Entender como o cérebro processa informação é um dos problemas mais fascinantes da ciência de nossos dias. Para resolvê-lo, é fundamental estudarmos os mecanismos de representação e transmissão de informação de um único neurônio, a unidade fundamental de processamento do cérebro. Grande parte dos neurônios representa a informação por seqüências de pulsos elétricos, ou potenciais de ação. Nós usamos a mosca como modelo para estudar como a arborização dendrítica do neurônio influencia a quantidade de informação transmitida pela estrutura temporal da seqüência de pulsos. O mapeamento retinotópico da informação no sistema visual da mosca permite que as arborizações dendríticas de certos neurônios sejam estimuladas localmente através da região que corresponde a essa localização no campo visual. Nós apresentamos imagens em movimento em várias regiões do campo visual da mosca e medimos a resposta do neurônio H1, sensível a movimentos horizontais. Usando a teoria da informação, calculamos a quantidade de informação transmitida para cada uma dessas regiões do campo visual e a relacionamos com outras propriedades do neurônio, como por exemplo a sensibilidade espacial e eficiência. Nossos resultados sugerem que a arborização dendrítica influencia a codificação temporal de maneira significativa, indicando que o neurônio pode usar a estrutura temporal da sequência de pulsos para codificar outros parâmetros do estímulo, ou para aumentar a confiabilidade da codificação dependendo da região excitada / Understanding how the brain processes information about the outside world is one of the most fascinating problems of modern science. This involves the analysis of information representation and transmission in the fundamental processing element of the brain - the neuron. In the cortex neurons represent information by sequences of electrical pulses, or spikes. We use the fly as a model to study how the amount of information transmitted by the temporal structure of the spike trains depends on the neuron\'s dendritic arborization. The retinotopic mapping of information in the fly\'s visual system allows the stimulation of specific regions of the neuron\'s dendritic tree through the visual stimulation of the respective region in the visual field of the fly. We show an image moving in the preferred direction of the motion-sensitive neuron H1 in specific regions of the fly\'s visual field and measure the electrical response of the neuron. Using information theory, we calculate the amount of information transmitted for each of these regions and compare it with other properties of the neuron, for example, the spatial sensitivity. Our results suggest that the dendritic arborization influentes the temporal coding in a significant way, indicating that the neuron could use the temporal structure of the spike train to codify other parameters of the stimulus, or to increase the reliability of the code depending on the excited region

Os animais utilizam, entre outras coisas, a informação visual que chega na forma de padrões do fluxo óptico para se locomover, desviar de obstáculos, localizar um predador ou uma presa. Esta informação permite ao animal estimar seu próprio movimento e o movimento de outros objetos no campo visual. Nós usamos a mosca como modelo para estudar como um de seus neurônios, o H1 ? que é sensível a movimentos horizontais ?, codifica e diferencia o movimento da mosca ou seja, como ele diferencia os movimentos de rotação e translação. Assim como a grande parte dos neurônios, o H1 representa a informação por seqüências de pulsos elétricos, ou potenciais de ação. Queremos estudar principalmente a interação deste neurônio com o H1 contralateral e com as células CH e HS. Para isso pintamos o olho direito da mosca e acrescentamos um anteparo entre os monitores para isolarmos o H1 esquerdo e comparamos a resposta obtida desta forma com a obtida quando os dois H1 recebiam o fluxo óptico. Apresentamos imagens em movimento em dois monitores na região mais sensível do campo visual da mosca, sincronizadas de forma a simular movimentos de rotação e translação medindo sempre a resposta do neurônio H1 esquerdo. Acrescentamos também várias defasagens entre os estímulos apresentados e repetimos os experimentos para estímulos com tempos de correlação diferentes. Usando a teoria da informação, calculamos a quantidade de informação transmitida para cada um destes casos e analisamos como cada uma destas situações influencia não só a quantidade de informação transmitida pelo trem de pulsos mas a própria estrutura deste trem de pulsos. Nossos resultados sugerem que a interação entre os dois H1 diminui a quantidade de informação transmitida pela estrutura temporal da seqüência dos pulsos e que apesar de a defasagem não alterar a quantidade de informação total do trem de pulsos, ela altera a quantidade de informação transmitida pela estrutura de determinados eventos. / Animals use, among others things, the visual information arriving in the form of optic flow patterns to move itself, to deviate from obstacles, to locate a predator or a prey. This information allows the animal to estimate its own movement and the movement of other objects in the visual field. We use the fly as model to study how one of its neurons, the H1 - that is sensitive to horizontal movements -, codifies and differentiates the movement of the fly, i.e., how it distinguishes the movements of rotation and translation. The H1 represents the information as sequences of electric pulses, or action potentials. We want to study the interaction of this neuron with the contralateral H1 and with CH and HS cells. To do that we paint the right eye of the fly and add a screen between the monitors to isolate the left H1 and we compare the response obtained with the response, when both H1 received the optic flow. We show images moving in the most sensitive region of the fly\'s visual field synchronized so as to simulate rotational and translational movements and measure the left H1 response. We also add delays between the stimuli presented to each eye and repeat the experiments for stimuli with different correlation times. Using information theory, we calculate the amount of information transmitted for each one of these situations and analyze how they influence the amount of information transmitted by the spike train, as well as the structure of this spike train. Our results suggest that the interaction between both H1 reduces the amount of information transmitted by the time structure of the sequence of spikes and that, although the delay does not modify the amount of total information of the spike train, it modifies the amount of information transmitted by the structure of particular events.

Corujas se destacam por suas habilidades visuais e auditivas. Pouco é conhecido sobre a neuroanatomia do seu telencéfalo. Assim, caracterizamos através de técnicas histo/imunoistoquímicas o telencéfalo da coruja-da-igreja. Os núcleos da base foram delineados através da sua intensa imunomarcação para DARPP-32 e tirosina hidroxilase. Áreas sensoriais primárias tálamorrecipientes, como o entopálio (E), L2 do Field L auditório e o núcleo basorostral palial, foram caracterizadas pela quase ausência de DARPP-32 e alta atividade da citocromo oxidase (CO). As pseudo-camadas do Wulst visual foram delineadas com uma combinação de métodos, incluindo a ativação da CO, e imunomarcação para DARPP-32. O Wulst visual e o Field L se destacaram como regiões enormes, enquanto o E se revelou menor. Os dados sugerem que a morfologia de muitas regiões telencefálicas da coruja-da-igreja é semelhante àquela em outras aves. Contudo, o Wulst e o Field L se destacaram por seu tamanho e grau de organização, refletindo a importância do sistema visual e auditivo no comportamento de corujas. / Owls possess exceptional visual and auditory capacities. There is only limited information about the neuroanatomy of their forebrain. Thus, we characterized by histo/immunohistochemical techniques the forebrain of the barn owl. The basal ganglia were delineated by their intense immunostaining for DARPP-32 and tyrosine hydroxylase. Primary thalamorecipient sensory areas, such as the entopallium (E), L2 of the auditory Field L and the basorostral palial nucleus were characterized by the almost absence of DARPP-32 and their high citocrome oxidase (CO) activity. The pseudo layers of the visual Wulst were delineated by a combination of methods, including CO activity and immunostaining for DARPP-32.The Wulst and Field L were outlined by their huge size, whereas the E was small. These data suggest that the morphology of many telencephalic regions of the barn owl is similar to that in other birds. However, the Wulst and Field L were highlighted by their size and degree of organization, reflecting the importance of the visual and auditory system for the behavior of owls.

The problem of spatial disorientation during flight of aircraft is briefly described, as are definitions of the phenomenon. Traditional countermeasure efforts that are often directed towards changes in the central visual field are reconsidered in favour of presentation of information in the peripheral visual field. It is proposed to use optic flow to support spatial orientation, as well as to omit such information from the central visual field. An experiment was conducted, and results showed that forward visual flow gave very important spatial information. The flow could be cropped to a certain degree in the periphery (horizontally), as well as parts of the central presentation could be omitted without decreasing effects in the experiment. Implications relevant to possible implementations in aircraft are discussed.

This thesis aims to understand the learning mechanisms which underpin the process of visual object recognition in the primate ventral visual system. The computational crux of this problem lies in the ability to retain specificity to recognize particular objects or faces, while exhibiting generality across natural variations and distortions in the view (DiCarlo et al., 2012). In particular, the work presented is focussed on gaining insight into the processes through which transformation-invariant visual representations may develop in the primate ventral visual system. The primary motivation for this work is the belief that some of the fundamental mechanisms employed in the primate visual system may only be captured through modelling the individual action potentials of neurons and therefore, existing rate-coded models of this process constitute an inadequate level of description to fully understand the learning processes of visual object recognition. To this end, spiking neural network models are formulated and applied to the problem of learning transformation-invariant visual representations, using a spike-time dependent learning rule to adjust the synaptic efficacies between the neurons. The ways in which the existing rate-coded CT (Stringer et al., 2006) and Trace (Földiák, 1991) learning mechanisms may operate in a simple spiking neural network model are explored, and these findings are then applied to a more accurate model using realistic 3-D stimuli. Three mechanisms are then examined, through which a spiking neural network may solve the problem of learning separate transformation-invariant representations in scenes composed of multiple stimuli by temporally segmenting competing input representations. The spike-time dependent plasticity in the feed-forward connections is then shown to be able to exploit these input layer dynamics to form individual stimulus representations in the output layer. Finally, the work is evaluated and future directions of investigation are proposed.

La grande variété de morphologie de l’appareil visuel chez les arthropodes en fait un groupe unique pour l’étude de la diversité et l'évolution du système visuel. Cependant, la plupart de nos connaissances sur le développement et l'architecture neurale du système visuel provient de quelques organismes modèles. Mon projet vise à contribuer à l'étude de la diversité et de l'évolution du système visuel des arthropodes en étudiant l'œil du crustacé Parhyale hawaiensis; axé sur son développement, sa neuro-architecture et sa fonction. En particulier, mon travail vise à caractériser la structure du système visuel, à cartographier les connexions entre les photorécepteurs (PR) et le lobe optique (LO) et à comprendre les adaptations fonctionnelles de l'œil, par rapport aux yeux des autres arthropodes.Une description de l'anatomie de base du système visuel a été réalisée au moyen de la microscopie électronique, par immunomarquage et par la production de lignées de transgénique. J'ai trouvé que Parhyale possède un œil composé de type apposition avec 8 (chez les nouveau-nés) à 50 (chez les adultes) ommatidies, chacun formée par 5 PR (R1-R5). Nous avons trouvé deux opsines, nommés Ph-Opsin1 et Ph-Opsin2, exclusivement exprimés dans la rétine. En utilisant la séquence génomique comme guide, j'ai cloné des séquences régulatrices en amont de chaque gène d’opsine et généré des rapporteurs transgéniques qui récapitulent les patterns d'expression de Ph-Opsin1 et de Ph-Opsin2. Ces rapporteurs ont révélé que R1-R4 exprime Ph-Opsin1 tandis que R5 exprime le Ph-Opsin2.Immunomarquage ainsi que l'imagerie des deux lignées transgéniques, ont montré que les PR envoient de longues projections depuis la rétine au LO. Trois neuropiles optiques ont été identifiés: la lamina, la medulla et un neuropile plus profond qui est probablement la lobula plate ou la lobula. En suivant les projections axonales des PR dans le cerveau, révélant que tous les PR se projettent dans la lamina. Ceci diffère de ce qui a été montré chez les diptères et les crustacés, où au moins un PR par ommatidie projette ses axones dans la medulla.La microscopie électronique a montré que les rhabdomères des deux paires de PR, R1 + R3 et R2 + R4, sont orthogonalement alignés les uns aux autres dans chaque ommatidie, et que le rhabdome ne tourne pas. Ces caractéristiques rendent les PR intrinsèquement sensibles aux directions spécifiques de la lumière polarisée. Par conséquent, j'ai essayé de comprendre si Parhyale réagît à la lumière polarisée, au moyen d'expériences comportementales. Les données que j'ai recueillies suggèrent que Parhyale sont phototactiques pour la lumière blanche mais ne montrent aucune réponse à la lumière polarisée dans ces essais expérimentaux. Les problèmes potentiels liés à ces tests de comportement sont discutés.Enfin, je montre que l'œil de Parhyale s'adapte rapidement à différentes conditions d'intensité lumineuse. Ceci est obtenu par le mouvement des granules de pigments, situés à l'intérieur des PR, et par des changements morphologiques de la membrane basale du PR.Ce projet est pionnier dans l'étude du système visuel chez Parhyale. C'est la première fois que des outils génétiques ont été introduits pour étudier le système visuel de crustacés. Il établit Parhyale comme un puissant système expérimental pour des études in vivo de développement des yeux composé et de ciblage axonal du système visuel, un champ actuellement dominé par des études sur une seule espèce de mouche. / The wide diversity of eye designs present in arthropods makes them a unique group for studying the diversity and evolution of the visual system. However, most of our knowledge on the development and the neural architecture of the visual system comes from few model organisms. My project aims to contribute to the study of the diversity and evolution of the arthropod visual system by studying the eye of the crustacean Parhyale hawaiensis; focusing on its development, neuroarchitecture and function. In particular, my work aims to characterize the structure of the visual system, to map the connections between photoreceptors (PR) and optic lobe (OL) and to understand the functional adaptations of the eye, in relation to the eyes of other arthropods.A description of the basic anatomy of the visual system was performed by means of electron microscopy, immunostainings and by generating transgenic reporter lines. I found that Parhyale has an apposition-type compound eye with 8 (in hatchlings) to 50 (in adults) ommatidia, each one formed by 5 PR cells (R1-R5).Two opsins were found in Parhyale, named Ph-Opsin1 and Ph-Opsin2, which are exclusively expressed in the retina. Using the genome sequence as a guide, I cloned upstream regulatory sequences from each opsin genes and generated transgenic reporters that recapitulate the expression patterns of Ph-Opsin1 and Ph-Opsin2. These reporters revealed that R1-R4 express Ph-Opsin1 while R5 expresses Ph-Opsin2.Immunostainings and live imaging of the two transgenic lines showed that PR cells send long projections from the retina to the OL, via an optic nerve. Three optic neuropils were identified: lamina, medulla and a deeper neuropil, possibly the lobula or lobula plate. Following the axonal projections of the PR into the brain, revealed that all PR project to the lamina. This differs from what has been shown in dipterans and crustaceans, where at least one PR per ommatidium projects to the medulla. Electron microscopy showed that the rhabdomeres of two pairs of PR, R1+R3 and R2+R4, are orthogonally aligned to each other in each ommatidium, and that the rhabdom does not rotate. These features render the PR intrinsically sensitive to specific directions of light polarisation. Therefore, I tried to understand whether and how Parhyale respond to polarised light. I developed two experimental setups to address whether Parhyale shows behavioural responses triggered by light polarisation. The data I have collected suggest that Parhyale are phototactic to dim white light but show no response to polarised light in these specific experimental assays. Potential problems with these behavioural assays are discussed.Finally I show that the eye of Parhyale quickly adapts to different conditions of light intensity. This is achieved by movement of the shielding pigment granules, located inside the PR cells and by morphological changes of the PR basal membrane.This project is pioneering the study of the visual system in Parhyale. It is the first time that genetic tools have been introduced to study the crustacean visual system. It establishes Parhyale as a powerful experimental system for in vivo studies of compound eye development and axonal targeting, a field currently dominated by studies in a single species of fruitfly.

Understanding how individual photoreceptor cells factor in the spectral sensitivity of a visual system is essential to explain how they contribute to the visual ecology of the animal in question. Existing methods that model the absorption of visual pigments use templates which correspond closely to data from thin cross-sections of photoreceptor cells. However, few modeling approaches use a single framework to incorporate physical parameters of real photoreceptors, which can be fused, and can form vertical tiers. Akaike’s information criterion (AIC c ) was used here to select absorptance models of multiple classes of photoreceptor cells that maximize information, given visual system spectral sensitivity data obtained using extracellular electroretinograms and structural parameters obtained by histological methods. This framework was first used to select among alternative hypotheses of photoreceptor number. It identified spectral classes from a range of dark-adapted visual systems which have between one and four spectral photoreceptor classes. These were the velvet worm, Principapillatus hitoyensis , the branchiopod water flea, Daphnia magna , normal humans, and humans with enhanced S-cone syndrome, a condition in which S-cone frequency is increased due to mutations in a transcription factor that controls photoreceptor expression. Data from the Asian swallowtail, Papilio xuthus , which has at least five main spectral photoreceptor classes in its compound eyes, were included to illustrate potential effects of model over-simplification on multi-model inference. The multi-model framework was then used with parameters of spectral photoreceptor classes and the structural photoreceptor array kept constant. The goal was to map relative opsin expression to visual pigment concentration. It identified relative opsin expression differences for two populations of the bluefin killifish, Lucania goodei . The modeling approach presented here will be useful in selecting the most likely alternative hypotheses of opsin-based spectral photoreceptor classes, using relative opsin expression and extracellular electroretinography.

Please read the abstract in the section 00front of this document Copyright 1990, University of Pretoria. All rights reserved. The copyright in this work vests in the University of Pretoria. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of the University of Pretoria. Please cite as follows: Barnard, G 1990, High quality coding and reconstruction for transmission of single video images, MEng dissertation, University of Pretoria, Pretoria, viewed yymmdd < http://upetd.up.ac.za/thesis/available/etd-10312007-110001/ > / Dissertation (M Eng (Electronic Engineering))--University of Pretoria, 2007. / Electrical, Electronic and Computer Engineering / unrestricted

京都大学 / 新制・課程博士 / 博士(生命科学) / 甲第23340号 / 生博第458号 / 新制||生||61(附属図書館) / 京都大学大学院生命科学研究科高次生命科学専攻 / (主査)教授 松田 道行, 教授 見学 美根子, 教授 今吉 格 / 学位規則第4条第1項該当 / Doctor of Philosophy in Life Sciences / Kyoto University / DFAM

No description available.