Live learning neuronal networks : plasticity of bursts


Stegenga, Jan (2009) Live learning neuronal networks : plasticity of bursts. thesis.

open access
Abstract:We studied the changes in activity of neuronal networks (±50000 cells) cultured on multi electrode arrays (MEAs), due to either spontaneous development of the network or stimulation. An MEA allows recording and stimulation from 60 sites in a two dimensional neural network of dissociated cortical cells. The activity consists of short (50 ms to 1 s) bursts in which most neurons are active. In chapter 2, we show that the profiles of the instantaneous firing rate of neurons during bursts (burst profiles) are highly neuron-specific and are preserved across consecutive bursts. This stability was maintained on a time-base of several hours and paralleled development of the physical structure. In chapter 3, we applied interventions consisting of single stimuli (single electrode single pulse), stimulus pairs, trains of stimuli and paired trains of stimuli. We compared the difference between burst profiles measured before and after the interventions with spontaneous development. The inconsistent results were compared to literature and we concluded that bursting stabilized the network, raising the threshold for change. In chapter 4 we used random stimuli applied to ±10 electrodes to establish rhythmic activity (4 Hz) without network bursts. Only when trains of stimuli were applied at the peak of the rhythm, upregulation of the responses to test stimuli was observed, as well as an increased change in burst profiles. Therefore, different signaling occurs dependent on the phase of the rhythm. This shows the importance of natural activity patterns in general and rhythmic activity in particular. We show in chapter 5 that feedback of the response to single stimuli can be used to cultivate initially random change such that it becomes part of the normal repertoire of activity patterns in a culture. Due to feedback, the responses to stimuli and burst profiles were changed. Thus, the stability of bursts can be overcome by applying external stimulation to reach a state without bursts, or by cultivating small changes through feedback.
Item Type:Thesis
Electrical Engineering, Mathematics and Computer Science (EEMCS)
Research Group:
Link to this item:
Official URL:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page