Animal behavior has evolved in a stimulus-rich environment, in which multiple sensory modalities are combined to produce behavior. However, the neural mechanisms underlying this integration are still poorly understood. My goal is to characterize these mechanisms using the larval zebrafish as a model system. Using a combination of video-projected visual stimulation and motor-induced water flow, I will describe composed visual/mechanical multisensory behaviors in freely swimming larva. The relative contribution of each sensory modality to behavioral outputs will be analyzed by varying stimulus strength and using mutant strains with impaired visual and/or mechanical synaptic activity. Next, I will take advantage of the optical accessibility of the zebrafish larva to analyze the complete neural network underlying multisensory behavior. Using a fictive-environment setup, I will perform whole-brain 2-photon calcium imaging of transgenic larva expressing calcium indicators in all neurons. At the same time, I will correlate this information with motor activity recorded from electrical signals in tail motoneurons. Finally, I will describe the contribution of multisensory brain areas through laser ablations. Altogether, these experiments will provide novel and important insights into the basic mechanisms by which the vertebrate central nervous system integrates composed sensory information and translates it into behavior.