Adaptation is one of the basic phenomena of biology, while adaptability is an important feature for neural network. Young barn owl can well adapt its visual and auditory integration to the environmental change, such as prism wearing. At first, a mathematical model is introduced by the related study in biological experiment. The model well explained the mechanism of the sensory map realignment through axongenesis and synaptogenesis. Simulation results of this model are consistent with the biological data. Thereafter, to test the model’s application in hardware, the model is implemented into a robot. Visual and auditory signals are acquired by the sensors of the robot and transferred back to PC through bluetooth. Results of the robot experiment are presented, which shows the SC model allowing the robot to adjust visual and auditory integration to counteract the effects of a prism. Finally, based on the model, a silicon Superior Colliculus is designed in VLSI circuit and fabricated. Performance of the fabricated chip has shown the synaptogenesis and axogenesis can be emulated in VLSI circuit. The circuit of neural model provides a new method to update signals and reconfigure the switch network (the chip has an automatic reconfigurable network which is used to correct the disparity between signals). The chip is also the first Superior Colliculus VLSI circuit to emulate the sensory map realignment.

Intercepter une cible en mouvement est un défi spatiotemporel relevé par de nombreuses espèces animales. Ici, nous nous sommes intéressés à la capture fovéale lors de l’orientation saccadique du regard vers une cible visuelle mobile chez le singe vigile. La théorie actuelle propose que l’interception saccadique fasse intervenir deux voies neuronales. Une première voie acheminerait, via le colliculus supérieur profond (CSp), un signal de position échantillonnée au générateur saccadique. La seconde, via le cervelet, fournirait une commande supplémentaire sur la base des signaux de mouvement visuel. Une étude comportementale a été réalisée afin d’analyser la dynamique de l’influence de ces signaux sur la trajectoire saccadique et a permis de mettre en évidence une influence continue. Dans une seconde étude, nous avons testé la robustesse du système oculomoteur à une perturbation spatiotemporelle inattendue (par microstimulation électrique du CSp) et montré la présence de saccades de correction précises. L’ensemble de nos résultats plaide pour une représentation continue du but des saccades d’interception. / Intercepting a visual moving target is a spatiotemporal challenge for the brain achieved by various species. Here, we investigated the foveal capture of a moving target by saccadic gaze shifts in the awake monkey. The current theory proposes that the saccadic interception involves two neural pathways. A first pathway would convey to the saccade burst generator a sampled target position signal through the superior colliculus (SC). The second one, through the cerebellum, would convey an additional command on the basis of motion-related signals. A behavioral experiment was performed to analyze the influence of motion-related signals on the saccade dynamics and allowed showing a continuous influence. In a second study, we tested the robustness of the oculomotor system to an unexpected spatiotemporal perturbation (by electrical microstimulation in the deep SC) and showed the presence of accurate correction saccades. Our results argue for a continuous representation of the saccade goal.

Le Colliculus Supérieur (SC) est bien connu pour son rôle dans la génération des saccades oculaires. Sa connectivité réciproque avec les Ganglions de la Base (BG) a récemment été mise en évidence, ainsi que son implication active dans les processus de sélection saccadique. Indépendamment du rôle des BG dans la sélection en général, les hypothèses traditionnelles suggèrent que la sélection au sein du SC résulte d'inhibitions latérales réciproques. Notre modèle propose de développer une autre hypothèse, dans laquelle les BG jouent un rôle important pour la sélection des cibles des saccades par un circuit purement sous-cortical SC-BG. En partant du constat que les profils d'activités des populations neuronales du SC peuvent être assimilés à ceux des neurones accumulateurs stochastiques des modèles phénoménologiques de sélection par course, ce nouveau modèle propose que la sélection observée dans le SC résulte non pas d'interactions latérales dans le SC, mais d'un processus de course vers un seuil de sélection dans les couches colliculaires intermédiaires, seuil dynamiquement déterminé par une désinhibition sélective opérée par les BG dans le cadre d'une boucle de rétrocontrôle double des BG vers le SC. Ce modèle reproduit divers profils d'activité neuronaux observés in-vivo, ainsi que les profils de sélection de plusieurs tâches expérimentales relatives à la discrimination entre de nombreux stimuli similaires. Son neuromimétisme lui permet de proposer diverses prédictions sur les substrats neurologiques de ses composants, et les substrats neurologiques de divers phénomènes particuliers de sélection comme les saccades moyennes ou le remote distractor effect. / The Superior Colliculus (SC) is well-known for its role in the generation of ocular saccades. Its reciprocal connectivity with the Basal Ganglia (BG) has recently been highligted, as well as its active involvement in the saccadic selection processes. Yet, the most common hypothesis explaining its role in saccade target selection neglect the potential implication of the BG and focus on unproven networks of lateral reciprocal inhibitions within the SC. We propose a model based on a the SC-BG loop hypothesis, in the framework of a purely subcortical saccade selection process Since the activity profiles of specific SC neurons populations can be seen as equivalent to the stochastic accumulators of the phenomelogical race models, we propose that the selection operated bythe SC is based not on lateral inhibitions within its maps, but on a race to a selection threshold in the intermediate layers of the SC, the threshold being dynamically set by the selective disinhibition of the SC maps by a double closed loops with the BG. Our model is able to reproduce various in-vivo neuronal activity profiles, as well as the selection distributions observed in various experimental setups involving the discrimination between numerous identical stimuli. Our model's neuromimetism level allows us to propose predictions about the neuronal substrates of the model's components, as well as the substrates for various selection phenomenons such as average saccades and the remote distractor effect.

A circuitaria neuro-anatômica envolvida com a organização de respostas de medo inclui muitos sítios subcorticais diferentes, dois dos quais são a região dorsal da matéria cinzenta periaquedutal (dPAG) e as camadas intermediárias e profundas do colículo superior (i/dlSC). Este estudo investigou como as células destas duas regiões mudam de atividade em um rato que foi colocado diante de seu predador, o gato, comparado com outras situações onde o risco menor. 11 ratos wistar machos passaram por cirurgia para implantação de um microdriver contendo 8 tetrodos, que detectou 39 células na dPAG de 7 ratos e 44 células nas i/dlSC de 4 ratos. A atividade das células foi registrada em uma condição de base e em quatro condições experimentais distintas: o rato confinado, o rato diante de uma novidade, diante de um predador e diante do contexto predatório. Os dados coletados foram analisados por scripts de MATLAB e cada célula foi classificada para sua responsividade às condições experimentais, à mudança do comportamento do rato, à velocidade e à posição do rato no aparato. Na dPAG as células se mostraram mais responsivas a condição do gato do que as demais condições experimentais, tanto em número de células como em aumento de atividade celular. Uma quantidade menor de células mostrou-se responsiva a mudanças de comportamento, particularmente os comportamentos de defesa. Nas i/dlSC as células também se mostraram mais responsivas para o gato, seguido do contexto, em ambos os casos com diminuição de atividade celular, que foi mais intensa no gato. Uma quantidade menor de células pareceu responder de forma similar para as diferentes mudanças comportamentais. Interpretamos esses dados propondo que a dPAG possui uma função sinalizadora de medo, fortemente ativada durante o gato, e que as i/dlSC são inibidas diante do gato a fim de evitar a captação de estímulos irrelevantes para lidar com a situação de risco. / The neuroanatomical circuitry involved with the organization of fear responses includes a great number of subcortical sites, two of which are the dorsal periaqueductal gray (dPAG) and the intermediate and deep layers of the superior colliculus (i/dlSC). This study investigated how the cellular activity in these regions changed in a rat that was exposed to one of its predators, a cat, when compared with lower risk situations. 11 male wistar rats underwent surgery to implant a microdriver containing 8 tetrodes, which detected 39 cells in the dPAG of 7 rats and 44 cells in the i/dlSC of 4 rats. Cell activity was registered in a basal condition and then again in for experimental conditions: the rat confined in a smaller space, the rat facing a novel stimulus, the rat facing a cat and the rat facing the predatory context. Collected data was analyzed using MATLAB and each cell was classified according to its responsivity to the experimental conditions, to the rats switches in behavior, to the rats speed and position in the experimental apparatus. In the dPAG, cells were shown to be more responsive to the cat condition over the other conditions, both in terms of number of responsive cells and intensity of increase in firing activity. A smaller number of cells were responsive to behavior switching, being especially sensitive to the initiation of defensive behaviors. In the i/dlSC, cells were more responsive to the cat, followed by the predatory context, in both cases with a general reduction of cellular activity that was more intensive for the cat. A smaller number of cells was responsive in similar ways for behavior switching across experimental conditions. We interpret this data by proposing that the dPAG acts signaling a state of fear, particularly during a high risk situation, and the i/dlSC are inhibited during riskier situations so as to block irrelevant environmental stimuli.

Das Ziel dieser Arbeit war es, die Modulation der GABAergen synaptischen Transmission in den visuellen Schichten des Colliculus superior der Maus durch das Neurotrophin BDNF zu charakterisieren. Hierzu wurden bdnf+/+ und -/- Mäuse kurz vor und nach der Augenöffnung in einer die Morphologie erhaltenden Schnittpräparation elektrophysiologisch und molekularbiologisch untersucht. Das Fehlen von BDNF veränderte das Präparat hinsichtlich Neurondichte und -größe nicht. Ebenso blieben der Membranwiderstand und die Ganzzellkapazität unbeeinflusst von der chronischen Abwesenheit von BDNF. Im Gegensatz dazu zeigten sich deutliche funktionelle Defizite im Entladungsverhalten und in der GABAergen Hemmung. Durch Registrierung von Aktionspotentialen wurde demonstriert, dass BDNF für die Aufrechterhaltung der Netzwerkaktivität erforderlich ist. Durch Applikation eines GABAA-Rezeptor-Inhibitors konnte die Suppression der GABAergen Hemmung durch BDNF als zugrunde liegender Mechanismus aufgedeckt werden. Daraufhin durchgeführte Ganzzellableitungen bestätigten dies und legten einen postsynaptischen, TrkB-vermittelten Mechanismus der BDNF-Wirkung nahe. Es war möglich, den Einfluss der chronischen Abwesenheit von BDNF durch akute lokale Superfusion von BDNF vollständig aufzuheben. Die ausschließlich postsynaptische Blockade der PKC reichte aus, dies zu verhindern. Hierdurch wird unterstrichen, dass in diesem Präparat der Angriffspunkt von BDNF an der GABAergen Synapse auf der postsynaptischen Seite liegt. Um den genauen Wirkungsmechanismus von BDNF an der GABAergen Synapse zu beleuchten, wurde die mRNA-Expression der GABAA-Rezeptor-Untereinheiten alpha 1-3 untersucht. Diese ist in Anwesenheit von BDNF höher. Demzufolge sollte eine reduzierte Expression dieser Untereinheiten in bdnf-/- Tieren zu einer verringerten Rezeptoranzahl und somit zur Sättigung postsynaptischer Rezeptoren führen. Durch die Analyse von Amplitude und Kinetik GABAerger IPSC und die Applikation von Zolpidem wurde dies bestätigt. Demnach führt die Abwesenheit von BDNF zur Aufregulation der GABAergen Inhibition, obwohl die Rezeptorzahl in der Postsynapse wahrscheinlich niedriger ist. Als zentraler Mechanismus der akuten BDNF-Wirkung kommt deshalb am ehesten eine PKC-vermittelte Phosphorylierung und nachfolgende Veränderung des Desensitisierungsverhaltens in Betracht. Außerdem muss an eine Reduktion der Öffnungswahrscheinlichkeit oder der Leitfähigkeit des Rezeptors gedacht werden. Unmittelbar vor der Augenöffnung hatte die BDNF-Defizienz keinerlei Einfluss auf die GABAerge Hemmung. Es ist also davon auszugehen, dass BDNF im CS erst nach der Augenöffnung eine wesentliche Rolle in der Modulation der GABAergen Synapsen und damit in der Kontrolle der Netzwerkaktivität spielt. Dies steht in Einklang mit der Vorstellung, dass sowohl Translation und Freisetzung als auch Transport von BDNF durch neuronale Aktivität reguliert werden. Der Ablauf der neurotrophinergen Regulation im Colliculus superior stellt sich wie folgt dar: Durch Aktivation colliculärer Afferenzen wird BDNF vermehrt freigesetzt. BDNF reduziert nun zunächst über die Modulation von Rezeptoreigenschaften die GABAerge Hemmung und disinhibiert die Netzwerkaktivität. Längerfristig kommt es über eine vermehrte Expression von GABAA-Rezeptor-Untereinheiten zum Anstieg der Rezeptorzahl, damit zur Wiederherstellung der GABAergen Hemmung und letztlich zu einer Reduktion der Netzwerkaktivität. BDNF ist also in der kritischen Zeitperiode der Augenöffnung, wenn das Mustersehen einsetzt, ein wichtiger Faktor in der Regulation neuronaler Aktivität. / The aim of the study was to characterise the influence of the neurotrophin BDNF on the GABAergic synaptic transmission in the visual layers of the mouse superior colliculus. Acute slices prepared from bdnf+/+ and -/- mice shortly before and after eye opening were employed in the experiments. The absence of BDNF altered neither the density or size of neurons nor their membrane resistance or whole cell capacity. However, registration of action potentials revealed a decreased firing rate in the absence of BDNF. Stronger disinhibition induced by application of a GABAA receptor blocker suggested an enhanced GABAergic inhibition as an underlying mechanism. This assumption was confirmed by performing whole cell experiments. Further analysis indicated a postsynaptic enhancement of GABAergic synaptic transmission in the absence of BDNF. In bdnf+/+ slices, blockade of BDNF signalling through the TrkB receptor strengthened GABAergic synaptic transmission. Contrariwise, superfusion of exogenous BDNF in bdnf-/- suppressed GABAergic synaptic transmission. An exclusively postsynaptic block of the PKC abolished the effect of BDNF application. Therefore, a BDNF induced, TrkB mediated, PKC dependent suppression of the GABAergic synaptic transmission at the postsynaptic site can be assumed. To further elucidate the mechanism of BDNF action the expression of the GABAA receptor subunits alpha 1-3 mRNA was studied. In the presence of BDNF an elevated expression was observed. A lower expression of these subunits in bdnf-/- slices could result in a reduced number of GABAA receptors. During synaptic transmission they may become saturated. Two observations support this idea: In bdnf-/- slices application of Zolpidem does not induce an increase of GABAergic IPSC amplitude as present in bdnf+/+ slices. Amplitude and decay kinetics of GABAergic IPSCs correlate in bdnf-/- but not +/+ slices. Therefore, absence of BDNF may strengthen GABAergic synaptic transmission although mediated by a reduced number of postsynaptic receptors. A BDNF induced PKC dependent receptor phosphorylation followed by a change in receptor desensitisation is most likely the underlying mechanism. Nevertheless, a reduction of the receptor opening probability or a reduced receptor conductance have to be considered as well. Shortly before eye opening BDNF deficiency had no impact on GABAergic inhibition. In accordance with the general idea that expression and release of BDNF is activity dependent, this finding suggests that BDNF controls network activity by modulating GABAergic synaptic transmission only after eye opening. Taken together, the following sequence of BDNF action in the superior colliculus during eye opening can be proposed: Activation of collicular inputs triggers an increased BDNF release. BDNF suppresses GABAergic synaptic transmission leading to a disinhibition of network activity. Later, the increased expression of GABAA receptor subunits by prolonged BDNF release results in an increased GABAA receptor number, the recovery of the GABAergic inhibition and finally a gently tuned network activity. In summary, BDNF constitutes a major regulator of neuronal activity in the critical period of eye opening marking the transition to pattern vision.

Damage to visual pathways can lead to irreversible blindness. Posterior visual pathways, located within a watershed area, are predisposed to hypoperfusion/reperfusion injury. In a novel rat model of bilateral common carotid artery occlusion (BCCAO), oxidative injury to the superior colliculus (SC), a major visual center within the watershed area was evaluated, in addition to its effects on retinal ganglion cells (RGCs). Nitrotyrosine, a footprint of peroxynitrite-mediated oxidative injury in the SC, and microtubule-associated protein 2, a dendrite marker in the retina, were assessed using immunofluorescence and confocal microscopy. Nitrotyrosine-immunoreactivity in the SC was increased 2 weeks after BCCAO compared to controls. Microtubule-associated protein 2-immunoreactivity in the central inner plexiform layer was reduced 3 weeks after BCCAO compared to controls. Global incomplete cerebral hypoperfusion/reperfusion induced oxidative injury in the SC and retrograde RGC dendritic changes. This suggests that cerebrovascular injury affecting the posterior visual pathways may contribute to vision loss in patients.

Damage to visual pathways can lead to irreversible blindness. Posterior visual pathways, located within a watershed area, are predisposed to hypoperfusion/reperfusion injury. In a novel rat model of bilateral common carotid artery occlusion (BCCAO), oxidative injury to the superior colliculus (SC), a major visual center within the watershed area was evaluated, in addition to its effects on retinal ganglion cells (RGCs). Nitrotyrosine, a footprint of peroxynitrite-mediated oxidative injury in the SC, and microtubule-associated protein 2, a dendrite marker in the retina, were assessed using immunofluorescence and confocal microscopy. Nitrotyrosine-immunoreactivity in the SC was increased 2 weeks after BCCAO compared to controls. Microtubule-associated protein 2-immunoreactivity in the central inner plexiform layer was reduced 3 weeks after BCCAO compared to controls. Global incomplete cerebral hypoperfusion/reperfusion induced oxidative injury in the SC and retrograde RGC dendritic changes. This suggests that cerebrovascular injury affecting the posterior visual pathways may contribute to vision loss in patients.

Inhibition of return (IOR) is a cognitive phenomenon whereby reaction times (RTs) are slower to cued relative to uncued targets at cue-target onset asynchronies (CTOAs) greater than approximately 300 ms. One important theory of IOR proposes that there are two mutually exclusive forms of IOR, with an attentional/perceptual form arising when the oculomotor system is actively suppressed, and a motoric form arising when it is engaged (Taylor & Klein, 2000). Other theories propose that IOR is the result of multiple, additive neural mechanisms (Abrams & Dobkin, 1994). Here, we have performed computational simulations and empirical investigations in an attempt to reconcile these two competing theories. Using a dynamic neural field (DNF) model of the intermediate layers of the superior colliculus (iSC), we have modeled both a sensory adaptation mechanism of IOR, and a motoric mechanism resulting from the aftereffects of saccadic eye movements. Simulating these mechanisms, we replicated behavior and neurophysiology in a number of variations on the traditional cue-target paradigm (Posner, 1980). Predictions driven by these simulations have led to the proposal of many behavioral and neuroimaging experiments which further examine the plausibility of a 2-mechanisms theory of IOR. Contrary to our original predictions, we demonstrated that saccades are biased away from cued targets in a paired target saccade averaging paradigm, even at short CTOAs. In paradigms thought to recruit both sensory and motoric mechanisms, we robustly demonstrated that there are at least two independent, additive mechanisms of IOR when tasks require saccadic responses to targets. When similar paradigms were tested with manual responses to targets, additivity effects did not hold, implying that the motoric mechanism of IOR does not transfer from the oculomotor to skeletomotor systems. Furthermore, across numerous experiments using event-related potential (ERP) techniques, we have demonstrated that P1 component reductions are neither necessary, nor sufficient, for the behavioral exhibition of IOR. We propose that a comprehensive framework for behavioral IOR must include (at least) four independent neural mechanisms, differentially active depending on circumstances, including sensory adaptation, saccadic aftereffects, local inhibition, and cortical habituation.

Game theory outlines optimal response strategies during mixed-strategy competitions in which available actions are selected probabilistically. The neural processes involved in choosing individual strategic actions, however, remain poorly understood. Here, actions need to be selected (1) in the absence of sensory instruction or reward cues and (2) independent of previous events. This thesis examines the neural processes involved in action selection during mixed-strategy competition. To do so, we both measured and manipulated presaccadic activity in the primate superior colliculus (SC), a structure involved in the generation of orienting saccadic eye movements, during a strategic game. The first study tested whether the SC is involved in choosing saccades under strategic conditions. Monkeys were free to choose either of two saccade targets as they competed against a computer opponent during the mixed-strategy game ‘matching-pennies’. The accuracy with which pre-saccadic SC activity predicted upcoming choice gradually increased in the time leading up to the saccade. Probing the SC with supra-threshold stimulation demonstrated that these evolving signals were functionally involved in preparing strategic saccades. Finally, sub-threshold stimulation of the SC increased the likelihood that contralateral saccades were selected. In the second study, we compared the influence of previous actions and rewards on updating premotor activity in the SC in the strategic condition where eliciting stochastic responses was optimal and in a non-strategic condition where stochastic responses were also elicited but through explicit instruction. To avoid exploitation by opponents during mixed-strategy competitions one should select behaviors unpredictably, that is, independent of previous choices and their outcomes. The iterative updating of neural processes involved in selecting actions to produce mixed-strategy behaviors, however, remain poorly understood In both tasks, premotor activity and behavior were shaped by past actions and rewards with more recent events exerting the largest influence. Importantly, these sequential effects were attenuated under strategic conditions suggesting that updating of selection processes is not entirely automatic but can be tailored to different decision-making contexts. Together our results highlight the active role played by the brain in choosing strategic actions. / Thesis (Master, Neuroscience Studies) -- Queen's University, 2009-01-30 17:11:21.002

Neuroeconomic models take economic theory literally, interpreting hypothesized quantities as observables in the brain in order to provide insight into choice behaviour. This thesis develops a model of the neural decision process in strategic games with a unique mixed strategy equilibrium. In such games, players face both an incentive to best-respond to valuations and to act unpredictably. Similarly, we model choice as the result of the interaction between action value and the noise inherent in networks of spiking neurons. Our neural model generates any ratio of choices through the specification of action value, including the equilibrium ratio, and provides an explanation for why we observe equilibrium behaviour in some contexts and not others. The model generalizes to a random-utility model which gives a structural specification to the error term and makes action value observable in the spike rates of neurons. Action value is measured in the spike activity of the Superior Colliculus (SC) while monkeys play a saccade version of matching pennies. We find SC activity predicts upcoming choices and is influenced by the history of events in the game, correlating with a behaviourally-established model of learning, and choice simulations based on neural measures of value exhibit similar biases to our behavioural data. A neural measure of value yields a glimpse at how valuations are updated in response to new information and compared stochastically, providing us with unique insight into modelling choice in strategic games. / Thesis (Ph.D, Economics) -- Queen's University, 2011-06-14 14:48:36.226