Remote controls illusion: strange interactions across space cannot be explained by simple contrast filters

Separation of luminance and contrast modulation in steady-state visual evoked potentials

Neurons in the retina and early visual cortex respond primarily to local luminance contrast rather than overall luminance energy. The distinction between luminance and contrast processing is revealed in its most striking form by the contrast asynchrony paradigm: two discs with bright and dark surrounds modulate in luminance. When the discs modulate at 3-6 Hz, there is a percept of antiphase flicker even though the luminance modulation of the patches is in phase. To establish the neural basis of this perceptual phenomenon, we conducted a study using steady-state visual evoked potentials (SSVEPs) aiming to identify specific contrast and luminance signals. Deconstructing contrast asynchrony into its constituent elements, we displayed eight discs modulating sinusoidally from dark to bright on one of three backgrounds (bright, midgray, dark). In the first experiment, disc modulation and background luminances spanned a narrow range (30-34 cd/m2) to avoid VEP saturation (Weber contrast ≤15.5%) at two frequencies: 3 Hz, falling inside the contrast asynchrony temporal range, and 7.14 Hz, falling outside this range. In the second experiment, luminances and contrasts spanned a large range (0-64 cd/m2) at three frequencies (3, 5, 7.14 Hz) to evaluate the degree to which VEP response non-linearities would affect observed data patterns. With lower contrast modulation at 3 Hz, SSVEP amplitudes and phases correspond to the temporal signatures of contrast - not luminance - modulation. However, at higher frequencies and/or contrasts, this orderly pattern was largely replaced by more complex patterns that no longer directly corresponded to the luminance or contrast of the stimulus.

The Perpetual Diamond: Contrast Reversals Along Thin Edges Create the Appearance of Motion in Objects

The Perpetual Diamond produces motion continuously and unambiguously in one direction despite never physically changing location. The phenomenon consists of a steady, mid-luminance diamond bordered by four thin edge strips and a surrounding background field. The direction of motion is determined by the relative phases of the luminance modulation between the edge strips and the background. Because the motion is generated entirely by changing contrast signals between the edge strips and background, the stimulus is a valuable tool for tests of spatial contrast, temporal contrast, contrast gain, and color contrast. We demonstrate that observers see motion even when the edge strips subtend only seconds of arc on the retina (which is less than the frequently reported 10 minutes of arc) and that perceived motion is due entirely to changes in the difference in contrast phase modulation, independent from the luminance phase.