White noise analysis is a powerful tool for characterizing the input-output relationships of sensory neurons. If the probability that a neuron fires a spike that can be expressed as a function of a low-dimensional linear projection of a visual stimulus, we can (usually) characterize the neuron’s input-output relationship using the techniques of spike-triggered averaging and spike-triggered covariance.
These techniques have shown us that some neurons in the primary visual cortex of macaques respond to visual stimuli in an interesting way. They respond to one color and are suppressed by the opponent color (e.g. they fire action potentials in response to blue blob but not yellow, and vice versa). They are also sensitive to luminance edges: they respond better to their preferred color when it is superimposed on a luminance edge than when it is not.
We would like to understand how this combination of chromatic and luminance signals is implemented by neural circuitry and what computational goal it accomplishes to help us see.