Abstract
Digital Light Processing (DLP) offers potential for use as a visual stimulator. At the heart of this technology is a Digital Micromirror Device (DMD) consisting of an M by N array of tiny mirrors. Each individual mirror switches between two discrete angles at high frequencies. Images are produced by controlling the duty cycle of each mirror independently, selectively passing different intensities of light down an optical path to the eye. Previous implementations of DLP technology for vision research have used a three-chip system in which the chips were precisely aligned and primaries were produced by passing a bright, broadband source through dichroic filters (Packer et al). When compared to the predecessor technology, Cathode Ray Tube (CRT), DLP offers greater range in luminance, an expanded gamut, and pixel independence (Packer et al). Here we present a new visual stimulator design based on a commercially available one-chip DLP-based system in which the RGB channels were presented serially within each frame. It is a Maxwellian view system with a point source of light composed of nine spectrally distinct LEDs with peak wavelengths at 382, 405, 470, 505, 530, 570, 594, 660, and 695 nanometers. During each R, G, and B screen update any combination of the nine primaries can be added together in any combination of intensities. The one chip design has the advantage of being much simpler, more compact and cost effective. The use of a single small light source makes the system very versatile with regard to the intensity and spectral character of the light output. In the present implementation a broad range of primaries allows for gamuts suited for animals for example rodents which have an S-cone peak of 360nm (Jacobs, Neitz, & Deegan). Also, because VGA signals are accepted by the projector, common visual stimulation generators such as the Cambridge Research Systems VSG and ViSaGe can be used.