Rapid Analysis of Visual Receptive Fields by Iterative Tomography.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Visual System Research

Background:

• Early visual receptive fields exhibit linear (center-surround) and nonlinear (direction selectivity) properties.
• Traditional methods often study these components separately using distinct stimuli and models.
• Characterizing both linear and nonlinear receptive field aspects simultaneously is crucial for a comprehensive understanding.

Purpose of the Study:

• To develop and validate a novel method for rapid and simultaneous characterization of linear and nonlinear visual receptive field components.
• To apply this method to map receptive fields in primate and rodent early visual systems.
• To correlate neural receptive field maps with underlying neuronal anatomy.

Main Methods:

• Utilized Non-negative Matrix Factorization (NNMF) and Iterative Reconstruction Tomography (IRT) on spiking and voltage potential responses to flashed bars.
• Applied the NNMF-IRT method to individual neurons and ensembles in marmoset lateral geniculate nucleus and mouse retina.
• Combined NNMF-IRT with patch-clamp recording and dye injection in rodent retina.

Main Results:

• Achieved high-resolution receptive field maps for individual neurons and ensembles.
• Demonstrated that the first two IRT components represent standard center-surround receptive fields (DoG-equivalent) in primate geniculate.
• Revealed spatiotemporal receptive field structure of nonstandard on/off-rectifying receptive fields.
• Successfully mapped spatial receptive fields to specific retinal neuron anatomy in rodents.

Conclusions:

• NNMF-IRT is a rapid and flexible framework for analyzing visual receptive fields.
• The method effectively distinguishes and maps both linear and nonlinear receptive field components.
• This approach offers a powerful tool for advancing research in the early visual system across species.