The linear relationship between stimulus intensity commanded by the software and the output luminance of the monitor was confirmed in two ways: (1) with a light meter (Konica Minolta, Model LS-100, Tokyo, Japan) and (2) by performing a fast Fourier transform
on visual stimuli photographed with a Dalsa 1M30 CCD camera (Dalsa Corporation, Waterloo, Ontario, Canada) (Rosenberg et al., 2010 and Zhang et al., 2007). Stimuli were viewed from 40 cm and presented as full screen images on a 40 × 30 cm CRT monitor with a pixel resolution of 800 × 600 and a refresh NVP-BGJ398 nmr rate of 100 Hz. Stimuli consisted of high contrast (80% Michelson) drifting or contrast-reversing sinusoidal gratings as well as three-component interference patterns (Equation 1; Figure 1A). In this equation, ωC is the vector defining the carrier spatiotemporal frequencies, ωE is the vector defining the envelope spatiotemporal frequencies, and χ is the vector defining the space and time dimensions (x,y,t). When measuring interference pattern responses with a drifting carrier, the carrier and envelope TFs were constrained to be whole multiples of each other. Without this constraint, the computation time and memory resources to construct and save the stimuli would have been too prohibitive to tailor the stimuli to the cell being studied. Fixing the carrier and envelope TFs to be whole multiples of each other meant constructing only one cycle of each stimulus (at
most 36 frames) rather than a unique frame for each refresh of the selleck kinase inhibitor monitor (200 frames per stimulus; the 100 Hz and monitor refresh rate times the 2 s stimulus duration). Stimuli were presented statically for either 250 ms or 1 s before drifting for a period lasting 1, 2, or 3 s. Firing rates were calculated over the drift duration. Each stimulus was presented between 4 and 12 times. Baseline activity was measured during the presentation of gray screens whose luminance matched the mean luminance of the other stimuli. equation(1) I(x,y,t)=cos(ωC⋅χ)+0.5⋅cos([ωC−ωE]⋅χ)+cos([ωC+ωE]⋅χ)=cos(ωC⋅χ)⋅[1+cos(ωE⋅χ)] SF tuning was measured using between 5 and 9 stimuli. The SFs of the sinusoidal gratings, which were presented
either in isolation or as components of contrast-reversing gratings or interference patterns, could range between 0.02 and 4.0 cyc/°, but rarely exceeded 3.0 cyc/°. The component SFs of the interference patterns did not exceed 2.0 cyc/°. In the LGN experiments, SF tuning curves were fit with difference of Gaussians (Enroth-Cugell and Robson, 1966). In the cortical experiments, SF tuning curves were fit with log-Gaussians. TF tuning was measured using either 6 or 9 stimuli. The TFs of the sinusoidal gratings, which were either presented in isolation or as components of contrast-reversing gratings or interference patterns, typically ranged between 0 and 25 cyc/s. One component of the interference patterns with a carrier TF of 25 cyc/s was higher (25 cyc/s + the envelope TF).