Interaction of STDP and metaplasticity in modelling heterosynaptic plasticity.
Although neuroscientists have still not found a comprehensive mechanism to underlie learning and memory, many investigations suggest that long term potentiation (LTP) and long term depression (LTD) are involved in establishment of learning and memory. As a consequence of certain neural activity, neurons need to modulate the activity of the synapse or the properties of ion channels, therefore, they use a mechanism called homeostatic plasticity to balance their activity and control their firing rate. Two forms of plasticity phenomena that are necessary for plasticity regulation are homosynaptic and heterosynaptic plasticity. In the dentate granule cell, induction of homosynaptic LTP in the activated pathway is accompanied by heterosynaptic LTD in the inactivated pathway. Because, the dentate granule cell shows changes in synaptic strengths, we used this cell to test the following hypotheses. The first hypothesis we propose is, with plasticity and metaplasticity models introduced in this thesis, and the modification of an average postsynaptic spike, we can reproduce homosynaptic LTP and concurrent heterosynaptic LTD. The second hypothesis is the metaplasticity generated after a high frequency stimulation (HFS) reduces the level of synaptic plasticity caused by a second HFS. To test these hypotheses we use computer simulation and combine the nearest-neighbor spike time dependent plasticity (STDP) and metaplasticity rules accompanied with noisy spontaneous activity and the nine compartmental model of a granule cell. For this study we use the experimental data from Abraham et al.(2001), Abraham et al. (2007) and Bowden et al. (2012). With the method mentioned above our model is able to reproduce homosynaptic LTP in the activated pathway and heterosynaptic LTD in the neighboring inactivated pathway. We also show, due to the metaplasticity effects of the plasticity generated from the first HFS, the same magnitude of LTP and LTD will not occur in both pathways during the second HFS. Our finding supports the assertion that the combination of our metaplasticity and nearest-neighbor STDP rules can be a reliable choice to reproduce synaptic plasticity in the dentate granule cell neuron. Our investigation also supports the idea that metaplasticity modulates synaptic plasticity and prevents the synapse from extreme increases, therefore, the same magnitude of synaptic plasticity will not occur during the second stimulation.
Advisor: Lubica, Benuskova; Jörg, Frauendiener
Degree Name: Doctor of Philosophy
Degree Discipline: Computer science
Publisher: University of Otago
Keywords: Metaplasticity; plasticity; STDP; LTP; LTD; heterosynaptic; homosynaptic
Research Type: Thesis