All stimuli were generated and controlled by a PC computer running Matlab (MathWorks, Natick, MA, United States) and Psychtoolbox (version 3.0) (Brainard, 1997 (link); Pelli, 1997 (link)). Dichoptic stimuli were rendered on a 3D-ready gamma-corrected computer monitor (ASUS VG278HE; refresh rate: 144 Hz; resolution: 1,920 × 1,080 pixels; background luminance: 54 cd/m2), with participants viewing through a pair of polarized glasses (NVIDIA 3D shutter glasses). The viewing distance was 114 cm, and a chin-forehead rest was used to secure the head position. All experiments were conducted in a dimly lit room (<5 lx).
The “signal” stimuli consisted of oriented ( ± 45° from vertical) sinusoidal gratings at six spatial frequencies (0.5, 1, 2, 4, 8, and 16 c/d) with random phase. Each grating consisted of eight cycles. The size of gratings was inversely proportional to the spatial frequency (i.e., 16°, 8°, 4°, 2°, 1°, and 0.5°), keeping the number of cycles the same across different spatial frequencies. Pixel intensities of the random “noise” stimuli were sampled from a Gaussian distribution (μ = 0, σ = 0.33). Noise stimuli in each trial were sampled independently. The size of the noise mask was the same as that of the signal grating (Chen et al., 2014 (link)). To minimize edge effects, a half-Gaussian ramp was added around the stimulus.
Free full text: Click here