The long-term goal of research in the Bi Lab is to discover and understand fundamental rules underlying the development of neural circuits and their function in learning and memory. Using cultured neurons, with a combination of different experimental and theoretical approaches, we study neural development and plasticity at different levels of organization: from single synaptic boutons to small neuronal networks. Through these studies, we hope to gain insights into both the normal functions of brain circuits and diseases associated with their dysfunction, especially those related to learning and memory.

   At the synapse level, our research aims at identifying quantitative rules of activity dependent synaptic plasticity, and understanding the underlying cellular and molecular mechanisms. Along this line, we have extended the study of spike-timing-dependent plasticity (STDP) by showing that the induction of STDP is modular, and that the integration of potentiation and depression modules is nonlinear and timing dependent. We also use a local-puffing technique to specifically study STDP and heterosynaptic interaction at identified single synaptic boutons. Other ongoing projects explore the homeostatic modulation of STDP and the interaction between STDP and short-term synaptic dynamics.

   At the network level, we are interested in evoked reverberatory activity in cultured networks. We found that recurrent excitation drives network reverberation, as predicted by many theoretical studies. Surprisingly, the oft-neglected asynchronous phase of neurotransmitter release plays a central role in sustaining such reverberation. More importantly, consistent with the cell assembly hypothesized by Donald Hebb as an elementary circuit for thought processes and learning, we have observed that repeated activation can stabilize reverberatory activity in the networks in vitro. We are also developing a new technique of light-directed focal stimulation of neurons grown on silicon chips to further study network dynamics.