Derivation of the visual contrast response function by maximizing information rate.

Area of Science:

• Neuroscience
• Computational Neuroscience
• Vision Science

Background:

• Neural responses to stimuli often exhibit an S-shaped function when plotted against stimulus intensity.
• The hyperbolic ratio equation is a common model for these S-shaped neural responses.
• Neuronal responses in early vision to varying contrast levels serve as a key example.

Purpose of the Study:

• To derive the hyperbolic ratio equation with a response exponent of two.
• To explain the underlying principles of neural information processing and cost.
• To elucidate the mechanisms of contrast gain control and normalization in early visual neurons.

Main Methods:

• Derivation of the hyperbolic ratio equation by balancing information rate and neural costs.
• Modeling neural costs as a function of synaptic strength and spike rate.
• Relating maximal response and semisaturation constant to the stimulus ensemble.

Main Results:

• The hyperbolic ratio equation with a response exponent of two was exactly derived.
• Neural costs, dependent on synaptic strength and spike rate, play a crucial role.
• Maximal response and semisaturation constant were linked to stimulus properties.

Conclusions:

• The balance between information rate and neural costs precisely models neuronal S-shaped output.
• This framework explains contrast gain control and normalization observed in early vision.
• Synaptic strength and spike rate are key determinants of neural response characteristics.