Mapping Computation on Neural Circuits

Nervous systems evolved some 600 million years ago and quickly diversified, while facing the same physical constraints of life on Earth and responding to the shared pressures of natural selection. This combination of genetic diversification and shared pressures led to functional convergence (i.e., a need to find solutions to shared problems) and a diversity of solutions (linked to each clade’s or species’ individual history). Therefore, an interesting problem in neuroscience is identifying the levels of description and understanding at which common principles can be defined.

In this Presidential Lecture, Gilles Laurent will describe several examples of neural and circuit computation studied by his lab over recent decades in insects (vision and olfaction), cephalopods (camouflage and texture matching) and reptiles (sleep and circadian rhythms). He will illustrate levels of description that might be suited to identifying shared principles and present the need to diversify our models of neuroscientific study if we wish to identify such seemingly general principles of neural circuit computation.