![]() |
|
In winter 2004/2005, Jennifer Smolinski and I examined the numerical competence of wild Northern Mockingbirds (Mimus polyglottos). Five individually color-banded mockingbirds on established winter territories on the campus of Xavier University in southwestern Ohio were presented with transparent acrylic bird feeders designed to test numerical competence. Each feeder had six holes drilled on either side so that bamboo sticks could be easily inserted and removed. To obtain mealworms from the feeder, a mockingbird needed to remove all sticks on one side of the apparatus. If mockingbirds possessed numerical competence, we hypothesized that they would remove sticks from the side of the feeder that contained fewer sticks, thus more quickly and easily receiving a food reward. The birds were presented with trials involving random placement and number of sticks: initially one stick on one side and six sticks on the other side (1 vs. 6), then 2 vs. 5, and 3 vs. 4. Statistical analysis of the data was carried out using one-tailed sign tests. In the 1 vs. 6 and 2 vs. 5 trials, all five subjects showed a statistically significant preference for the side of the feeder containing fewer sticks (all P<0.05). All five subjects failed, however, to discriminate between 3 vs. 4 (all P>0.05). The results of this study demonstrate the ability of wild birds to solve optimal foraging problems on the basis of number. Click here for an illustrated example. In winter 2005/2006, Hank Kerschen, Sirisha Manyam, Mike Horejs, and I wanted to further investigate mockingbirds' abilities to discriminate 3 sticks versus 4 sticks. We developed a new opaque feeder designed to allow the mealworms to be released from only one side. in this way if the bird made the wrong choice it had to pull all the sticks from the correct side in order to get the food. You can see our abstract and presentation from the 2005 NCUR meetings. Below is a video demonstrating the new feeder:
Extra-pair paternity |