INTERFACING LIVING NEURAL CIRCUITS TO STUDY THE EMERGENCE OF ORIENTATION SELECTIVITY

A prominent feature of the primary visual cortex of mammals is the selectivity of neurons to oriented edges. While primates and carnivores arrange orientation selective neurons into columnar domains, rodents distinct from this nearly invariant pattern exhibit an interspersed layout of orientation selectivity. In both architectures orientation selectivity is thought to arise from the convergence of afferent input from a non-selective prior stage of the visual pathway. To better understand the emergence of orientation selectivity and the potential transitions between different arrangements we designed a synthetic model of the early visual pathway that combines computational and biological components. This hybrid system allows to test different thalamo-cortical connection schemes and their influence on orientation selectivity in living neuronal tissue. To this end, we express a light-sensitive ion-channel in neurons and use a holographic stimulation system to generate spatial-temporal patterns that mimic thalamic input computationally obtained from a model of the afferent connectome. The target were either primary cell cultures from cortical neurons or acute tangential brain slices of visual cortex. Using this system we measured a loss of orientation bias with decreasing size of orientation modules with significant orientation tuning remaining even in the absence of specific inputs. By using reverse correlation we linked this intrinsic orientation bias to interactions within the cortical circuit, highlighting the potential of intracortical connections to drive the emergence of orientation selectivity. Hybrid systems as ours can provide valuable insights into processing in the visual cortical pathway and its evolutionary transformation.