This study investigates the properties of motor pools in the human motor control system. The simulations carried out as part of this study used two biologically inspired neuronal models to simulate networks with properties similar to those observed in the human motor system (Burke, 1991). The Synchronous neuronal model developed as part of this study explicitly models the input/output spike train and frequency relationship of each neuron. The motor pool simulations were carried out using the INSIGHT TOO simulation software developed as part of this study. INSIGHT TOO is a flexible neural design tool that allows the visual interactive design of network connectivity and has the power of a node specification language similar to that of BASIC that allows multi-layer, multi-model networks to be simulated. The simulations have shown that the motor pools are capable of reproducing commonly observed physiological properties during normal voluntary reaching movements. As a result of these findings a theoretical model of ballistic voluntary motor action was proposed called the Recruitment Model. The Recruitment model utilises the "recruitment" principle known to exist in motor pools and applies this distributed processing methodology to the higher levels of motor action to explain how complex structures similar to the human skeletal system might be controlled. A simple version of the Recruitment Model is simulated showing an animation of a running "stick man". This simulation demonstrates some of the principles necessary to solve problems relating to synergy formation.

We study the appearance of a novel phenomenon for coupled identical bursters: synchronized bursts where there are changes of spike synchrony within each burst. The examples we study are for normal form elliptic bursters where there is a periodic slow passage through a Bautin (codimension two degenerate Andronov-Hopf) bifurcation. This burster has a subcritical Andronov-Hopf bifurcation at the onset of repetitive spiking while the end of burst occurs via a fold limit cycle bifurcation. We study synchronization behavior of two Bautin-type elliptic bursters for a linear direct coupling scheme as well as demonstrating its presence in an approximation of gap-junction and synaptic coupling. We also find similar behaviour in system consisted of three and four Bautin-type elliptic bursters. We note that higher order terms in the normal form that do not affect the behavior of a single burster can be responsible for changes in synchrony pattern; more precisely, we find within-burst synchrony changes associated with a turning point in the spontaneous spiking frequency (frequency transition). We also find multiple synchrony changes in similar system by incorporating multiple frequency transitions. To explain the phenomenon we considered a burst-synchronized constrained model and a bifurcation analysis of the this reduced model shows the existence of the observed within-burst synchrony states. Within-burst synchrony change is also found in the system of mutually delaycoupled two Bautin-type elliptic bursters with a constant delay. The similar phenomenon is shown to exist in the mutually-coupled conductance-based Morris-Lecar neuronal system with an additional slow variable generating elliptic bursting. We also find within-burst synchrony change in linearly coupled FitzHugh-Rinzel 2 3 elliptic bursting system where the synchrony change occurs via a period doubling bifurcation. A bifurcation analysis of a burst-synchronized constrained system identifies the periodic doubling bifurcation in this case. We show emergence of spontaneous burst synchrony cluster in the system of three Hindmarsh-Rose square-wave bursters with nonlinear coupling. The system is found to change between the available cluster states depending on the stimulus. Lyapunov exponents of the burst synchrony states are computed from the corresponding variational system to probe the stability of the states. Numerical simulation also shows existence of burst synchrony cluster in the larger network of such system.

Arrays of broadly-selective chemical sensors, also known as electronic noses, have been developed during the past two decades as a low-cost and high-throughput alternative to analytical instruments for the measurement of odorant chemicals. Signal processing in these gas-sensor arrays has been traditionally performed by means of statistical and neural pattern recognition techniques. The objective of this dissertation is to develop new computational models to process gas sensor array signals inspired by coding and learning mechanisms of the biological olfactory system. We have used a neurodynamic model of the olfactory system, the KIII, to develop and demonstrate four odor processing computational functions: robust recovery of overlapping patterns, contrast enhancement, background suppression, and novelty detection. First, a coding mechanism based on the synchrony of neural oscillations is used to extract information from the associative memory of the KIII model. This temporal code allows the KIII to recall overlapping patterns in a robust manner. Second, a new learning rule that combines Hebbian and anti-Hebbian terms is proposed. This learning rule is shown to achieve contrast enhancement on gas-sensor array patterns. Third, a new local learning mechanism based on habituation is proposed to perform odor background suppression. Combining the Hebbian/anti-Hebbian rule and the local habituation mechanism, the KIII is able to suppress the response to continuously presented odors, facilitating the detection of the new ones. Finally, a new learning mechanism based on anti-Hebbian learning is proposed to perform novelty detection. This learning mechanism allows the KIII to detect the introduction of new odors even in the presence of strong backgrounds. The four computational models are characterized with synthetic data and validated on gas sensor array patterns obtained from an e-nose prototype developed for this purpose.

Arrays of broadly-selective chemical sensors, also known as electronic noses, have been developed during the past two decades as a low-cost and high-throughput alternative to analytical instruments for the measurement of odorant chemicals. Signal processing in these gas-sensor arrays has been traditionally performed by means of statistical and neural pattern recognition techniques. The objective of this dissertation is to develop new computational models to process gas sensor array signals inspired by coding and learning mechanisms of the biological olfactory system. We have used a neurodynamic model of the olfactory system, the KIII, to develop and demonstrate four odor processing computational functions: robust recovery of overlapping patterns, contrast enhancement, background suppression, and novelty detection. First, a coding mechanism based on the synchrony of neural oscillations is used to extract information from the associative memory of the KIII model. This temporal code allows the KIII to recall overlapping patterns in a robust manner. Second, a new learning rule that combines Hebbian and anti-Hebbian terms is proposed. This learning rule is shown to achieve contrast enhancement on gas-sensor array patterns. Third, a new local learning mechanism based on habituation is proposed to perform odor background suppression. Combining the Hebbian/anti-Hebbian rule and the local habituation mechanism, the KIII is able to suppress the response to continuously presented odors, facilitating the detection of the new ones. Finally, a new learning mechanism based on anti-Hebbian learning is proposed to perform novelty detection. This learning mechanism allows the KIII to detect the introduction of new odors even in the presence of strong backgrounds. The four computational models are characterized with synthetic data and validated on gas sensor array patterns obtained from an e-nose prototype developed for this purpose.

Modelar a dinâmica de neurônios é relevante em estudos de neurociências. Neste trabalho, propõe-se um modelo computacional de neurônio baseado no comportamento dos canais iônicos presentes na sua membrana. O modelo combina elementos microscópicos, como o comportamento dos canais individuais, com elementos macroscópicos, como a tensão ao longo de um trecho de membrana. Simulações foram realizadas com o objetivo de reproduzir dados biológicos e resultados obtidos de modelos teóricos clássicos da área. Foi possível reproduzir com boa concordância o potencial de ação, o fenômeno da adaptação, a curva da corrente de entrada versus a frequência de disparos e o potencial excitatório pós-sináptico. / Modelling the dynamics of neurons is relevant in studies on neurosciences. In this work, a computational model of neuron based on the behavior of the ionic channels found in its membrane is proposed. The model comprises microscopic elements, as the behavior of the individual channels, and macroscopic elements, as the tension along a membrane patch. Simulations were performed with the aim of reproducing biological data and results derived from classical theoretical models of the field. It was possible to reproduce with good agreement the action potential, the phenomenon of adaptation, the curve of the input current versus the spike frequency, and the excitatory postsynaptic potential.

Modelar a dinâmica de neurônios é relevante em estudos de neurociências. Neste trabalho, propõe-se um modelo computacional de neurônio baseado no comportamento dos canais iônicos presentes na sua membrana. O modelo combina elementos microscópicos, como o comportamento dos canais individuais, com elementos macroscópicos, como a tensão ao longo de um trecho de membrana. Simulações foram realizadas com o objetivo de reproduzir dados biológicos e resultados obtidos de modelos teóricos clássicos da área. Foi possível reproduzir com boa concordância o potencial de ação, o fenômeno da adaptação, a curva da corrente de entrada versus a frequência de disparos e o potencial excitatório pós-sináptico. / Modelling the dynamics of neurons is relevant in studies on neurosciences. In this work, a computational model of neuron based on the behavior of the ionic channels found in its membrane is proposed. The model comprises microscopic elements, as the behavior of the individual channels, and macroscopic elements, as the tension along a membrane patch. Simulations were performed with the aim of reproducing biological data and results derived from classical theoretical models of the field. It was possible to reproduce with good agreement the action potential, the phenomenon of adaptation, the curve of the input current versus the spike frequency, and the excitatory postsynaptic potential.

Made available in DSpace on 2016-03-15T19:37:40Z (GMT). No. of bitstreams: 1 Alex Martins.pdf: 1863982 bytes, checksum: 63a3f4efd397697e6bc129fa070520d5 (MD5) Previous issue date: 2011-10-19 / Synchronization can be understood as a temporal organization of events, able of emerging in complex systems, as neural networks. Here, random graph and cellular automaton (CA) are used to represent neural networks, in order to investigate the occurrence of synchronism in such networks. The network coupling topology is of Newman-Watts type, formed by regular lattice with additional random connections. Two parts with this structure are connected by random links. Results obtained from numerical simulations with this model indicate variety of oscillatory behavior: there are cases in which both parts oscillate with equal, multiple and submultiple periods; and cases without oscillation. Investigations were performed concerning the relation among oscillatory behavior and maximum activity, the time to reach such an activity, the minimum average path length, size of the network, the percentage of random connections added and the rules of the CA state transition. Synchronous behavior was found in more than 75% of 28000 simulations accomplished. The system dynamics is influenced more by variations on the number of time steps in which a cell remains firing than by alterations on the lattice size or on the percentage of the randomly added links. / Pode-se entender sincronismo como uma organização temporal de eventos, possível de emergir em sistemas complexos, como redes neurais. Aqui, usam-se grafo aleatório e autômato celular (AC) para representar redes neurais, a fim de investigar a ocorrência de sincronismo em tais redes. A topologia de acoplamento da rede é do tipo Newman-Watts, formada por uma grade regular com ligações aleatórias acrescentadas. Duas partes com essa estrutura são conectadas por ligações aleatórias. Resultados obtidos por simulações numéricas com esse modelo indicam diversidade de comportamento oscilatório: há casos em que as duas partes oscilam em períodos iguais, múltiplos e submúltiplos; e casos sem oscilação. Investigaram-se as relações entre comportamento oscilatório e a atividade máxima, o tempo para se alcançar essa atividade, o comprimento do caminho mínimo médio, o tamanho da rede, a porcentagem de ligações aleatórias adicionadas, e as regras de transição de estado do AC. Comportamento síncrono foi encontrado em mais de 75% das 28.000 simulações realizadas. A dinâmica do sistema é mais influenciada por variações no número de passos de tempo em que a célula permanece disparando do que por alterações no tamanho do reticulado ou no percentual das ligações aleatórias adicionais.

<p>Neurodynamics refers to the mechanical and physiological components of</p> <p>the nervous system and the interconnections between them (Shacklock, 1995).</p> <p>This is a phase 1 pilot trial investigating the immediate effect of a neurodynamic</p> <p>treatment as compared to a sham treatment in eight participants with low back</p> <p>pain. Primary outcome measures included: H-reflex latency and nerve</p> <p>conduction velocity. Secondary outcome measures included: the sitting slump</p> <p>test and visual analog scale for pain following a neurodynamic treatment</p> <p>compared to a sham treatment on eight participants with low back pain. T-tests</p> <p>were used to analyze any differences between the groups at baseline and post-</p> <p>intervention. No statistically significant differences were observed between the</p> <p>groups at baseline. Statistically significant differences were noted post-</p> <p>intervention between the treatment groups for H-reflex latency (t(5)=4.323,</p> <p>p=0.008) and the unaffected leg sitting slump test (t(5)=3.402, p=0.019). The H-</p> <p>reflex latency increased for the group following the neurodynamic treatment and</p> <p>decreased following the sham treatment. This was not expected and is of</p> <p>interest due to the possible mechanisms that may be underlying these</p> <p>phenomena. Despite the small sample size used in this study, differences were</p> <p>observed and displayed trends that were unanticipated. These between-group</p> <p>differences are of interest but require further investigation using a larger sample</p> <p>population. Sample size calculations for future studies based on the primary</p> <p>outcome measures yielded a sample of 2008 participants. This accounted for</p> <p>both a 20% difference between the two groups and a 20% dropout rate. Future</p> <p>studies need to investigate the most beneficial length of time, type and dosage of</p> <p>neurodynamic treatments, as well as, the most appropriate times to assess the</p> <p>outcome measures. Comparison to controls would be beneficial in subsequent</p> <p>studies.</p> / Master of Science Rehabilitation Science (MSc)

El aprendizaje visuomotor condicional es un paradigma en el que las asociaciones estímulo-respuesta se aprenden a través de una recompensa. Un experimento típico se desarrolla de la siguiente forma: cuando se presenta un estímulo a un sujeto, éste debe decidir qué acción realizar de entre un conjunto. Una vez seleccionada la acción, el sujeto recibirá una recompensa en el caso de que la acción escogida sea correcta. En este tipo de tareas interactúan distintas regiones cerebrales, entre las que destacan el córtex prefrontal, el córtex premotor, el hipocampo y los ganglios basales. El objetivo de este proyecto consiste en estudiar la dinámica neuronal subyacente a esta clase de tareas a través de modelos computacionales. Proponemos el término processing pathway para describir cómo ejecuta esta tarea el cerebro y explicar los roles e interacciones entre las distintas áreas cerebrales. Además, tratamos el procesamiento anómalo en una hipótesis neurodinámica de la esquizofrenia. / Conditional visuomotor learning is a paradigm in which stimulus-response associations are learned upon reward feedback. A typical experiment is as follows: Upon a stimulus presentation, a subject has to decide which action to choose among a number of actions. After an action is selected, the subject receives reward if the action was correct. Several interacting brain regions work together to perform this task, most prominently the prefrontal cortex, the premotor cortex, the hippocampus, and the basal ganglia. Using computational modeling, we analyze and discuss the neurodynamics underlying this task. We propose the term processing pathway to describe how the brain performs this task and detail the roles and interactions of the brain regions. In addition, we address aberrant processing in a neurodynamical hypothesis of schizophrenia.

In der vorliegenden Arbeit werden rekurrente neuronale Netze im Hinblick auf ihre Eignung zur Bewegungssteuerung autonomer Roboter untersucht. Nacheinander werden Oszillatoren für Vierbeiner, homöostatische Ringmodule für segmentierte Roboter und monostabile Neuromodule für Roboter mit vielen Freiheitsgraden und komplexen Bewegungsabläufen besprochen. Neben dem mathematisch-theoretischen Hintergrund der Neuromodule steht in gleichberechtigter Weise deren praktische Implementierung auf realen Robotersystemen. Hierzu wird die funktionale Einbettung ins Gesamtsystem ebenso betrachtet, wie die konkreten Aspekte der zugrundeliegenden Hardware: Rechengenauigkeit, zeitliche Auflösung, Einfluss verwendeter Materialien und dergleichen mehr. Interessante elektronische Schaltungsprinzipien werden detailliert besprochen. Insgesamt enthält die vorliegende Arbeit alle notwendigen theoretischen und praktischen Informationen, um individuelle Robotersysteme mit einer angemessenen Bewegungssteuerung zu versehen. Ein weiteres Anliegen der Arbeit ist es, aus der Richtung der klassischen Ingenieurswissenschaften kommend, einen neuen Zugang zur Theorie rekurrenter neuronaler Netze zu schaffen. Gezielte Vergleiche der Neuromodule mit analogen elektronischen Schaltungen, physikalischen Modellen und Algorithmen aus der digitalen Signalverarbeitung können das Verständnis von Neurodynamiken erleichtern. / How recurrent neural networks can help to make autonomous robots move, will be investigated within this thesis. First, oscillators which are able to control four-legged robots will be dealt with, then homeostatic ring modules which control segmented robots, and finally monostable neural modules, which are able to drive complex motion sequences on robots with many degrees of freedom will be focused upon. The mathematical theory of neural modules will be addressed as well as their practical implementation on real robot platforms. This includes their embedding into a major framework and concrete aspects, like computational accuracy, timing and dependance on materials. Details on electronics will be given, so that individual robot systems can be built and equipped with an appropriate motion controller. It is another concern of this thesis, to shed a new light on the theory of recurrent neural networks, from the perspective of classical engineering science. Selective comparisons to analog electronic schematics, physical models, and digital signal processing algorithms can ease the understanding of neural dynamics.