Peixes elétricos de campo fraco comunicam-se por meio de descargas elétricas de forma de onda estereotipada, variando somente o intervalo entre pulsos de acordo com a informação a ser transmitida. Esse mecanismo de codificação é similar ao encontrado em diversos circuitos de neurônios conhecidos, o que torna esses animais excelentes modelos para o estudo de sistemas de comunicação naturais, permitindo experimentos que envolvem tanto aspectos comportamentais como neuroetológicos. É um desafio realizar análises de dados coletados de mais de um peixe nadando livremente, pois os padrões de descargas de órgão elétrico (DOE) dependem da posição dos animais e de suas orientações com relação aos eletrodos de medida. Contudo, como cada peixe emite uma forma de onda de DOE característica, ferramentas computacionais podem ser empregadas para associar cada DOE ao respectivo peixe. Neste trabalho, descrevemos um método computacional capaz de reconhecer DOEs de pares de peixes usando vetores de características normalizados, obtidos aplicando a transformada de Fourier e a transformada complexa de dupla árvore de pacote wavelet. Empregamos máquinas de vetores de suporte como classificadores, e um algoritmo de regra de continuidade permite resolver problemas causados por DOEs sobrepostas e saturação de sinais. Procedimentos de validação com Gymnotus sp. mostraram que as DOEs podem ser atribuídas corretamente a cada peixe com apenas dois erros por milhão de descargas. Para permitir que esse processo de discriminação ocorra em tempo real, implementamos uma arquitetura de hardware dedicada e maciçamente paralela em um field programmable gate array (FPGA) para executar a etapa de maior esforço computacional do algoritmo de discriminação. Como resultado, obtivemos um sistema híbrido de hardware e software de tempo real que foi capaz de atender a um requisito de latência máxima de 1 ms, o que permite mimetizar o tempo de resposta de importantes sistemas sensoriais elétricos de Gymnotus sp. Com o auxílio de nossa instrumentação, diversos experimentos com realimentação poderão ser propostos, permitindo que um modelo computacional interaja com dois peixes em uma preparação in vivo naturalística. / Pulse-type weakly electric fishes communicate through electrical discharges with a stereotyped waveform, varying solely the interval between pulses according to the information being transmitted. This simple codification mechanism is similar to the one found in various known neuronal circuits, which renders these animals as good models for the study of natural communication systems, allowing experiments involving behavioral and neuroethological aspects. Performing analysis of data collected from more than one freely swimming fish is a challenge since the detected electric organ discharge (EOD) patterns are dependent on each animal´s position and orientation relative to the electrodes. However, since each fish emits a characteristic EOD waveform, computational tools can be employed to match each EOD to the respective fish. In this work we describe a computational method able to recognize fish EODs from dyads using normalized feature vectors obtained by applying Fourier and dual-tree complex wavelet packet transforms. We employ support vector machines as classifiers, and a continuity constraint algorithm allows us to solve issues caused by overlapping EODs and signal saturation. Validation procedures with Gymnotus sp. showed that EODs can be assigned correctly to each fish with only two errors per million discharges. In order to allow this discrimination process to occur in real-time, we implemented a massively parallel application-specific hardware architecture in a field programmable gate array (FPGA) to run the discrimination algorithm step whose computational effort is the highest among the others. As a result, we obtained a hardware and real-time software co-design that was able to meet a maximum latency requirement of 1 ms, allowing it to mimic the response time of major electrical sensory systems of Gymnotus sp. Our instrumentation will enable the proposal of several novel experiments with closed-loop feedback, allowing a computer model to interact with two fish in a naturalistic in vivo preparation.

Peixes elétricos de campo fraco comunicam-se por meio de descargas elétricas de forma de onda estereotipada, variando somente o intervalo entre pulsos de acordo com a informação a ser transmitida. Esse mecanismo de codificação é similar ao encontrado em diversos circuitos de neurônios conhecidos, o que torna esses animais excelentes modelos para o estudo de sistemas de comunicação naturais, permitindo experimentos que envolvem tanto aspectos comportamentais como neuroetológicos. É um desafio realizar análises de dados coletados de mais de um peixe nadando livremente, pois os padrões de descargas de órgão elétrico (DOE) dependem da posição dos animais e de suas orientações com relação aos eletrodos de medida. Contudo, como cada peixe emite uma forma de onda de DOE característica, ferramentas computacionais podem ser empregadas para associar cada DOE ao respectivo peixe. Neste trabalho, descrevemos um método computacional capaz de reconhecer DOEs de pares de peixes usando vetores de características normalizados, obtidos aplicando a transformada de Fourier e a transformada complexa de dupla árvore de pacote wavelet. Empregamos máquinas de vetores de suporte como classificadores, e um algoritmo de regra de continuidade permite resolver problemas causados por DOEs sobrepostas e saturação de sinais. Procedimentos de validação com Gymnotus sp. mostraram que as DOEs podem ser atribuídas corretamente a cada peixe com apenas dois erros por milhão de descargas. Para permitir que esse processo de discriminação ocorra em tempo real, implementamos uma arquitetura de hardware dedicada e maciçamente paralela em um field programmable gate array (FPGA) para executar a etapa de maior esforço computacional do algoritmo de discriminação. Como resultado, obtivemos um sistema híbrido de hardware e software de tempo real que foi capaz de atender a um requisito de latência máxima de 1 ms, o que permite mimetizar o tempo de resposta de importantes sistemas sensoriais elétricos de Gymnotus sp. Com o auxílio de nossa instrumentação, diversos experimentos com realimentação poderão ser propostos, permitindo que um modelo computacional interaja com dois peixes em uma preparação in vivo naturalística. / Pulse-type weakly electric fishes communicate through electrical discharges with a stereotyped waveform, varying solely the interval between pulses according to the information being transmitted. This simple codification mechanism is similar to the one found in various known neuronal circuits, which renders these animals as good models for the study of natural communication systems, allowing experiments involving behavioral and neuroethological aspects. Performing analysis of data collected from more than one freely swimming fish is a challenge since the detected electric organ discharge (EOD) patterns are dependent on each animal´s position and orientation relative to the electrodes. However, since each fish emits a characteristic EOD waveform, computational tools can be employed to match each EOD to the respective fish. In this work we describe a computational method able to recognize fish EODs from dyads using normalized feature vectors obtained by applying Fourier and dual-tree complex wavelet packet transforms. We employ support vector machines as classifiers, and a continuity constraint algorithm allows us to solve issues caused by overlapping EODs and signal saturation. Validation procedures with Gymnotus sp. showed that EODs can be assigned correctly to each fish with only two errors per million discharges. In order to allow this discrimination process to occur in real-time, we implemented a massively parallel application-specific hardware architecture in a field programmable gate array (FPGA) to run the discrimination algorithm step whose computational effort is the highest among the others. As a result, we obtained a hardware and real-time software co-design that was able to meet a maximum latency requirement of 1 ms, allowing it to mimic the response time of major electrical sensory systems of Gymnotus sp. Our instrumentation will enable the proposal of several novel experiments with closed-loop feedback, allowing a computer model to interact with two fish in a naturalistic in vivo preparation.

Animals must sense their surroundings to update their internal representations of the external environment, and exploratory behaviours such as sensory sampling are influenced by past experiences. This thesis investigates how voluntary sensory sampling activities undergo learning-dependent changes. Studies of freely behaving animals impose two major challenges: 1) the accuracy of biological measurements is compromised by movement-induced artifacts, and 2) large degrees of freedom in unrestrained behaviours confound well-controlled studies. Pulse-type weakly electric fish (WEF) are an ideal choice to study adaptive sensory sampling from unrestrained animals, since they generate readily observable and quantifiable sensory capture events expressed by discrete pulses of electric organ discharges (EODs). To study the voluntarily movements and sensory sampling while animals navigated in darkness, we developed three novel experimental techniques to track movements and detect sensory sampling from a freely behaving WEF: 1) an EOD detector to remotely and accurately measure the sensory sampling rate, 2) an electrical tracking method to track multiple WEF using their own EODs, and 3) visual tracking algorithm for robust body tracking through water under infrared illumination. These techniques were successfully applied to reveal novel sensory sampling behaviours in freely exploring Gymnotus sp. Cortical activity precedes self-initiated movements by several seconds in mammals; this observation has led into inquiries on the nature of volition. Here we demonstrate the sensory sampling enhancement also precedes self-initiated movement by a few seconds in Gymnotus sp. Next, we tested whether these animals can be trained to learn a location of food using electrically detectable landmarks and, if so, whether they can use their past experiences to optimize their sensory sampling. We found that animals revisited the missing food location with high spatial accuracy, and they intensified their sensory sampling near the expected food location by increasing the number of EOD pulses per unit distance travelled.

Sensorische Nervenzellen kodieren Informationen über die Umwelt mittels elektrischer Impulse, sogenannte Aktionspotentiale oder Spikes. Diese werden weitergeleitet zu postsynaptischen Neuronen im zentralen Nervensystem, welche unterschiedliche Auslesestrategien verwenden. Integratorzellen summieren alle ankommenden Aktionspotentiale auf, wodurch sie die Gesamtaktivität einer präsynaptischen Population messen. Koinzidenzdetektoren hingegen, werden nur durch das synchrone Feuern der zuführenden Neuronenpopulation aktiviert. Die grundlegende Frage dieser Dissertation lautet: Welche Information eines zeitabhängigen Signals kodieren die synchronen Spikes einer Neuronenpopulation im Vergleich zu der Summe all ihrer Aktionspotentiale? Hierbei verwenden wir die Theorie stochastischer Prozesse: wir berechnen Spektralmaße, die es ermöglichen Aussagen darüber zu treffen welche Frequenzkomponenten eines Signals vorwiegend transmittiert werden. Im Gegensatz zu früheren Studien, verstehen wir unter einem synchronen Ereignis nicht zwangsläufig, dass die gesamte Population simultan feuert, sondern, dass ein minimaler Anteil („Synchronizitätsschranke") gleichzeitig aktiv ist. Unsere Analyse zeigt, dass die synchrone Populationsaktivität als ein Bandpass-Informationsfilter agieren kann: die synchronen Spikes kodieren hauptsächlich schnelle Signalanteile. Damit stellt die Selektion simultaner Neuronenaktivität ein potentielles Mittel dar um gleichzeitig anwesende, konkurrierende Signale voneinander zu trennen. Dabei hängen die genauen Charakteristika der Informationsfilterung ausschlaggebend von der Synchronizitätsschwelle ab. Insbesondere zeigt sich, dass eine Symmetrie in der Schwelle vorliegt,die die Äquivalenz der Kodierungseigenschaften von synchronem Feuern und synchronem Schweigen offenlegt. Unsere analytischen Ergebnisse testen wir mittels numerischer Simulationen und vergleichen sie mit Experimenten am schwach elektrischen Fisch. / Populations of sensory neurons encode information about the environment into electrical pulses, so called action potentials or spikes. Neurons in the brain process these pulses further by using different readout strategies. Integrator cells sum up all incoming action potentials and are thus sensitive to the overall activity of a presynaptic population. Coincidence detectors, on the other hand, are activated by the synchronous firing of the afferent population. The main question of this thesis is: What information about a common time-dependent stimulus is encoded in the synchronous spikes of a neuronal population in comparison to the sum of all spikes? We approach this question within the framework of spectral analysis of stochastic processes, which allows to assess which frequency components of a signal are predominantly encoded. Here, in contrast to earlier studies, a synchronous event does not necessarily mean that all neurons of the population fire simultaneously, but that at least a prescribed fraction ('synchrony threshold') needs to be active within a small time interval. We derive analytical expressions of the correlation statistics which are compared to numerical simulations and experiments on weakly electric fish. We show that the information transmission of the synchronous output depends highly on the synchrony threshold. We uncover a symmetry in the synchrony threshold, unveiling the similarity in the encoding capability of the common firing and the common silence of a population. Our results demonstrate that the synchronous output can act as a band-pass filter of information, i.e. it extracts predominantly fast components of a stimulus. If signals in different frequency regimes are concurrently present, the selection of synchronous firing events can thus be a tool to separate these signals.

Made available in DSpace on 2016-04-28T14:16:38Z (GMT). No. of bitstreams: 1 Gerda Maisa Jensen.pdf: 2040855 bytes, checksum: 12775de75b437b43c13fd5771e89f26e (MD5) Previous issue date: 2008-12-17 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The Italian naturalist Lazzaro Spallanzani (1729-1799) took up the study of several issues of Natural History amongst which, the numbness phenomenon caused by fish called torpedoes in their preys or in human beings. Well known since the Ancient times once a fable, it became subsequently the object of inquiry of the natural philosophers from XVII and XVIII centuries. They witnessed the veracity of the phenomenon and described the morphology and the anatomic of these fish, trying to explain it origin and cause of the numbness. In this Dissertation, we present some of the inquiries on the subject conducted by Spallanzani, to ascertain whether they were based on the knowledge available in the period and if his observations and experiments were well planned and executed. To do so, we present some of the explanations for the phenomenon, ranging from mechanical and electrical hypotheses, for some predecessors and contemporaries of Spallanzani. Then, in order to contextualize the study in the general scope of the investigations of Spallanzani, details of his training and academic activities are given, with emphasis on his naturalistics travels and maintenance of museums of Natural History. Finally, the process of investigation that he conducted on the torpedoes, between the years 1780 and 1782 is detailed. The report of his observations and experiences as well as the results were summarized by Spallanzani in a letter dated February 23 of 1783, addressed to the Marchis Girolamo Lucchesini and published in the same year, in two journals aimed at disseminating scientific Italians, the Opuscoli scelti sulle Scienze e sulle Arti and the Giornale de Letterati. We compared this letter with to excerpts of the diaries, published only recently, about his naturalistic journeys that provided Spallanzani access and opportunity to conduct their studies on the torpedoes. Our analysis led us to conclude that Spallanzani took into account the past and contemporary studies on the phenomenon, his guided through their research hypotheses (mechanical and electrical) that subsidized these studies. Spallanzani concluded that although the shock caused by the fish does not produce sparks or express phenomena of attraction and repulsion (as occurred with static electricity studied at the time), the numbness caused by torpedoes in their preys, in the hands of fishermen and scholars was an electrical phenomenon of the same nature as manifest in the atmosphere or the bottle of Leyden / O naturalista italiano Lazzaro Spallanzani (1729-1799) dedicou-se ao estudo de diversos temas da História Natural, dentre os quais o do fenômeno de entorpecimento causado em presas ou em seres humanos, por peixes chamados torpedos. Conhecido pelos sábios da Antigüidade, o fenômeno chegou a ser considerado, posteriormente, uma fábula, até que filósofos naturais dos séculos XVII e XVIII tomaram-no como objeto de investigação. Atestaram não apenas a veracidade do fenômeno, como procederam a descrições morfológicas e anatômicas desses peixes, procurando explicar a origem e causa do fenômeno. Nesta Dissertação, apresentamos algumas das investigações sobre o tema realizadas por Spallanzani, objetivando averiguar se estavam baseadas nos conhecimentos disponíveis no período e se suas observações e experiências foram bem planejadas e executadas. Para isso, são apresentadas algumas das explicações fornecidas para o fenômeno, variando entre hipóteses mecânicas e elétricas, por alguns antecessores e contemporâneos de Spallanzani. Em seguida, a fim de contextualizar esse estudo no âmbito geral das pesquisas de Spallanzani, são indicados elementos de sua formação e atividades acadêmicas, com ênfase nas suas viagens naturalísticas e manutenção de museus de História Natural. Por fim, é detalhada uma fase de investigações que ele realizou sobre os torpedos, entre os anos de 1780 e 1782. O relato de suas observações e experiências e os resultados obtidos foram resumidos por Spallanzani em carta datada de 23 de fevereiro de 1783, endereçada ao Marchese Girolamo Lucchesini e publicada, no mesmo ano, em dois periódicos italianos voltados à divulgação científica, o Opuscoli scelti sulle Scienze e sulle Arti e o Giornale de Letterati. Comparamos essa carta com trechos dos diários, apenas recentemente publicados, das viagens naturalísticas que propiciaram a Spallanzani o acesso e a possibilidade de realizar seus estudos sobre os torpedos. Nossa análise nos levou a concluir que Spallanzani levou em consideração os estudos anteriores e contemporâneos sobre o fenômeno, bem como guiou sua investigação pelas hipóteses (mecânica e elétrica) que subsidiavam esses estudos. Spallanzani concluiu que apesar do choque causado pelo peixe não produzir faísca nem manifestar fenômenos de atração e repulsão (como ocorria com a eletricidade estática estudada na época), o entorpecimento causado pelos torpedos nas presas, nas mãos dos pescadores e de estudiosos era um fenômeno elétrico, de mesma natureza que o manifesto na atmosfera ou na garrafa de Leyden

One of the most powerful and important applications that the brain accomplishes is solving the sensory "cocktail party problem:" to adaptively suppress extraneous signals in an environment. Theoretical studies suggest that the solution to the problem involves an adaptive filter, which learns to remove the redundant noise. However, neural learning is also in its infancy and there are still many questions about the stability and application of synaptic learning rules for neural computation. In this thesis, the implementation of an adaptive filter in the brain of a weakly electric fish, A. Leptorhynchus, was studied. It was found to require a cerebellar architecture that could supply independent frequency channels of delayed feedback and multiple burst learning rules that could shape this feedback. This unifies two ideas about the function of the cerebellum that were previously separate: the cerebellum as an adaptive filter and as a generator of precise temporal inputs.

One of the most powerful and important applications that the brain accomplishes is solving the sensory "cocktail party problem:" to adaptively suppress extraneous signals in an environment. Theoretical studies suggest that the solution to the problem involves an adaptive filter, which learns to remove the redundant noise. However, neural learning is also in its infancy and there are still many questions about the stability and application of synaptic learning rules for neural computation. In this thesis, the implementation of an adaptive filter in the brain of a weakly electric fish, A. Leptorhynchus, was studied. It was found to require a cerebellar architecture that could supply independent frequency channels of delayed feedback and multiple burst learning rules that could shape this feedback. This unifies two ideas about the function of the cerebellum that were previously separate: the cerebellum as an adaptive filter and as a generator of precise temporal inputs.

One of the most powerful and important applications that the brain accomplishes is solving the sensory "cocktail party problem:" to adaptively suppress extraneous signals in an environment. Theoretical studies suggest that the solution to the problem involves an adaptive filter, which learns to remove the redundant noise. However, neural learning is also in its infancy and there are still many questions about the stability and application of synaptic learning rules for neural computation. In this thesis, the implementation of an adaptive filter in the brain of a weakly electric fish, A. Leptorhynchus, was studied. It was found to require a cerebellar architecture that could supply independent frequency channels of delayed feedback and multiple burst learning rules that could shape this feedback. This unifies two ideas about the function of the cerebellum that were previously separate: the cerebellum as an adaptive filter and as a generator of precise temporal inputs.

Cette thèse concerne principalement la résolution des problèmes d’inversion dynamiquedans le cadre des systèmes dynamiques non linéaires. Ainsi, un ensemble de techniquesbasées sur l’utilisation des trains de mesures passées et sauvegardées sur une fenêtreglissante, a été développé. En premier lieu, les mesures sont utilisées pour générerune famille de signatures graphiques, qui constituent un outil de classification permettantde discriminer les diverses valeurs des variables à estimer pour un système non linéairedonné. Cette première technique a été appliquée à la résolution de deux problèmes : leproblème d’électolocation d’un robot doté du sens électrique et le problème d’estimationd’état dans les systèmes à dynamiques non linéaires. Outre ces deux applications, destechniques d’inversion à horizon glissant spécifiques au problème de diagnostic des défautsd’éoliennes dans le cadre d’un benchmark international ont été développées. Cestechniques sont basées sur la minimisation de critères quadratiques basés sur des modèlesde connaissance. / This thesis mainly concerns the resolution of dynamic inverse problems involvingnonlinear dynamical systems. A set of techniques based on the use of trains of pastmeasurements saved on a sliding window was developed. First, the measurements areused to generate a family of graphical signatures, which is a classification tool, in orderto discriminate between different values of variables to be estimated for a given nonlinearsystem. This technique was applied to solve two problems : the electrolocationproblem of a robot with electrical sense and the problem of state estimation in nonlineardynamical systems. Besides these two applications, receding horizon inversion techniquesdedicated to the fault diagnosis problem of a wind turbine proposed as an internationalbenchmark were developed. These techniques are based on the minimization of quadraticcriteria based on knowledge-based models.

One of the most powerful and important applications that the brain accomplishes is solving the sensory "cocktail party problem:" to adaptively suppress extraneous signals in an environment. Theoretical studies suggest that the solution to the problem involves an adaptive filter, which learns to remove the redundant noise. However, neural learning is also in its infancy and there are still many questions about the stability and application of synaptic learning rules for neural computation. In this thesis, the implementation of an adaptive filter in the brain of a weakly electric fish, A. Leptorhynchus, was studied. It was found to require a cerebellar architecture that could supply independent frequency channels of delayed feedback and multiple burst learning rules that could shape this feedback. This unifies two ideas about the function of the cerebellum that were previously separate: the cerebellum as an adaptive filter and as a generator of precise temporal inputs.