Computational neuroscience of homeostatic plasticity:
Modelling structural plasticity in prefrontal and
hippocampal networks
Markus Butz
Abteilung Neuroanatomie
Fakultät für Biologie
Montag, 25.04.2005, 16 Uhr c.t., H 10
Progressive and regressive events during development and in the mature brain are crucial for
maintaining functional conditions in biological neural networks in a homoeostatic manner i.e.
stabilizing neuronal activity. From increasing or decreasing synaptic receptor densities and
altering synaptic efficiency over synthesizing and abolishing synaptic contacts up to cell
proliferation and apoptosis etc., structural processes permanently reorganize the connectivity
among nerve cells within mutually every part of the nervous system. In the mature brain, different
regions show characteristic strategies of plasticity according to their functional needs. The
dentate gyrus, entrance region of the hippocampal formation, shows a dramatic but functional
reorganisation involving an ongoing synaptogenesis and cell turnover. In contrast, the prefrontal
cortex shows a milder, predominantly synaptic reorganization. By the use of our computational
modelling approach we are able to show that cell proliferation and synaptogenesis together
contribute to a homoeostatic stability of hippocampal networks and enable them to adapt to a
dramatically changing input spectrum via perforant path fibres. In other words, the proportion of
new cells being synaptically integrated into a pre-existing network depend not only on the rate of
cell proliferation alone, but also on the current activity of the network. Prefrontal networks,
not endowed with cell turnover, suffer severe alterations of their connectivity pattern when
affected by an adverse afference spectrum, i.e. a juvenile trauma or a pharmacological
intoxication. Thus, the structural basis of a dysconnection syndrome as it can be seen in
schizophrenia may relay on a common mechanism effecting activity-dependent plasticity in a
functional and pathological situation.