Mice prefer rare food

Abstract

In the first experiment designed to test directly for frequency-dependent selection of food by a mammal, mice were presented with pairs of food in 9: 1 and 1:9 combinations. The foods used were crumbs of laboratory rat cake flavoured vanilla or peppermint and dyed green or brown: vanilla/green was presented with peppermint/brown, and vanilla/brown with peppermint/green. Vanilla was preferred to peppermint and green to brown: peppermint/brown was particularly unacceptable. Each type was preferred more when rare than when common. PREFERENCES changed over the five days of the experiment, the preference for greens becoming less strong. Since the experimental trials were carried out in total darkness, the pREFERENCES must have been based on non-visual clues, indicating that the dyes were not odourless and tasteless to mice, as they are to man.