Adaptive Spatiotemporal Receptive Field Estimation in the Visual Pathway
- 1 December 2002
- journal article
- Published by MIT Press in Neural Computation
- Vol. 14 (12) , 2925-2946
- https://doi.org/10.1162/089976602760805340
Abstract
The encoding properties of the visual pathway are under constant control from mechanisms of adaptation and systems-level plasticity. In all but the most artificial experimental conditions, these mechanisms serve to continuously modulate the spatial and temporal receptive field (RF) dynamics. Conventional reverse-correlation techniques designed to capture spatiotemporal RF properties assume invariant stimulus-response relationships over experimental trials and are thus limited in their applicability to more natural experimental conditions. Presented here is an approach to tracking time-varying encoding dynamics in the early visual pathway based on adaptive estimation of the spatiotemporal RF in the time domain. Simulations and experimental data from the lateral geniculate nucleus reveal that subtle features of encoding properties can be captured by the adaptive approach that would otherwise be undetected. Capturing the role of dynamically varying encoding mechanisms is vital to our understanding of vision on the natural setting, where there is absence of a true steady state.Keywords
This publication has 34 references indexed in Scilit:
- Receptive field structure of neurons in monkey primary visual cortex revealed by stimulation with natural image sequencesJournal of Vision, 2002
- Full identification of a linear-nonlinear system via cross-correlation analysisJournal of Vision, 2002
- Recursive stimulus reconstruction algorithms for real-time implementation in neural ensemblesNeurocomputing, 2001
- A simple white noise analysis of neuronal light responsesNetwork: Computation in Neural Systems, 2001
- Nonparametric approach to Wiener system identificationIEEE Transactions on Circuits and Systems I: Regular Papers, 1997
- Adaptation of retinal processing to image contrast and spatial scaleNature, 1997
- Nonparametric identification of Wiener systems by orthogonal seriesIEEE Transactions on Automatic Control, 1994
- A method for constructing data-based models of spiking neurons using a dynamic linear-static nonlinear cascadeBiological Cybernetics, 1993
- Nonparametric identification of Wiener systemsIEEE Transactions on Information Theory, 1992
- Normalization of cell responses in cat striate cortexVisual Neuroscience, 1992