Two forms of feedback inhibition determine the dynamical state of a small hippocampal network
F Zeldenrust, WJ Wadman - Neural networks, 2009 - Elsevier
F Zeldenrust, WJ Wadman
Neural networks, 2009•ElsevierPyramidal cells in the hippocampus are part of a small neuronal network that performs
computations on external input. The network consists of principal cells and various forms of
feedback inhibition. Experimental evidence indicates at least two functionally distinct
inhibitory feedback loops in the CA3 area of the hippocampus:(1) a loop in which O-LM
interneurons project to the distal dendrites of pyramidal cells with synapses that have slow
kinetics, and (2) a loop in which basket interneurons project to the somata of pyramidal cells …
computations on external input. The network consists of principal cells and various forms of
feedback inhibition. Experimental evidence indicates at least two functionally distinct
inhibitory feedback loops in the CA3 area of the hippocampus:(1) a loop in which O-LM
interneurons project to the distal dendrites of pyramidal cells with synapses that have slow
kinetics, and (2) a loop in which basket interneurons project to the somata of pyramidal cells …
Pyramidal cells in the hippocampus are part of a small neuronal network that performs computations on external input. The network consists of principal cells and various forms of feedback inhibition. Experimental evidence indicates at least two functionally distinct inhibitory feedback loops in the CA3 area of the hippocampus: (1) a loop in which O-LM interneurons project to the distal dendrites of pyramidal cells with synapses that have slow kinetics, and (2) a loop in which basket interneurons project to the somata of pyramidal cells with synapses that have fast kinetics. There is an interconnection between the two loops in the form of O-LM to basket interneuron inhibition and the configuration is further complicated by the presence of distinct propagation delays and short-term facilitation and depression of certain synapses in the two basic loops. In this study we investigated the consequences of various configurations of the circuit and modulations of the components of inhibition for the computation that the network can perform on its input. Gaussian noise was used as the input to the dendrite of the pyramidal cell and evoked two types of events: spikes or bursts. The event-triggered average (ETA) and the event-triggered covariance (ETC) were determined and the inter-event-intervals between spikes and bursts were analyzed. The ETA and ETC on the pyramidal cell show that this model behaves in first approximation as an activity integrator: with sufficient positive input, bursts as well as spikes are evoked. Which of the two is determined by the input just after the (first) spike: positive input results in a burst; negative input results in a spike. Stronger feedback inhibition, in the slow as well as in the fast loop, increases the event rate of the pyramidal cell. For a single input and large propagation delays, the interaction between the two feedback loops is not of great importance. The consequences of the presence of the slow and/or fast feedback inhibitory loop, with or without facilitation and depression, were analyzed in relation to synapse strength. Facilitation and depression are most relevant when their recovery time constant is of the same order as the mean inter-event interval. Short-term depression can stop activity in the fast loop after several fast spikes and can switch the network to a different state, thus functioning as a kind of ‘brake’ on the fast inhibitory feedback loop. Thus inhibition and the details of the microcircuit organization play an important role in the information processing of the small neuronal circuit.
Elsevier
以上显示的是最相近的搜索结果。 查看全部搜索结果