Sensory arrays made of coupled excitable elements can improve both their input sensitivity and dynamic range due to collective nonlinear wave properties. This mechanism is studied in a neural network of electrically coupled (e.g., via gap junctions) elements subject to a Poisson signal process. The network response interpolates between a Weber-Fechner logarithmic law, and a Stevens power law depending on the relative refractory period of the cell. Therefore, these nonlinear transformations of the input level could be performed in the sensory periphery simply due to a basic property: the transfer function of excitable media.