Fig. 7. An illustration of the effectiveness of the brightness controls used in the test of rat color vision. In successive experiments, a rat was required to discriminate between a test light of either 380nm or 510nm and a 510nm comparison light. Each of the test lights was presented an equal number of times at an intensity that had been pre-determined by the brightness equation procedure (see text) to be equally bright (arrow) to the comparison light as well as for values brighter (positive) or dimmer (negative) than the comparison light. Intensities are plotted in logarithmic units relative to the brightness match, and each plotted point shows performance over a total of 100 test trials. The broken line indicates the 99% confidence level. Note that the animal always successfully discriminates between 380nm and 510nm lights, but fails to discriminate between the 510nm test and comparison lights when the brightness cues are eliminated.