The work presented in this dissertation focuses on the use of computational and mathematical models to investigate how mammalian brains construct and maintain stable representations of space and location. Recordings of the activities of cells in the hippocampus and entorhinal cortex have provided strong, direct evidence that these cells and brain areas are involved in generating internal representations of the location of an animal in space. The emphasis of the first two portions of the dissertation are on understanding the factors that influence the scale and stability of these representations, both of which are important for accurate spatial navigation. In addition, it is argued in both cases that many of the computations observed in these systems emerge at least in part as a consequence of a particular type of network structure, where excitatory neurons are driven by external sources, and then mutually inhibit each other via interactions mediated by inhibitory cells. The first contribution of this thesis, which is described in chapter 2, is an investigation into the origin of the change in the scale of spatial representations across the dorsoventral axis of the hippocampus. Here it will be argued that this change in scale is due to increased processing of nonspatial information, rather than a dorsoventral change in the scale of the spatially-modulated inputs to this structure. Chapter 3 explores the factors influencing the dynamical stability of class of pattern-forming networks known as continuous attractor networks, which have been used to model various components of the spatial navigation systems, including head direction cells, place cells, and grid cells. Here it will be shown that network architecture, the amount of input drive, and the timescales at which cells interact all influence the stability of the patterns formed by these networks. Finally, in chapter 4, a new technique for analyzing neural data is introduced. This technique is a spike train similarity measure designed to compare spike trains on the basis of shared inhibition and bursts.

Les organismes montrent des variations de cycle de vie, de comportement et de physiologie. L'écologie évolutive cherche à les expliquer en investiguant les conséquences de ces variations pour l'écologie et l'aptitude phénotypique des animaux. Ainsi, certaines populations comportent des individus ayant un cycle de vie plus rapide que d'autres, caractérisé par une reproduction plus précoce, une fécondité plus élevée mais également une longévité plus faible. Une hypothèse récente, l'hypothèse du train de vie, prédit que ces différences de cycle de vie seraient associées à des différences de comportement constantes entre les individus lorsque ces traits sont impliqués dans les compromis évolutifs. Selon cette hypothèse, les individus investissant plus dans les comportements exploratoires devraient ainsi atteindre leur succès reproducteur maximum plus tard dans leur vie. Ceci peut provenir, d'une part, du fait que le train de vie des individus détermine leur personnalité; ou d'autre part, que la personnalité d'un individu détermine comment il complétera son cycle de vie. Il est possible de gagner certains indices sur les relations entre les traits comportementaux et biodémographiques à travers l'analyse des mécanismes de régulation qui leur sont commun, comme les gluco-corticoïdes (GC). Le but général de cette thèse de doctorat est de mieux comprendre les différences de patron d'exploration en testant l'hypothèse du train de vie. Selon cette hypothèse, les individus investissant le plus dans l'exploration de leur environnement devraient montrer un succès reproducteur maximal plus tardif et mobiliser plus de GC pour protéger leur survie face aux perturbations environnementales. J'analyse ainsi les relations entre le patron d'exploration exprimé par les individus dans un environnement nouveau, leur patron de reproduction au cours de leur vie, et le niveau de cortisol (le GC principal chez les mammifères) dans une population naturelle de tamias rayés (Tamias striatus) située au sud du Québec dans les monts Sutton. Je valide dans un premier temps une méthode permettant le suivi du niveau de cortisol de manière non-invasive. Je documente ensuite les relations entre le patron d'exploration des individus, leur docilité lors de manipulations sur le terrain et leur propension à être capturés dans les pièges. Je détermine la relation entre le patron d'exploration des individus et leur réactivité physiologique aux perturbations environnementales. Je documente ensuite le patron de reproduction des individus tout au long de leur vie et analyse sa relation avec leur patron d'exploration. Enfin, j'analyse le niveau de cortisol des individus en nature. Les individus ont été suivis sur 25 hectares de forêt décidue mature. Chaque été de 2005 à 2010 les individus ont été suivis par piégeage, observations focales et télémétrie. Les jeunes ont été capturés à l'émergence du terrier maternel, et assignés à leur père le plus probable à l'aide de marqueurs microsatellites. Les individus ont été soumis à des tests d'arène, quantifiant la réponse comportementale à un environnement nouveau. Durant ce test, certains individus expriment un patron d'exploration superficiel, caractérisé par une exploration importante en début de test mais une diminution tout aussi importante de l'exploration au court des secondes suivantes. À l'autre extrême, d'autres individus expriment un patron d'exploration plus méticuleux, avec un niveau d'exploration modéré mais constant au cours de la durée du test. Les animaux ont également été soumis à des tests de docilité, lors desquels nous avons relevé le nombre de secondes passées par l'animal à se débattre durant une minute lors des manipulations. Lors des captures en 2009, des échantillons fécaux ont été prélevés dans les trappes. La concentration de métabolites issus du cortisol a ensuite été analysée par ELISA compétitive en utilisant un anticorps polyclonal. La réactivité du système sympathique des individus a été quantifiée par l'analyse du rythme cardiaque lors d'un test de restriction. Les individus de la population d'étude montrant un patron d'exploration plus superficiel dans l'arène sont moins dociles. Ces individus sont également capturés plus fréquemment (mâles) ou plus loin de leur terrier (femelles). Les individus exprimant un patron d'exploration superficiel dans l'arène démontrent une réactivité du système sympathique plus importante, mais une variabilité plus faible de leur niveau de cortisol, suggérant qu'ils protègent ainsi leur survie de manière moins importante. En accord avec les prédictions, les individus plus superficiels dans leur exploration atteignent également leur succès reproducteur maximal plus tôt au cours de leur vie reproductive. Les variations d'abondance de nourriture d'une année à l'autre, à travers leurs effets sur l'âge à la première reproduction des individus contribuent à favoriser un meilleur succès reproducteur à vie des explorateurs superficiels parmi les individus nés durant les années de forte abondance de nourriture. À l'inverse, les explorateurs méticuleux ont un succès reproducteur plus élevé parmi les individus nés durant les années de faible abondance de nourriture. Les femelles ayant un patron d'exploration plus superficiel démontrent une variabilité plus faible de leur niveau de cortisol. Les femelles amenant un plus grand nombre de jeunes au sevrage ont également une variabilité en cortisol plus faible au cours de l'été. Ces résultats sont en accord avec l'hypothèse du train de vie et démontrent que le patron d'exploration exprimé par les animaux est associé à leur patron de reproduction en fonction de l'âge. De plus ils suggèrent que les fluctuations d'abondance des ressources d'une année à l'autre sont susceptibles de générer des pressions de sélection oscillantes qui pourraient maintenir la variabilité du patron d'exploration observée dans cette population. L'analyse du patron de cortisol des individus suggère enfin que le patron de reproduction et le patron d'exploration sont susceptibles de s'influencer l'un l'autre. Ces résultats contribuent à la compréhension des différences de personnalité par des résultats empiriques sur les relations entre les traits de personnalité, la physiologie et la biodémographie des individus. De telles études sont encore rares. Des analyses futures, détaillant d'une part la nature des coûts et des bénéfices associés au patron d'exploration des individus, et d'autre part investiguant les sources de variation contribuant à la variabilité du patron d'exploration observées dans cette population permettront de compléter les résultats présentés dans cette thèse. ______________________________________________________________________________

Neurons perform complex computations, communications and precise transmissions of information in the form of action potentials (APs). The high level of heterogeneity and complexity at all levels of organization within a neuron and the functional requirement of highly permeable cell membranes leave neurons exposed to damage when energy levels are insufficient for the active maintenance of ionic gradients. When energy is limiting the ionic gradient across a neuron’s cell membrane risks being dissipated which can have dire consequences. Other researchers have advocated “generalized channel arrest” and/or “spike arrest” as a means of reducing the neuronal permeability allowing neurons to adjust the demands placed on their electrogenic pumps to lower levels of energy supply. I investigated the consequences of hypoxia on the propagation of a train of APs down the length of a fast conducting axon capable of transmitting APs at very high frequencies. Under normoxic conditions I found that APs show conduction velocities and instantaneous frequencies nearly double that of neurons experiencing energy limiting hypoxic conditions. I show that hypoxia affects AP conduction differently for different lengths of axon and for APs of different instantaneous frequencies. Action potentials of high instantaneous frequency in branching lengths of axon within ganglia were delayed more significantly than those in non-branching lengths contained within the connective and fail preferentially in branching axon. I found that octopamine attenuates the effects of hypoxia on AP propagation for the branching length of axon but has no effect on the non-branching length of axon. Additionally, for energetically stable cells, application of the anti-diabetic medication metformin or the hyperpolarization-activated cyclic nucleotide-gated (HCN) channel blocker ZD7288 resulted in a reduced performance similar to that seen in neurons experiencing energetic stress. Furthermore both metformin and ZD7288 affect the shape of individual APs within an AP train as well as the original temporal sequence of the AP train, which encodes behaviourally relevant information. I propose that the reduced performance observed in an energetically compromised cell represents an adaptive mechanism employed by neurons in order to maintain the integrity of their highly heterogeneous and complex organization during periods of reduced energy supply. / Thesis (Master, Biology) -- Queen's University, 2011-10-07 14:41:46.972

No description available.

Wireless sensor network (WSN) has become a hot topic lately. By using WSN things that previously were difficult or impossible to measure has now become available. One of the main reasons using WSN for monitoring is to save money by cost optimization and/or increase safety by letting the user knowing the physical status of the monitored structure. This thesis considers four main topics, empirical testing of WSN in harsh environments, antenna designs, antenna measurements and radio environment emulation. The WSN has been tested in train environment for monitoring of ball bearings and inside jet engines to monitor strain of blades and temperatures. In total, two investigations have been performed aboard the train wagon and one in the jet engine. The trials have been successful and provide knowledge of the difficulties with practical WSN applications. The key issues for WSN are robust communication, energy management (including scavenging) and physical robustness. For the applications of WSN in harsh environments antennas has to be designed. In the thesis, two antennas has been designed, one for train environment and one for the receiver in the jet engine. In the train environment, a more isotropic radiation pattern is preferable; hence a small dual layered patch antenna is designed. The antenna is at the limit of being electrically small; hence slightly lower radiation efficiency is measured. For the WSN in the jet engine, a directive patch array is designed on an ultra-thin and flexible substrate. The thin substrate of the antenna causes rather lower radiation efficiency. But the antenna fulfils the requirements of being conformal and directive. In reverberation chambers are used to measure antennas, but there are difficulties to provide a realistic radio environment, for example outdoor or on-body. In this thesis, a large reverberation chamber is designed and verified. It enables measurement between 400 MHz and 3 GHz. Also, a sample selection method is designed to provide a post processing possibilities to emulate the radio environment inside the chamber. The method is to select samples from a data set that corresponds to a desired probability density function. The method presented in this thesis is extremely fast but the implementation of the method is left for future research. / WISENET / WiseJet

国立情報学研究所で電子化したコンテンツを使用している。

Rail track operators have traditionally used manual methods to construct train timetables. Creating a timetable can take several weeks, and so the process usually stops once the first feasible timetable has been found. It is suspected that this timetable is often far from optimal. Existing methods schedule track maintenance once the best train timetable has been determined and allow little or no adjustments to the timetable. This approach almost certainly produces suboptimal integrated solutions since the track maintenance schedule is developed with the imposition of the previously constructed train timetable. The research in this thesis considers operationally feasible methods to produce integrated train timetables and track maintenance schedules so that, when evaluated according to key performance criteria, the overall schedule is the best possible. This research was carried out as part of the Cooperative Research Centre for Railway Engineering and Technologies. We developed a method that uses a local search meta-heuristic called 'problem space search'. A fast dispatch heuristic repeatedly selects and moves a track possessor (train or maintenance task) through the network; this results in a single integrated schedule. This technique generates a collection of alternative feasible schedules by applying the dispatch heuristic to different sets of randomly perturbed data. The quality of the schedules is then evaluated. Thousands of feasible solutions can be found within minutes. We also formulated an integer programming model that selects a path for each train and maintenance task from a set of alternatives. If all possible paths are considered, then the best schedule found is guaranteed to be optimal. To reduce the size of the model, we explored a reduction technique called 'branch and price'. The method works on small example problems where paths are selected from a predetermined set, but the computation time and memory requirements mean that the method is not suitable for realistic problems. The main advantages of the problem space search method are generality and speed. We are able to model the operations of a variety of rail networks due to the representation of the problem. The generated schedules can be ranked with a user-defined objective measure. The speed at which we produce a range of feasible integrated schedules allows the method to be used in an operational setting, both to create schedules and to test different scenarios. A comparison with simulated current practice on a range of test data sets reveals improvements in total delay of up to 22%.

Rail track operators have traditionally used manual methods to construct train timetables. Creating a timetable can take several weeks, and so the process usually stops once the first feasible timetable has been found. It is suspected that this timetable is often far from optimal. Existing methods schedule track maintenance once the best train timetable has been determined and allow little or no adjustments to the timetable. This approach almost certainly produces suboptimal integrated solutions since the track maintenance schedule is developed with the imposition of the previously constructed train timetable. The research in this thesis considers operationally feasible methods to produce integrated train timetables and track maintenance schedules so that, when evaluated according to key performance criteria, the overall schedule is the best possible. This research was carried out as part of the Cooperative Research Centre for Railway Engineering and Technologies. We developed a method that uses a local search meta-heuristic called 'problem space search'. A fast dispatch heuristic repeatedly selects and moves a track possessor (train or maintenance task) through the network; this results in a single integrated schedule. This technique generates a collection of alternative feasible schedules by applying the dispatch heuristic to different sets of randomly perturbed data. The quality of the schedules is then evaluated. Thousands of feasible solutions can be found within minutes. We also formulated an integer programming model that selects a path for each train and maintenance task from a set of alternatives. If all possible paths are considered, then the best schedule found is guaranteed to be optimal. To reduce the size of the model, we explored a reduction technique called 'branch and price'. The method works on small example problems where paths are selected from a predetermined set, but the computation time and memory requirements mean that the method is not suitable for realistic problems. The main advantages of the problem space search method are generality and speed. We are able to model the operations of a variety of rail networks due to the representation of the problem. The generated schedules can be ranked with a user-defined objective measure. The speed at which we produce a range of feasible integrated schedules allows the method to be used in an operational setting, both to create schedules and to test different scenarios. A comparison with simulated current practice on a range of test data sets reveals improvements in total delay of up to 22%.

Thèses (M.Sc.)--Université de Sherbrooke (Canada), 1997. / Titre de l'écran-titre (visionné le 20 juin 2006). Publié aussi en version papier.

by Alexander H. Levis and Michael Athans. / Bibliography: p. 27. / PB 177 669. / Prpared for the U.S. Dept. of Transportation under Contract C-85-65, DSR Project no. 76105.