Using a Free Online & Citizen-Science Project to Teach Observation & Quantification of Animal Behavior
Voss, Margaret A., Cooper, Caren B., The American Biology Teacher
There are logistical and conceptual challenges to teaching the techniques of behavioral observation that are fundamental to the field of behavioral ecology Logistical obstacles, such as housing captive animals or observing wild animals, can prevent instructors from implementing the many published laboratory exercises designed to teach students to observe, describe, and quantify animal behavior (e.g., Ploger, 2003). Conceptually, the complexity of behavioral repertoires often leads inexperienced students to inappropriately confound their observations through vague and subjective descriptions. Thus, teaching objectivity in the observation of animal behavior is a cornerstone of behavioral ecology with at least three components: understanding sources of observational bias, gaining familiarity with observational techniques, and developing proficient skills to accurately define and quantify observations.
We designed an introductory, college-level animal behavior laboratory to overcome these conceptual and logistical obstacles to teaching quantification of animal behavior. The exercise uses time-lapse photography in a free, online citizen-science project (CamClickr) developed and maintained by Cornell University's Lab of Ornithology In this lesson, students learn about multiple ways to describe behavior, create ethograms from observations gathered while participating in CamClickr, and use their ethograms to generate and test hypotheses about the evolution of observed behaviors. By using series of still images of behavior at discrete periods, the students are limited to structural descriptions of behavior. This limit is useful in learning fundamentals, as it averts the students' attention from the initial distraction of attempting to determine the consequences of observed behaviors. The exercise teaches them, through experience, the concepts of structure, consequence, and relational descriptions of behavior. The time-lapse images are also a useful way to illustrate the differences between scan sampling and focal-subject sampling, two common methods used to obtain standardized observations.
Although we have structured this exercise for an introductory college course, its scope and content are also appropriate for the ecology unit of a high school biology class. Ethograms depict patterns of animal behavior; as such, they are useful tools to permit students to generate--and, with appropriate replication, test--hypotheses about their observations. For example, students in our classes often notice differences in nestling provisioning rates between bird species. They then hypothesize about why provisioning rates might differ between species. Their hypotheses often focus on differences in the types of food delivered to the nestlings, the time of year the nestlings are hatching (late spring vs. early summer), and the body size of the species observed. In some cases, the students narrowly define hypotheses that can be tested by further examination of the time-lapse data combined with a little background research on the species in question (e.g., nestling provisioning rates for insectivores that specialize on aerial prey are much higher than those for insectivores that specialize on terrestrial prey). In all cases, the students' observations generate further inquiry and independent research. This inquiry-based approach fits nicely with both the standard of developing the "skills necessary to become independent inquirers about the natural world" (National Research Council, 1996) and the benchmark of developing critical-response skills (American Association for the Advancement of Science, 2009).
* 1 computer per 1-3 students
* Internet access
* Example observation data sheet (provided)
* Spreadsheet and graphing software such as MS Excel
The activity is done over two 3-hour laboratory sessions to allow enough time for the students to (1) engage in class discussion of key concepts in behavioral observations, (2) become familiar with the CamClickr project, (3) generate testable hypotheses, and (4) collect observations from CamClickr and create and use ethograms. …