in neuronal networks explained (cont)
Our research started with the simple insight that this kind of instability could be prevented if neurons could regulate their own excitability. For example, if neurons have a way of sensing how active they are and can adjust the amount of synaptic excitation they receive to maintain this activity within some set limits, that would prevent the correlation between any two neurons from rising unchecked. We were able to use a very simple kind of experiment in cell culture to show the existence of this kind of plasticity. We found that artificially increasing activity in networks of cultured cortical neurons caused the neurons to reduce the strength of all of their synaptic connections, which in turn reduced the amount of excitation the neurons received and restored activity to the appropriate level. Conversely, artificially lowering activity caused neurons to increase the strength of all of their synaptic connections, again bringing activity back into the right range. These results suggest that two forms of plasticity operate hand-in-glove to effectively store information in networks of neurons. Hebbian mechanisms can modify particular synaptic strengths according to correlations in activity and allow a child (and even an adult on occasion) to rapidly form associations between qualities, objects, and concepts. At the same time, homeostatic mechanisms provide stability to those same networks so that the fundamental characteristics of our brains are preserved.
While exploring the theoretical requirements for stable learning may seem quite abstract, this work is likely to have important practical consequences. The realization that homeostatic plasticity plays a critical role in learning and development is likely to open up a whole realm of possible avenues for treatment of developmental abnormalities. For example, epilepsy results when certain parts of the brain become overly active. An understanding of the mechanisms that normally keep neuronal activity within limits may provide some crucial insights into how such abnormal activity develops, and how it can be prevented.
Gina Turrigiano, winner of a 2000 MacArthur genius fellowship, is associate professor of biology at the Volen National Center for Complex Systems, Brandeis University.