First-order motion from contrast modulated noise?

The class of microbalanced motion stimuli is thought to contain no systematic directional biases in motion energy. The fact that we can see motion in such stimuli implies that models of human motion perception based on Fourier decomposition need to be revised. The validity of one widely studied class of microbalanced stimuli, contrast modulated noise, has recently been questioned. It has been proposed that stochastic local biases in the noise carrier give rise to luminance artifacts detectable by a Fourier energy mechanism. However, in this study we show that the response of a motion energy system to contrast modulated noise shows no directional bias over a number of carrier configurations. We conclude that this class of stimuli remains an important tool for researchers wishing to study non-Fourier motion.