Prediction of the Smallest Channel in Early Human Vision

Using the optimal filter description for the difference of two Gaussian (DOG) functions model for early human vision, an extrapolation procedure is used to predict the smallest channel. It is found that the excitatory center of this smallest channel has a diameter of 1 min 22 s. The procedure is based on two factors: the maximum spatial frequency of 60 cycles/° for human visual sensitivity, and the assumption of equal (spatial bandwidth × spatial interval) product.