Complexity lies halfway between stochasticity and determinism, suggesting that brain activity is neither fully random nor fully predictable but lives by the rules of nonlinear high- and low-complexity dynamics. One important aspect of brain function is noise-like activity (NLA), defined as background, electrical potential fluctuations in the nervous system distinct from spiking rhythms in the foreground. The objective of this thesis was to investigate the neurodynamical complexity of NLA recorded at the cellular and local network scales in in vitro preparations of mouse and human hippocampal tissue, under healthy and epileptiform conditions. In particular, it was found that neuronal NLA arises out of the physiological contributions of gap junctions and chemical synaptic channels and is characterized by a spectrum of complexity, ranging from high- to low-complexity, that was measured using methods from nonlinear dynamical systems theory. Importantly, the complexity of background, neuronal NLA was shown to depend on the degree of cellular interconnectivity to the surrounding local network. In addition, the complexity and multifractality of NLA was further studied at the cellular and local network scales in epileptiform transitions to seizure-like events, identifying emergent low-complexity and reduced multifractality (bordering on monofractal-type dynamics) in the pathological ictal state. Finally, dual intracellular recordings of hippocampal epileptiform activity were analyzed to measure NLA synchronicity, showing evidence for increased same- and cross-frequency correlations and increased phase synchronization in the pathological ictal state. Convergence towards increased phase synchrony manifested in lower frequency regions including theta (4-10 Hz) and beta (12-30 Hz), but also in higher frequency bands (gamma, 30-80 Hz). In summary, there is evidence to suggest that background NLA captures important neurodynamical information pertinent to the classification and identification of brain state transitions in healthy and epileptiform hippocampal dynamics, using sophisticated neuroengineering analyses of these physiological signals.

Complexity lies halfway between stochasticity and determinism, suggesting that brain activity is neither fully random nor fully predictable but lives by the rules of nonlinear high- and low-complexity dynamics. One important aspect of brain function is noise-like activity (NLA), defined as background, electrical potential fluctuations in the nervous system distinct from spiking rhythms in the foreground. The objective of this thesis was to investigate the neurodynamical complexity of NLA recorded at the cellular and local network scales in in vitro preparations of mouse and human hippocampal tissue, under healthy and epileptiform conditions. In particular, it was found that neuronal NLA arises out of the physiological contributions of gap junctions and chemical synaptic channels and is characterized by a spectrum of complexity, ranging from high- to low-complexity, that was measured using methods from nonlinear dynamical systems theory. Importantly, the complexity of background, neuronal NLA was shown to depend on the degree of cellular interconnectivity to the surrounding local network. In addition, the complexity and multifractality of NLA was further studied at the cellular and local network scales in epileptiform transitions to seizure-like events, identifying emergent low-complexity and reduced multifractality (bordering on monofractal-type dynamics) in the pathological ictal state. Finally, dual intracellular recordings of hippocampal epileptiform activity were analyzed to measure NLA synchronicity, showing evidence for increased same- and cross-frequency correlations and increased phase synchronization in the pathological ictal state. Convergence towards increased phase synchrony manifested in lower frequency regions including theta (4-10 Hz) and beta (12-30 Hz), but also in higher frequency bands (gamma, 30-80 Hz). In summary, there is evidence to suggest that background NLA captures important neurodynamical information pertinent to the classification and identification of brain state transitions in healthy and epileptiform hippocampal dynamics, using sophisticated neuroengineering analyses of these physiological signals.