Influence of Taste and Color Cues on Bathing by Starlings in Appetitive and Adversive Contexts: Implications for Animal Damage Control

The importance of color and taste in feeding and drinking by omnivorous birds is context·dependent and influenced by learning. Here, we report three experiments designed to assess the influence of such characteristics on starlings. In Experiment 1, eight starlings were given a choice between bathing in red or plain water and 0.15 M NaCI solution or plain water. The frequencies of bathing, drinking, and preening were recorded. Red water was avoided (p <: 0.05), but no preferences were observed between NaCI solution and plain water (p / 0.25). That 0.15 M NaCI was not avoided is surprising, because it is rejected by starlings when drinking. Perhaps starlings do not taste substances while bathing but continue to ingest substances that they would otherwise reject. In Experiment 2, we assessed these alternative explanations and also tested (a) whether starlings would bathe in colored water if plain water was unavailable, (b) whether starlings would show preferences among such colors, and (c) whether preferences courd be altered by learning. Twenty-four starlings were assigned to three conditions. Birds in the first condition were presented with red and blue baths and relative preferences for bathing in these colors were assessed. Birds in the second condition were presented with a blue bath and intubated with methiocarb or propylene glycol. Birds in the third condition were presented with a saccharin bath and intubated with methiocarb or propylene glycol. On the four days following treatment, birds in the second condition were given two-choice tests between red and blue baths. Those in the third condition were given two-choice tests between bathing in saccharin solution and plain water. Birds readily bathed in red and blue water when plain water was unavailable. After treatment, however, birds avoided blue water (p < 0.05), but aversions dissipated rapidly. Learned aversions for saccharin were also obtained (p < 0.05); these remained strong over all tests. Experiment 3 was designed to assess the differential importance of taste and color. Sixteen starlings were aSSigned to four groups. Two groups were food-deprived and then given dogfood in a red cup followed by a bath of 0.15 M NaCI or LiCI. The other two 99 100 groups were presented with a bath of 0.15 M NaCI or LiCI only, as a control. On the four days immediately following treatment, all groups were given two-choice feeding (red vs. blue food cups) and bathing (NaCI vs. plain water) tests. Aversions were expressed towards color in the feeding context (p <: 0.05) but not taste in the bathing context (p /' 0.25). We inferred that color cues in the feeding context overshadowed taste cues in the bathing context. The present results may have implications for control. Starlings will bathe even under harsh environmental conditions, and one control strategy might be to pair livestock feed with distinctive colors and provide lithium·adulterated bathing stations nearby. Starlings eating feed and bathing in the solution might form color aversions and subsequently avoid the food. Also, the use of such techniques might enhance already existing control, such as the use of starlicide baits. Depredating starlings would be directed toward such baits as birds feeding in the laboratory are directed towards food color combinations not explicitly paired with lithium-induced malaise.