To survive and prosper, any motile organism must extract relevant cues from its external environment and convert them into coherent internal representations. Responses elicited by the flow of internal representations form the basis of adaptive behaviours. The overall goal of this project is to elucidate how olfactory signalling and behavioural decisions emerge from the integration of molecular and cellular interactions. Despite the relatively good anatomical characterisation of insect and vertebrate olfactory systems, little is known about the neural basis of odour percept formation and their processing in the brain. We will address this problem in the simple and genetically tractable system of the fruit fly (Drosophila melanogaster) larva. An odorant stimulus can be specified by three variables: odour quality, odour intensity and time of occurrence. To chemotax, larvae are able to reliably perceive and integrate changes relative to these variables by using noisy detectors. We aim at identifying and functionally characterising the peripheral and central neuronal circuits involved in larval chemotaxis. A cross-disciplinary approach will be adopted to combine the power of fly molecular genetics, state-of-the-art engineering, electrophysiology and computational modelling to determine: how odour intensity is represented by the peripheral sensory neurons; where intensities measured at different times are stored and subsequently compared and whether the design principles of the larval olfactory system allow noise filtering. Clarifying how an elementary nervous system solves these fundamental questions will increasing our understanding of the relationships between neuronal circuits in the brain and behaviour.