Observer Distance and Identity Effects on Mixed-Species Flocks of Parids

Abstract

When studying non-human animals in the wild, the presence of a human researcher can potentially affect the behavior of the animal being observed. This phenomenon is known as the “observer effect,” and it has been demonstrated across a wide variety of species. Much of the research performed in the field on Carolina chickadees (Poecile carolinensis), tufted titmice (Baeolophus bicolor), and white-breasted nuthatches (Sitta carolinensis)—birds that flock together in the winter months throughout much of the eastern United States—is done with an observer present, but no research has explicitly tested the observer effect with respect to these flocks. Thus, using feeders supplied with seed, we measured seed-taking rates of 33 flocks of these species with a human researcher standing at distances from 3 to 20 m. We found that all three species were less likely to take seed when the human observer stood 3 or 5 m from the feeder, and chickadees were less likely to take seed with the observer at 10 m, compared to 15 m or the baseline condition of the human observer standing 20 m away. A follow-up experiment was conducted to assess the difference between the number of seeds taken with an observer at 5 and 20 m compared to the number taken when an observer was so far away as to be effectively out of sight (40 m). This additional experiment indicated that all three species took significantly fewer seeds when an observer was at 5 m, while the number of seeds taken at 20 and 40 m was not significantly different. Additionally, results from both experiments suggest potential effects of the specific observer standing at the different distances. Our results suggest that researchers studying these animals in the wild should conservatively maintain a minimum observer distance of 15 m so as not to affect the behavior of any of the three species within these mixed-species flocks and that researchers be aware of potential idiosyncratic observer effects in studies. The presence of a human observer can have a wide range of influences on non-human animal behavior—some quite long-lasting—and so should be carefully considered in observational and experimental research in ethology.