The mouse has become one of the most important mammalian targets for modern basic and bio-medical research. This thesis project is one of many efforts that are underway in the vision research community to understand mouse visual function. The goal of the work was to apply behavioral techniques to the visual performance of wild type (WT) and certain mutant mice addressing the question whether mouse vision is governed by the same basic psychophysical laws as those established for humans. We developed a novel non-invasive behavioral instrument/method that yields first-ever visual sensitivity measurements while the mouse performs a natural behavior, running. Experiments were designed to collect increment threshold data for the rods and cones of individual WT and mutant mice, using the method of constant stimuli. The experiments were carried out with small well-calibrated targets that stimulated (for the first time) the retina locally.

Our results show that the absolute threshold intensity response of mouse rods was more sensitive than that of UV-cones by a factor of about 3000 and more sensitive than that of M-cones by a factor of about 12,500. The spatial integrating area of the mouse rod retina extended to about 450 rods. It was possible to register a threshold response mediated by no more than 25 rods imaged by the test flash. A brief, 500 nm light flash mediated by rods at threshold produced about 10-15 photoisomerizations in the dark-adapted mouse retina. Threshold versus intensity (t.v.i.) curves obtained on WT mice with test and background stimuli of various wavelength combinations were displaced horizontally and vertically as determined by the spectral sensitivity of the rods. As previously reported by others, the spectral sensitivity of Gnat-1^-/- mice, that lack functional rods, could be fitted with two pigment templates, one that peaked at about 360 nm, the other at 508 nm. Isolation of UV- and M- cone responses in WT mice was achieved in the presence of backgrounds that produced about 25 - 50 and 200 - 300 photoisomerizations rod^-1 s^-1, respectively.

Due to limitations of the light source, the characteristics of the UV-cones could not be investigated. The M-cone increment threshold curve of the WT mouse showed a peculiar deviation from the Weber line. To shed light on this peculiarity, the threshold responses of Gnat-1^-/- mice to 510 nm stimuli and those of Gnat-2^-/- mice to 365 nm and 500 nm stimuli were determined. We found that the M-cones of Gnat-1^-/- mice desensitized profoundly in the presence of moderately bright lights suggesting that M-cones were not the mediators of vision in bright light. The increment threshold curve rose with a slope of 1.5 in the "Weber-adaptation" region. In Gnat-2 ^-/- mice all visual responses were mediated by rods. At a field intensity that produced 33,000 R* s^-1 and that is known to saturate human rods, mice threshold responses did not deviate from the Weber line. However, the minimum Weber-Fechner fraction was 0.03 (in the high intensity region). For WT mice the Weber fraction was 0.09. The adaptive behavior of the mice to high intensity background light may be explained by the operation of a mechanism of light adaptation that was recently discovered (Nikonov et al., 2006). We propose that the mouse is very much a nocturnal animal whose vision is dominated by rods, over a wide range of light levels. The mouse's rod system fundamentally resembles that of humans in its basic operations. It has achieved comparable absolute sensitivity as that of humans. The noise levels in the rod system of the mouse retina are about the same as in humans. Lastly, the rods of the mouse show Weber adaptation. Mouse rods are different from human rods in that they can operate in light conditions under which human rods would saturate. We believe that the present work has developed a number of critical benchmarks that could be helpful in the evaluation of physiological data from single cell and electroretinographical studies as well as the evaluation of data from molecular biological studies.