Hybrid Neural-Microelectronic Systems
Bringing living neurons and model systems together
Michael Sorensen, Carrie Williams, Kate Williams
 |
An aplysia californica looks forward to its participation in breathtaking scientific research |
Neural modeling and experimentation have been viewed traditionally as separate, albeit complementary, fields. Data from neural models suggest new experimental directions; experiments, in turn, suggest refinements to the model.
We are bringing models and experiments together, creating systems whose activity is a product of both the biological and the modeled components. In these hybrid systems, the model component allows us to manipulate the system in ways that are impossible with traditional neurophysiological techniques. The biological component, however, provides the system with a level of realism and relevance that cannot be achieved with modeling alone. For example, by replacing a single neuron in a biological neural network with a modeled neuron, we can investigate how the properties of that neuron affect the behavior of the network.
We use both invertebrate and vertebrate systems in our research. The small-scale networks in invertebrates, such as in the sea slug aplysia californica or the medicinal leech hirudo medicinalis, allow us a better understanding of how individual components function together to create a useful result. This understanding yields principles upon which neural networks organize their behavior. Vertebrate networks, such as the rat spinal cord, provide us with a means to investigate how these principles are applied in larger systems. Vertebrate networks also provide us with insight into how our technologies could be applied to the human nervous systems.
 |
By making changes in the parameters of a model neuron (red trace) we can alter the activity of a living neuron (blue trace). |