar atrophy, this article describes how prediction can also be understood as the cerebro-cerebellar system’s capacity to rapidly shift attention to manipulate cause-and-effect relationships embedded in language.MethodThe evolution of the capacity of ToM is described within the evolution of stone-tool making, language, and the origin of the phonological loop in verbal working memory. Specifically, it is argued that this evolutionary framework offers a way to get further inside the prediction process by illuminating how sub-vocal speech evolved during stone-tool evolution due to its adaptive refinement of early human ability to manipulate and hold in memory progressively more detailed cause-and-effect relationships in the origin of verbal working memory.ConclusionThe addition of sub-vocal speech/cause-and-effect relationship to the analysis of prediction provides an evolutionary model of the mechanisms of ToM, which, in turn, brings forward additional cerebro-cerebellar mechanisms which can (1) further support Clausi, Olivito, Lupo et al’s findings and (2) shed light on additional mechanisms that might further clarify what might be behind cerebellar dysfunction in the construction of ToM. Problems encountered by cerebellar degenerative atrophy patients with the Faux pas test and Advanced ToM task with unexpected events may stem from a combination of an inability (1) of their cerebellar internal models to rapidly switch attention among cause-and-effect elements of the stories and (2) to extend cerebellar internal models to the prediction of the resulting similar but unexpected events. That is, with both (1) and (2) occurring at the same time, alternative meanings of causes and effects might be missed in both automatic and consciously manipulated sub-vocal verbal working memory. A method to measure sub-vocal speech in this context is suggested.

The role of the cerebellum for feedback processing and behavioral switching in a reversal-learning task

Previous studies have reported cerebellar activations during error and reward processing. The present study investigated if the cerebellum differentially processes feedback depending on changes in response strategy during reversal learning, as is conceivable given its internal models for movement and thought. Negative relative to positive feedback in an fMRI-based reversal learning task was hypothesized to be associated with increased cerebellar activations. Moreover, increased activations were expected for negative feedback followed by a change in response strategy compared to negative feedback not followed by such a change, and for first positive feedback after compared to final negative feedback before a change, due to updating of internal models. As predicted, activation in lobules VI and VIIa/Crus I was increased for negative relative to positive feedback, and for final negative feedback before a change in response strategy relative to negative feedback not associated with a change. Moreover, activation was increased for first positive feedback after relative to final negative feedback before a change. These findings are consistent with updating of cerebellar internal models to accommodate new behavioral strategies. Recruitment of posterior regions in reversal learning is in line with the cerebellar functional topography, with posterior regions involved in complex motor and cognitive functions.

A comprehensive gaze stabilization controller based on cerebellar internal models

Gaze stabilization is essential for clear vision; it is the combined effect of two reflexes relying on vestibular inputs: the vestibulocollic reflex (VCR), which stabilizes the head in space and the vestibulo-ocular reflex (VOR), which stabilizes the visual axis to minimize retinal image motion. The VOR works in conjunction with the opto-kinetic reflex (OKR), which is a visual feedback mechanism that allows the eye to move at the same speed as the observed scene. Together they keep the image stationary on the retina.In this work, we implement on a humanoid robot a model of gaze stabilization based on the coordination of VCR, VOR and OKR. The model, inspired by neuroscientific cerebellar theories, is provided with learning and adaptation capabilities based on internal models. We present the results for the gaze stabilization model on three sets of experiments conducted on the SABIAN robot and on the iCub simulator, validating the robustness of the proposed control method. The first set of experiments focused on the controller response to a set of disturbance frequencies along the vertical plane. The second shows the performances of the system under three-dimensional disturbances. The last set of experiments was carried out to test the capability of the proposed model to stabilize the gaze in locomotion tasks. The results confirm that the proposed model is beneficial in all cases reducing the retinal slip (velocity of the image on the retina) and keeping the orientation of the head stable.

The Origin of Mathematics and Number Sense in the Cerebellum: with Implications for Finger Counting and Dyscalculia

BackgroundMathematicians and scientists have struggled to adequately describe the ultimate foundations of mathematics. Nobel laureates Albert Einstein and Eugene Wigner were perplexed by this issue, with Wigner concluding that the workability of mathematics in the real world is a mystery we cannot explain. In response to this classic enigma, the major purpose of this article is to provide a theoretical model of the ultimate origin of mathematics and “number sense” (as defined by S. Dehaene) that is proposed to involve the learning of inverse dynamics models through the collaboration of the cerebellum and the cerebral cortex (but prominently cerebellum-driven). This model is based upon (1) the modern definition of mathematics as the “science of patterns,” (2) cerebellar sequence (pattern) detection, and (3) findings that the manipulation of numbers is automated in the cerebellum. This cerebro-cerebellar approach does not necessarily conflict with mathematics or number sense models that focus on brain functions associated with especially the intraparietal sulcus region of the cerebral cortex. A direct corollary purpose of this article is to offer a cerebellar inner speech explanation for difficulty in developing “number sense” in developmental dyscalculia.ResultsIt is argued that during infancy the cerebellum learns (1) a first tier of internal models for a primitive physics that constitutes the foundations of visual-spatial working memory, and (2) a second (and more abstract) tier of internal models based on (1) that learns “number” and relationships among dimensions across the primitive physics of the first tier. Within this context it is further argued that difficulty in the early development of the second tier of abstraction (and “number sense”) is based on the more demanding attentional requirements imposed on cerebellar inner speech executive control during the learning of cerebellar inverse dynamics models. Finally, it is argued that finger counting improves (does not originate) “number sense” by extending focus of attention in executive control of silent cerebellar inner speech.DiscussionIt is suggested that (1) the origin of mathematics has historically been an enigma only because it is learned below the level of conscious awareness in cerebellar internal models, (2) understandings of the development of “number sense” and developmental dyscalculia can be advanced by first understanding the ultimate foundations of number and mathematics do not simply originate in the cerebral cortex, but rather in cerebro-cerebellar collaboration (predominately driven by the cerebellum).ConclusionIt is concluded that difficulty with “number sense” results from the extended demands on executive control in learning inverse dynamics models associated with cerebellar inner speech related to the second tier of abstraction (numbers) of the infant’s primitive physics.

Cerebellar tDCS Modulates Neural Circuits during Semantic Prediction: A Combined tDCS-fMRI Study.

It has been proposed that the cerebellum acquires internal models of mental processes that enable prediction, allowing for the optimization of behavior. In language, semantic prediction speeds speech production and comprehension. Right cerebellar lobules VI and VII (including Crus I/II) are engaged during a variety of language processes and are functionally connected with cerebral cortical language networks. Further, right posterolateral cerebellar neuromodulation modifies behavior during predictive language processing. These data are consistent with a role for the cerebellum in semantic processing and semantic prediction. We combined transcranial direct current stimulation (tDCS) and fMRI to assess the behavioral and neural consequences of cerebellar tDCS during a sentence completion task. Task-based and resting-state fMRI data were acquired in healthy human adults (n = 32; μ = 23.1 years) both before and after 20 min of 1.5 mA anodal (n = 18) or sham (n = 14) tDCS applied to the right posterolateral cerebellum. In the sentence completion task, the first four words of the sentence modulated the predictability of the final target word. In some sentences, the preceding context strongly predicted the target word, whereas other sentences were nonpredictive. Completion of predictive sentences increased activation in right Crus I/II of the cerebellum. Relative to sham tDCS, anodal tDCS increased activation in right Crus I/II during semantic prediction and enhanced resting-state functional connectivity between hubs of the reading/language networks. These results are consistent with a role for the right posterolateral cerebellum beyond motor aspects of language, and suggest that cerebellar internal models of linguistic stimuli support semantic prediction.SIGNIFICANCE STATEMENT Cerebellar involvement in language tasks and language networks is now well established, yet the specific cerebellar contribution to language processing remains unclear. It is thought that the cerebellum acquires internal models of mental processes that enable prediction, allowing for the optimization of behavior. Here we combined neuroimaging and neuromodulation to provide evidence that the cerebellum is specifically involved in semantic prediction during sentence processing. We found that activation within right Crus I/II was enhanced when semantic predictions were made, and we show that modulation of this region with transcranial direct current stimulation alters both activation patterns and functional connectivity within whole-brain language networks. For the first time, these data show that cerebellar neuromodulation impacts activation patterns specifically during predictive language processing.

The prominent role of the cerebellum in the learning, origin and advancement of culture.

BackgroundVandervert described how, in collaboration with the cerebral cortex, unconscious learning of cerebellar internal models leads to enhanced executive control in working memory in expert music performance and in scientific discovery. Following Vandervert’s arguments, it is proposed that since music performance and scientific discovery, two pillars of cultural learning and advancement, are learned through in cerebellar internal models, it is reasonable that additional if not all components of culture may be learned in the same way. Within this perspective strong evidence is presented that argues that the learning, maintenance, and advancement of culture are accomplished primarily by recently-evolved (the last million or so years) motor/cognitive functions of the cerebellum and not primarily by the cerebral cortex as previously assumed. It is suggested that the unconscious cerebellar mechanism behind the origin and learning of culture greatly expands Ito’s conception of the cerebellum as “a brain for an implicit self.”ResultsThrough the mechanism of predictive sequence detection in cerebellar internal models related to the body, other persons, or the environment, it is shown how individuals can unconsciously learn the elements of culture and yet, at the same time, be in social sync with other members of culture. Further, this predictive, cerebellar mechanism of socialization toward the norms of culture is hypothesized to be diminished among children who experience excessive television viewing, which results in lower grades, poor socialization, and diminished executive control.ConclusionIt is concluded that the essential components of culture are learned and sustained not by the cerebral cortex alone as many traditionally believe, but are learned through repetitious improvements in prediction and control by internal models in the cerebellum. From this perspective, the following new explanations of culture are discussed: (1) how culture can be learned unconsciously but yet be socially in sync with others, (2) how the recent evolutionary expansion of the cerebellum was involved in the co-evolution of earliest stone tools and language—leading to the cerebellum-driven origin of culture, (3) how cerebellar internal models are blended to produce the creative, forward advances in culture, (4) how the blending of cerebellar internal models led to human, multi-component, infinitely partitionable and communicable working memory, (5) how excessive television viewing may represent a cultural shift that diminishes the observational learning of internal models of the behavior of others and thus may result in a mild, parallel version of Schmahmann’s cerebellar cognitive affective syndrome.

Generalized Role for the Cerebellum in Encoding Internal Models: Evidence from Semantic Processing

The striking homogeneity of cerebellar microanatomy is strongly suggestive of a corresponding uniformity of function. Consequently, theoretical models of the cerebellum's role in motor control should offer important clues regarding cerebellar contributions to cognition. One such influential theory holds that the cerebellum encodes internal models, neural representations of the context-specific dynamic properties of an object, to facilitate predictive control when manipulating the object. The present study examined whether this theoretical construct can shed light on the contribution of the cerebellum to language processing. We reasoned that the cerebellum might perform a similar coordinative function when the context provided by the initial part of a sentence can be highly predictive of the end of the sentence. Using functional MRI in humans we tested two predictions derived from this hypothesis, building on previous neuroimaging studies of internal models in motor control. First, focal cerebellar activation-reflecting the operation of acquired internal models-should be enhanced when the linguistic context leads terminal words to be predictable. Second, more widespread activation should be observed when such predictions are violated, reflecting the processing of error signals that can be used to update internal models. Both predictions were confirmed, with predictability and prediction violations associated with increased blood oxygenation level-dependent signal in the posterior cerebellum (Crus I/II). Our results provide further evidence for cerebellar involvement in predictive language processing and suggest that the notion of cerebellar internal models may be extended to the language domain.

Coupling internal cerebellar models enhances online adaptation and supports offline consolidation in sensorimotor tasks

The cerebellum is thought to mediate sensorimotor adaptation through the acquisition of internal models of the body-environment interaction. These representations can be of two types, identified as forward and inverse models. The first predicts the sensory consequences of actions, while the second provides the correct commands to achieve desired state transitions. In this paper, we propose a composite architecture consisting of multiple cerebellar internal models to account for the adaptation performance of humans during sensorimotor learning. The proposed model takes inspiration from the cerebellar microcomplex circuit, and employs spiking neurons to process information. We investigate the intrinsic properties of the cerebellar circuitry subserving efficient adaptation properties, and we assess the complementary contributions of internal representations by simulating our model in a procedural adaptation task. Our simulation results suggest that the coupling of internal models enhances learning performance significantly (compared with independent forward and inverse models), and it allows for the reproduction of human adaptation capabilities. Furthermore, we provide a computational explanation for the performance improvement observed after one night of sleep in a wide range of sensorimotor tasks. We predict that internal model coupling is a necessary condition for the offline consolidation of procedural memories.

Cerebellar Internal Models: Implications for the Dexterous Use of Tools

Humans have remarkable abilities in the dexterous use of tools to extend their physical capabilities. Although previous neuropsychological and functional neuroimaging studies have mainly focused on the contribution of frontal-parietal cerebral networks to skills for tool use, dense anatomical and functional connections are known to exist between the frontal-parietal regions and the lateral cerebellum, suggesting that the cerebellum also supports the information processing necessary for the dexterous use of tools. In this article, we review functional and structural imaging studies reporting that the cerebellum is related to the learning acquisition of neural mechanisms representing input-output properties of controlled objects, including tools. These studies also suggest that such mechanisms are modularly organized in the cerebellum corresponding to the different properties of objects, such as kinematic or dynamic properties and types of tools, and that they enable humans to flexibly cope with discrete changes in objects and environments by reducing interference and combining acquired modules. Based on these studies, we propose a hypothesis that the cerebellum contributes to the skillful use of tools by representing the input-output properties of tools and providing information on the prediction of the sensory consequences of manipulation with the parietal regions, which are related to multisensory processing, and information on the necessary control of tools with the premotor regions, which contribute to the control of hand movements.

From ‘Understanding the Brain by Creating the Brain’ towards manipulative neuroscience

Ten years have passed since the Japanese 'Century of the Brain' was promoted, and its most notable objective, the unique 'creating the brain' approach, has led us to apply a humanoid robot as a neuroscience tool. Here, we aim to understand the brain to the extent that we can make humanoid robots solve tasks typically solved by the human brain by essentially the same principles. I postulate that this 'Understanding the Brain by Creating the Brain' approach is the only way to fully understand neural mechanisms in a rigorous sense. Several humanoid robots and their demonstrations are introduced. A theory of cerebellar internal models and a systems biology model of cerebellar synaptic plasticity is discussed. Both models are experimentally supported, but the latter is more easily verifiable while the former is still controversial. I argue that the major reason for this difference is that essential information can be experimentally manipulated in molecular and cellular neuroscience while it cannot be manipulated at the system level. I propose a new experimental paradigm, manipulative neuroscience, to overcome this difficulty and allow us to prove cause-and-effect relationships even at the system level.

Towards manipulative neuroscience based on brain-network -interface

Event Abstract Back to Event Towards manipulative neuroscience based on brain-network -interface Mitsuo Kawato1* 1 ATR Computational Neuroscience Labs, Japan The cerebellar internal model theory postulates that the cerebellar cortex acquires many internal models of controlled objects, dynamical processes in the external world dependent on long-term depression (LTD) of Purkinje cells. It predicts that the climbing fiber inputs to Purkinje cells carry the feedback motor command and can supervise learning of inverse dynamics models. Many experimental supports were obtained from the ventral paraflocculus of the cerebellum during monkey control of ocular following responses. fMRI studies mapped forward and inverse models of manipulated objects and tools in the cerebellar cortex. Kinetic models of LTD [1,2] suggest a cascade of excitable and bistable dynamical processes, which may resolve plasticity-stability dilemma at single spine level. That is, even a single pulse of climbing fiber input combined with an early train of several parallel fiber pulses can induce Ca2+ induced Ca2+ release via IP3 receptors on ER. The MAPK positive feedback loop leaky integrates resulting large Ca2+ elevation and if it crosses the threshold then the state moves to the depressed equilibrium. These models explain diverse LTD experiments and clearly demonstrate that LTD is a supervised learning rule, and not anti-Hebbian as erroneously characterized. The MAPK positive feedback loop model [1] was recently supported by a Ca2+ photo-uncaging and imaging experiment [3] that supports LTD all-or-none character. In [3], the most important information within the system, Ca2+ can be measured and manipulated directly, thus the theory was quite rigorously proved. At the system level, the cerebellar internal models for arm movements are still debated and we need to develop a method to directly manipulate information. In our "Understanding brain by Creating Brain"-approach, high-performance humanoid robots have been developed, and sensory-motor coordination problems were investigated. We explored several computational theories such as MOSAIC, imitation learning, biologically motivated robot biped locomotion, modular and hierarchical reinforcement learning models. Our paradigms include computational-model based imaging and non-invasive decoding of neural representations. However, we got frustrated that most experiments can show just temporal correlation between data and theory but not causality, unlike the above spine-level LTD story. We hope that manipulative neuroscience based on real time feedback of decoded neural information could be a resolution to this difficulty. Hierarchical Bayesian approach to combine fMRI and MEG, or NIRS and EEG, or neural decoding with machine learning techniques, and real-time control of robots with decoded information are expected to be key technological elements. Demonstrations of real-time robot hand control by decoded information from fMRI (http://www.atr.jp/html/topics/press_060526_e.html) in collaboration with Honda, control of a humanoid robot locomotion-like movements from monkey brain in collaboration with Miguel Nicolelis [4], prediction of wrist joint angular acceleration from the primary motor cortex currents estimated by hierarchical Bayesian MEG, or estimation of visual target velocity from MST currents, etc could be technical bases for this new approach.

Towards manipulative neuroscience based on Brain Network Interface

In ATR Computational Neuroscience Laboratories, we proposed several computational models such as cerebellar internal models, MOSAIC, and modular and hierarchical reinforcement-learning models. Some of these models can quantitatively reproduce subject behaviors given sensory inputs and reward and action sequences that subjects received and generated. These computational models possess putative information representation such as error signals for internal models and action stimulus dependent reward prediction, and they can be used as explanatory variables in neuroimaging and neurophysiology experiments. We named this approach computational-model-based neuroimaging, as well as computational-model-based neurophysiology. This new approach is very appealing since it is likely the only method with which we can explore neural representations remotely from either sensory or motor interfaces. However, sometimes the limitation of a mere temporal correlation between the theory and data became apparent, so we started to develop a new paradigm, “manipulative neuroscience”, where physical causality is guaranteed.