web  www.oist.jp

Gordon Arbuthnott Unit Brain Mechanisms for Behaviour Unit. Abstract Our aim is to understand, at the systems neuroscience level, the mechanisms that allow the brain to control behavioural responses. It now seems likely that the neurotransmitter, dopamine, has a vital role to play in this regard. However the brain mechanisms, underlying the changes in behaviour that can be engendered by training, remain obscure. We use anatomical, physiological, pharmacological and molecular genetic techniques to understand how dopamine acts on neuronal systems to change learning and performance. The endeavour will be a collaborative one involving other scientists at OIST in testing behaviour, in delineating the biology of the neural systems involved, and in developing theoretical models of brain systems capable of modifying behavioural outcomes. Research Strategies Recently we showed that the synapses on the major output neurons of the striatum can be reduced in number by removal of dopamine and are doubled in number in mice lacking one class of calcium channels. We have so far only been able to show the anatomical effect at the two extremes, but in normal brain we expect this structural action of dopamine to play an important role in the normal sculpting of striatal cell structure to match the physiological requirements of changing motor strategies. With Jeff Wickens and the Neurobiology Research Unit we will develop methods with which to study these dynamic changes in synaptic connectivity. We have developed a cell culture system, where only the corticostriatal system is present, in which to examine the action of dopamine on individual neuronal connections. Such a system would allow us to ‘grow’ neuronal assemblies more like the networks modelled in mathematical simulations. Electrophysiological as well as imaging methods applied to these ‘simple’ networks will allow us to test some of the assumptions in mathematical simulations. By developing methods to interact dynamically with such ‘grown-to-order’ neuronal networks we intend to test their computational power for example in the control of robots like the cyber-rodents used by Kenji Doya’s Neural Computation Unit. Research Goals Dopamine neurons are implicated in a wide variety of neurological and psychiatric disease states and we hope that our scientific knowledge of their basic biology will both, allow a deeper understanding of their role in these diseases, and suggest treatment strategies that respect the complex systems of which they form such a vital part.