The external noise normalized gain profile of spatial vision

Area of Science:

• Vision science
• Computational neuroscience
• Psychophysics

Background:

• The contrast sensitivity function (CSF) is commonly used to model visual system gain, but it is influenced by processing inefficiencies.
• Using CSF directly as a gain profile in observer models may be problematic for accurately predicting spatial pattern perception.

Purpose of the Study:

• To re-evaluate the gain profile of the visual system using the external noise paradigm and the perceptual template model (PTM).
• To identify key processing components, including nonlinearity and internal noise, that shape the CSF.

Main Methods:

• Application of the external noise paradigm to normalize visual system gain.
• Utilizing the perceptual template model (PTM) to analyze experimental data across various spatial frequencies and conditions.
• Characterizing internal additive and multiplicative noise parameters.

Main Results:

• The PTM successfully accounted for 92.8% of the variance in experimental data.
• The external noise-normalized gain profile of the visual system was found to be relatively flat across spatial frequencies.
• Key visual system properties, including gain, nonlinearity, and noise, were characterized.

Conclusions:

• The perceptual template model provides a more accurate representation of the visual system's front-end gain profile than the CSF.
• A flat gain profile is a more appropriate model for visual system processing across spatial frequencies.
• These findings have significant implications for understanding both normal and impaired spatial vision.