Understanding how the perception of an environment emerge from the collective dynamics of the massive assembly of neurons that constitutes the vertebrate brain is arguably one of the most challenging scientific question of our time. The multi-scale nature of the brain allows for several complex processing, among which the integration of sensory inputs plays a central role. Each sensory system transduces a different form of energy and provides the brain with an independent picture of the environment, which has two major consequences: (i) animals are subjected to a barrage of information for which the brain must continuously evaluate the relative priorities, and (ii) the brain can use the information from different sensory channels to enhance event detection. The aim of this proposal is to build a set of experimental and analytical tools that will allow, for the first time, to record brain-wide neuronal activity during multi-modal integration tasks and to decipher the complex neural processes involved.
To this aim, we plan to use a specific animal model, zebrafish larvae. Zebrafish, which was originally developed as a model for embryo development and tissue regeneration, has recently emerged as a similarly important model system for neuroscience. The rapid development of calcium imaging – tightly related to the progress of genetics – has led to remarkable improvements on the number of neurons whose activity can be simultaneously recorded. Recently, Single-Plane Illumination Microscopy (SPIM) approaches yielded even more spectacular progress allowing to record simultaneously the activity of approximately 85% of the neurons of an intact zebrafish larva, yet with cellular resolution.
These groundbreaking advances in the field of functional imaging allow us to reconsider long-standing question in neurosciences with a fresh perspective. Here, we propose to study some fundamental issues related to multi-sensory integration with brain-wide neuronal activity recordings during a simple stimulation paradigm. In the framework of a collaboration with experienced theoreticians, we will test and develop models of network inference on extensive datasets, and explore to what extent the zebrafish brain uses probabilistic approaches to handle the intrinsic variability of sensory inputs.
We believe that the convergence between new experimental and theoretical tools convey great promises to unveil fundamental mechanisms of multi-sensory integration in the vertebrate brain.
Monsieur Raphaël CANDELIER (Laboratoire Jean Perrin)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
LJP Laboratoire Jean Perrin
Help of the ANR 282,400 euros
Beginning and duration of the scientific project: September 2016 - 36 Months