Loading...

Conceptual Evaluation of Brain Internal Models and Modeling their Uncertainty

Sirjani, Nasim | 2018

470 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 51804 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Jahed, Mehran
  7. Abstract:
  8. In the literature of motor behavior, rapid and harmonious human movements and adaptation to limb dynamics are described with reference to internal models. Concepts of internal models in neuroscience cognitive sciences and computational modeling have been investigated and their structure and performance have been evaluated by behavioral, neurophysiological and medical imaging data. The purpose of the theory of internal models is to explain the mechanism of human movement. Based on this theory, the central nervous system uses internal models to predict and adjust movements and reduce uncertainty. Uncertainty refers to the unknown dynamics of the system due to the lack of access to accurate information about the system model and the effect of external disturbances on it. In this reseach, the role of internal models in modeling of uncertainty in the form of a disturbance is presented and in this regard, a comprehensive model is presented. Therefore, for the unknown dynamic modeling, a “bias adaptive component” is used in the inverse internal model and used within the "Neural Engineering Framework" (NEF). To evaluate the performance of the inverse model in the unknown dynamics model, the movements of the hand in straight and circular paths are used in the presence of an external field acting as a disturbance. The results show that the system is not able to trace the path correctly if the inverse model is not used in the system. As a result, the reverse internal model in the cerebellum seems to reduce uncertainty in the human motor system
  9. Keywords:
  10. Uncertainty ; Unknown Dynamics ; Internal Model Principle ; Reverse Internal Model ; Motion System

 Digital Object List

 Bookmark

No TOC