In computer-generated spatiotemporal noise every stimulus frame contains a new static noise sample. The spectral density of white spatiotemporal noise is calculated by multiplying the squared rms contrast of noise by the product of the noise check area and the exposure duration of each noise check. When the exposure duration of each noise check is gradually increased, the spectral density of spatiotemporal noise increases, reaching its maximum when noise becomes static. In static spatial noise both stimulus and noise checks have the same duration. The signal-to-noise ratio is known to be constant at detection threshold. Detection thresholds should thus increase in proportion to the spectral density of spatiotemporal noise, which increases with the duration of the noise checks. We measured detection thresholds for stationary cosine gratings embedded in spatiotemporal noise. The exposure duration of the noise checks was increased from one frame duration to the total exposure duration of the stimulus grating. Noise was thus gradually transformed from spatiotemporal to static spatial noise. The contrast energy threshold increased in proportion to the spectral density of spatiotemporal noise up to a noise check duration found to be equal to the integration time for the stimulus grating without noise. After this, energy thresholds remained constant in spite of the increase in the spectral density of spatiotemporal noise. This suggests that the masking effect of spatiotemporal noise increases with the duration of noise checks up to the critical duration marking the saturation of the temporal integration of the signal.
Effects of weak spatiotemporal noise on a bistable one-dimensional system
