Confiability among inexperienced observers of rest-activity behaviours of (Macaca arctoides) in exterior captivity.

The behavioral sciences rely to a great extent in the sensory perception of observers, Recently, the American Journal of Primatology devoted an entire issue to the state of art of research on the activity-rest periods of non human primates, finding that the use of high-sensitive video recording equipments is a non obtrusive option to study behavior in socially established groups. In orden to perform quantitative studies it is necessary to assess the behaviors observed by different observers using reliability indexes, in the following paper four resting behaviors related to sleep patterns in an group of stumptail macaques housed outdoors were defined: 0 off (outside of the camera focus); 1 resting with mioclonus; 2 resting without mioclonus; 3 transition; and 4 awake. Three naive observes from a nightly video recording of a single macaque evaluated these behaviors. After a first trial the given definitions of behaviors were discussed obtaining an operational definition for each one. Afterwards, a B-hours video recording was observed, and the interobservers reliability was assessed by means of Cohen's Kappa. The following results were obtained: Kappa for behavior 0 was of 0.943; 0.785 for behavior 1; 0.857 for behavior 2; 0.590 for behavior 3; and 0.859 for behavior 4. These data provided a measure of the easiness in recognizing and assessing each behavior.