Towards a biophysical basis of spike based inference

Abstract

Event Abstract Back to Event Towards a biophysical basis of spike based inference Cells in the retina, thalamus, and sensory cortices integrate inputs from thousands of synaptic afferents. Sensory stimulation varies over time and in response to specific events in this stream, these cells emit spikes. What are the generic dynamical properties of sensory neurons allowing them to extract information from their input and signal it efficiently? How can these properties be realized by biophysical mechanisms? We previously derived dynamical equations for an optimal “Bayesian” spiking neuron. We used a generative model (GM) to describe how synaptic input is caused by the dynamic stimulus. This formulation allowed us to derive in a principled way an equation for optimalsynaptic integration. In addition, we supposed that the model is self consistent, i.e. that the same GM can be used to estimate the stimulus from the output spike train. This implied a spike generation mechanism based on an adapting firing threshold. Since the model is designed to optimally transmit information, it can be used as a tool to understand neural processing. It provides a normative yardstick against which to compare empirical spike data. Furthermore, it helps to understand how different components in the complex neural dynamics affect sensory processing. The Bayesian model, however, is motivated by optimality principles rather than biological realism. In order to bridge the gap between normative and biophysical models, we analyzed its dynamics for different input regimes where the dynamics can be described by mechanisms known from biophysical models. Our results suggest that elements necessary to instantiate optimal information transfer are (a) a leak current adjusting the time constant to the temporal stimulus statistics, (b) a voltage-dependent depolarizing term preventing the membrane potential to become too hyperpolarized, and (c) spike based adaptation increasing the membrane conductance after each spike. Signatures of these elements have been found in thalamic and cortical neurons: besides a variety of leak currents, many cells show hyperpolarisation activated depolarizing I_h currents and Ca2+ controlled K-currents represent a significant source of spike based adaptation. Such detailed descriptions contrast with stochastic models of sensory neurons like the Linear Nonlinear Poisson model. We argue that Bayesian spiking neurons are both more efficient at transferring information and provide additional insight into biophysical detail. Nevertheless, they are efficient to simulate and account for the apparent Poisson-like response variability of cortical neurons. Our analysis provides important links between biophysical mechanisms and GMs describing the stimulus statistics e.g. how often stimuli appear and how long they last. The probabilistic interpretation clarifies the functional significance of different currents and how they enable cells to adapt to different input regimes. Showing how information transmission breaks down if specific components are blocked forms a basis to understand how natural neuromodulators or pharmacological substances acting on the corresponding channels affect sensory processing. By putting emphasis on the inference task as the ”problem to be solved” and linking it to more phenomenological descriptions of neural dynamics we can advance our understanding of the link between the biophysics of sensory neurons and perception. Conference: Computational and systems neuroscience 2009, Salt Lake City, UT, United States, 26 Feb - 3 Mar, 2009. Presentation Type: Poster Presentation Topic: Poster Presentations Citation: (2009). Towards a biophysical basis of spike based inference. Front. Syst. Neurosci. Conference Abstract: Computational and systems neuroscience 2009. doi: 10.3389/conf.neuro.06.2009.03.268 Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters. The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated. Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed. For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions. Received: 04 Feb 2009; Published Online: 04 Feb 2009. Login Required This action requires you to be registered with Frontiers and logged in. To register or login click here. Abstract Info Abstract The Authors in Frontiers Google Google Scholar PubMed Related Article in Frontiers Google Scholar PubMed Abstract Close Back to top Javascript is disabled. Please enable Javascript in your browser settings in order to see all the content on this page.