Analyzing Neural Responses to Natural Signals: Maximally Informative Dimensions
From MaRDI portal
Abstract: We propose a method that allows for a rigorous statistical analysis of neural responses to natural stimuli which are non-Gaussian and exhibit strong correlations. We have in mind a model in which neurons are selective for a small number of stimulus dimensions out of a high dimensional stimulus space, but within this subspace the responses can be arbitrarily nonlinear. Existing analysis methods are based on correlation functions between stimuli and responses, but these methods are guaranteed to work only in the case of Gaussian stimulus ensembles. As an alternative to correlation functions, we maximize the mutual information between the neural responses and projections of the stimulus onto low dimensional subspaces. The procedure can be done iteratively by increasing the dimensionality of this subspace. Those dimensions that allow the recovery of all of the information between spikes and the full unprojected stimuli describe the relevant subspace. If the dimensionality of the relevant subspace indeed is small, it becomes feasible to map the neuron's input-output function even under fully natural stimulus conditions. These ideas are illustrated in simulations on model visual and auditory neurons responding to natural scenes and sounds, respectively.
Recommendations
Cites work
- scientific article; zbMATH DE number 1164155 (Why is no real title available?)
- A simple white noise analysis of neuronal light responses
- Computation in a Single Neuron: Hodgkin and Huxley Revisited
- Estimation of Entropy and Mutual Information
- Neural coding and decoding: communication channels and quantization
- Statistics of natural time-varying images
- The statistics of natural images
Cited in
(24)- Characterizing responses of translation-invariant neurons to natural stimuli: maximally informative invariant dimensions
- Scalable Variational Inference for Low-Rank Spatiotemporal Receptive Fields
- How Close Are We to Understanding V1?
- Model-based decoding, information estimation, and change-point detection techniques for multineuron spike trains
- Space-by-time tensor decomposition for single-trial analysis of neural signals
- Context-dependent coding in single neurons
- Learning quantitative sequence-function relationships from massively parallel experiments
- Equitability, mutual information, and the maximal information coefficient
- A novel nonparametric approach for neural encoding and decoding models of multimodal receptive fields
- Multi-input, multi-output neuronal mode network approach to modeling the encoding dynamics and functional connectivity of neural systems
- Identifying functional bases for multidimensional neural computations
- Periodic clustering of simple and complex cells in visual cortex
- Variable selection for naïve Bayes classification
- Estimating Information Rates with Confidence Intervals in Neural Spike Trains
- An estimate of mutual information that permits closed-form optimisation
- Spatio-chromatic information available from different neural layers via gaussianization
- System identification of \textit{Drosophila} olfactory sensory neurons
- Modeling convergent ON and OFF pathways in the early visual system
- Learning quadratic receptive fields from neural responses to natural stimuli
- Parametric inference in the large data limit using maximally informative models
- Nonlinear modeling of neural interaction for spike prediction using the staged point-process model
- Spike-triggered covariance: geometric proof, symmetry properties, and extension beyond Gaussian stimuli
- NeuroXidence: reliable and efficient analysis of an excess or deficiency of joint-spike events
- Neural decoding with kernel-based metric learning
This page was built for publication: Analyzing Neural Responses to Natural Signals: Maximally Informative Dimensions
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4819822)
