Drinking behavior in nursery pigs: determining the accuracy between an automatic water meter versus human observers

Area of Science:

• Animal Behavior
• Agricultural Science
• Measurement Technology

Background:

• Accurate long-term behavioral event recording is crucial but often labor-intensive and costly.
• Traditional human observation methods for behavioral studies can be inefficient.
• Development of automatic devices for recording behavioral events offers a potential alternative.

Purpose of the Study:

• To evaluate the accuracy of human observers using Observer software (OBS) in recording pig drinking behavior.
• To compare OBS accuracy against an automatic water meter (WM) for time spent at the waterer and visit frequency.
• To assess the reliability of traditional observational methods in behavioral research.

Main Methods:

• Individually housed nursery gilts were monitored using video cameras and an automatic water meter (WM).
• Drinking behavior was recorded by human observers using Observer software (OBS) from video files.
• Data from OBS and WM were analyzed using nonparametric methods and general linear models.

Main Results:

• The time of day significantly affected drinking behavior frequency and duration, irrespective of the recording method.
• OBS significantly underestimated the number of visits to the waterer (3.48 visits/h) compared to WM (4.94 visits/h).
• OBS significantly overestimated the time spent at the waterer (22.6 s/h) compared to WM (13.9 s/h), with weak predictive relationships (R² = 0.56–0.69).

Conclusions:

• The traditional Observer software (OBS) method for quantifying pig drinking behavior is potentially misleading.
• Automatic water meters (WM) provide a more accurate assessment of drinking behavior frequency and duration.
• Further validation of observational methods is necessary for accurate quantification of behavioral events in animal studies.