Font Size: 

Neurons grown in vitro (neuronal cultures) are an excellent model system to investigate a wealth of problems from a physics of complex systems perspective. Neuronal cultures are able to spontaneously activate, i.e., elicit network-wide action potentials without any external input, and the observed emerging activity is grounded on the intrinsic neuronal dynamics, the connectivity blueprint between neurons and noise. In this talk I will show how the combination of neuroengineering tools with imaging technology and statistical physics analysis allows to design different neuronal circuits and explore their intricacy, with the aim to design controllable in vitro system that approach brain-like dynamics. Specifically, I will first present some examples of engineered neuronal cultures and their monitoring through either calcium imaging or high-density multielectrode arrays. Engineering was conducted by using polymer molds that induce anisotropies in the connectivity of the networks, which favor the guidance of axons or neurons, promoting axonal directionality. Next, I will show how the analysis of the repertoire of activity patterns serves as proxy to evaluate the capacity of the neuronal networks to accommodate diverse dynamical states, and even operate in a state close to criticality. Finally, I will provide examples of the use of chemical or electrical stimulation to alter the dynamic repertoire of the networks and induce plasticity, an aspect of great importance to investigate computation in vitro.