Advanced Robotics

Eberle, Harry; Hayashi, Yoshikatsu; Kurazume, Ryo; Takei, Tomohiko; An, Qi

doi: 10.1080/01691864.2021.1943710pmid: N/A

Hyper-adaptability is an ability of humans and animals to adapt to large-scale changes in the nervous system or the musculoskeletal system, such as strokes and spinal cord injuries. Although this adaptation may involve similar neural processes with normal adaptation to usual environmental and body changes in daily lives, it can be fundamentally different because it requires ‘construction’ of the neural structure itself and ‘reconstitution’ of sensorimotor control rules to compensate for the changes in the nervous system. In this survey paper, we aimed to provide an overview on how the brain structure changes after brain injury and recovers through rehabilitation. Next, we demonstrated the recent approaches used to apply computational and neural network modeling to recapitulate motor control and motor learning processes. Finally, we discussed future directions to bridge the gap between conventional physiological and modeling approaches to understand the neural and computational mechanisms of hyper-adaptability and its applications to clinical rehabilitation.

Kambara, H.; Ogawa, H.; Takagi, A.; Shin, D.; Yoshimura, N.; Koike, Y.

doi: 10.1080/01691864.2021.1900913pmid: N/A

We can catch all sorts of falling objects like eggs or balls. We must predict the dynamical properties of objects to interact with them, but doing so precisely is difficult. The CNS is suggested to modulate the mechanical impedance of the musculoskeletal dynamics to accomplish robust control and overcome variability in environmental dynamics. In this study, we tested the hypothesis that musculoskeletal stiffness increases as the degree of variability in the environment increase through the motor adaptation process. We conducted a ball-catching task experiment in a virtual reality system where the load force of the ball changed every trial and measured the muscle activity in the wrist to estimate its joint stiffness. We found that group level wrist stiffness after adaptation monotonically increased against load force variability. Meanwhile, some participants showed a non-monotonic relationship between wrist stiffness and load force variability. The results of the experiment and computational simulations suggest that the CNS may adapt to a stochastic environment by modulating musculoskeletal stiffness level under the trade-off between movement accuracy and energetic cost.

Nishimura, Kotaro; Saracbasi, Ozge Ozlem; Hayashi, Yoshikatsu; Kondo, Toshiyuki

doi: 10.1080/01691864.2021.1913445pmid: N/A

Skilled musicians can improvise with first-time partners. Thus, the question arises how the adaptability to others can emerge through the mutual motor learning experience. We developed a two-person cooperative visuomotor task; an object was connected through virtual springs with the cursors controlled by the subjects. We instructed paired subjects to jointly control the object toward a specified target under a virtual force field. Experimental results suggest that a novice subject who was trained with a skill-level matched peer in the Learning phase showed significantly better adaptability to others in the successive Evaluation phase. Variety of the cooperative experience with others in the visuomotor task probably gave rise to high adaptability in the novice-to-novice group subjects, while the learning experience with an expert did not. We conclude that the motor skills acquired during mutual interactions with peers can lead to have an ability to tune the motor commands subject to the dynamics of the external environment and the behavior of the partners.

Morita, Tomoyo; Asada, Minoru; Naito, Eiichi

doi: 10.1080/01691864.2021.1886168pmid: N/A

In the central nervous system, regional neuronal inhibition plays important roles in functional segregation. Here, we showed how brain deactivation, which is a putative index of neuronal inhibition, develops and ages using a unimanual motor task. Healthy right-handed children (8–11 years), adolescents (12–15 years), young adults (20–24 years), and older adults (69–75 years; 21 participants in each group) underwent functional magnetic resonance imaging with their eyes closed while they performed 1-Hz alternating extension–flexion of the right wrist. In young adults, we found deactivations in the hand/arm section of the ipsilateral primary sensorimotor cortices (SM1) including the dorsal premotor cortex (interhemispheric inhibition), foot and face SM1 sections (cross-somatotopic inhibition), visual and auditory cortices (cross-modal inhibition), and precuneus and medial prefrontal cortex of the default mode network (DMN; DMN inhibition). Interhemispheric, cross-modal, and DMN inhibitions developed from childhood to adulthood, but cross-somatotopic inhibition showed no developmental changes. Conversely, interhemispheric, cross-somatotopic, and cross-modal inhibitions, but not DMN inhibition, decreased with aging. Thus, neuronal inhibition generally progresses with development and deteriorates with aging, with some noted regional differences. This was the first study to systematically describe the development and aging of brain deactivation, which may reflect regional neuronal inhibition.

Yoshida, Kazunori; An, Qi; Hamada, Hiroyuki; Yamakawa, Hiroshi; Tamura, Yusuke; Yamashita, Atsushi; Asama, Hajime

doi: 10.1080/01691864.2021.1917452pmid: N/A

Sit-to-stand motion is an important daily activity, and it is important to study the mechanism of the motion to improve the ability when it becomes weak. To study the mechanism, we hypothesized that muscle synergy generates muscle activity as a feedforward signal, which is modified by sensory input. This study focuses on determining the sensory input primarily used for modifying sit-to-stand motion. To obtain this, we built artificial neural network models that generate muscle activities based on sensory input and feedforward signals and analyzed the effect of each input on the output. The models were built for each motion phase. The input was information from vestibular and somatosensory input and averaged muscle synergy as feedforward signals, and the output was muscle synergy. As a result, it was revealed that humans may primarily use hip angle to bend forward, ankle and vertical foot reaction force to hip rise, ankle, knee, and lumber angles and vertical foot reaction force to extend body, and lumber angle to stabilize. This indicates the type of sensory input used to control each muscle synergy in each motion phase. The information should be used to modify the sit-to-stand motion in environmental conditions where the motion is performed.

Kogami, Hiroki; An, Qi; Yang, Ningjia; Wang, Ruoxi; Yoshida, Kazunori; Hamada, Hiroyuki; Yamakawa, Hiroshi; Tamura, Yusuke; Shimoda, Shingo; Yamasaki, Hiroshi; Yokoyama, Moeka; Alnajjar, Fady; Hattori, Noriaki; Takahashi, Koji; Fujii, Takanori;

Li, Dongdong; Kaminishi, Kohei; Chiba, Ryosuke; Takakusaki, Kaoru; Mukaino, Masahiko; Ota, Jun

doi: 10.1080/01691864.2021.1893218pmid: N/A

Strokes are the third leading cause of disability worldwide. Currently, several types of performance assessment, such as the Fugl–Meyer index, are used to explore the overall difference between cerebral infarction (CI) and cerebral hemorrhage (CH) post-stroke patients. However, these performance assessments ignore subtle differences in the limbs of patients, which could be helpful for rehabilitation training. This study was designed to determine and evaluate the differences between the limbs of CI and CH patients. First, we collected the kinematic data of patients and extracted the spatio-temporal features. Then, we developed four different models to classify the CI and CH patients, in which a linear support vector machine (LSVM) classifier method achieved an 80.1% classification accuracy. Finally, we calculated the decision boundary of the shoulder and ankle marker position features separately based on the LSVM model. From the decision boundary, we determined that the CI patients' shoulder position appeared to be anterior to that of the CH patients, and the CH patients had a wider stance width compared to the CI patients. Such findings can serve as guidance for doctors and help provide professional rehabilitation courses for post-stroke patients.

Yozu, Arito; Kaminishi, Kohei; Ishii, Daisuke; Omura, Yuichiro; Matsushita, Akira; Kohno, Yutaka; Chiba, Ryosuke; Ota, Jun

doi: 10.1080/01691864.2021.1948353pmid: N/A

Parkinson's disease (PD) is a common neurological disorder, which causes instability in standing balance. The risk factors for falls in PD include disease severity, medication state (On/Off state), and dual tasking. The aim of this study was to assess the effects of dopaminergic medication and dual tasking on standing balance in PD by evaluating not only the center of pressure but also muscle cocontraction. Two patients with PD were included in this pilot case study. Relational trends were found among physical symptoms, COP, and muscle cocontraction. Evaluations of COP and cocontraction may be useful in making decisions for rehabilitation prescription.

Inoue, Takahisa; Terada, Shin; Matsuzaki, Masanori; Izawa, Jun

doi: 10.1080/01691864.2021.1912637pmid: N/A

Computational models of motor learning have been examined experimentally with human participants using a robotic manipulandum that produces force perturbations on their reach trajectories. Here, we developed a small-scale robotic manipulandum for rodent studies. This device enabled us to conduct the same motor control experiments typically used for human study on rodents in order to examine the neural basis of the computational process of motor control and learning. We present the design of the manipulandum and controller system in addition to kinematic analysis of the link structures. We also confirmed the feasibility of the developed system for rodent studies.