Discriminating spontaneous locomotor play of dairy calves using accelerometers

摘要

Play behavior is a promising welfare indicator in dairycalves because it decreases in negative situations suchas pain or hunger and increases in positive contextssuch as in appropriate social environments. Directlymeasuring play is time consuming because it is performedin irregular bouts and can be inconsistent overdays. To facilitate automatic recording of play, previousstudies fitted triaxial accelerometers to the hind legs ofcalves and measured the velocity of movements in largearenas; high correlations were reported between verticalaxis peak duration and the duration of locomotor play.The current study aimed to validate accelerometers forrecording spontaneous locomotor play in calves’ homepens over longer periods. Data were collected from 48Holstein Friesian calves, housed in groups of 3 in pensof 10 m~2, at either 4 or 8 wk of age. Acceleration at thevertical axis of the hind leg was recorded at a rate of 1Hz. One active time period for each calf was randomlyselected (mean duration ± standard deviation = 34 ± 9min). From video of the corresponding time period, thefrequency of locomotor play events, consisting of run,turn, and buck/buck-kick, was recorded using behaviorsampling. Combined counts of play events were highlycorrelated (Pearson r = 0.91) with counts of accelerationpeaks. However, for calves with higher levels of locomotorplay, this method underestimated the extent ofplay. Alternatively, run, turn, and buck events obtainedfrom video were transformed by creating intervals of 10s and then classifying each 10-s interval as comprisingevents of play (“play”) or not comprising events of play(“no play”). The corresponding accelerometer data forall 10-s periods, equaling 10 consecutive readings each,were classified into play or no play by using quadraticdiscriminant analysis; 79% of periods with locomotorplay were correctly classified. Counts of observed playintervals correlated with the counts of play periodsfrom accelerometers (r = 0.87), but the discriminantanalysis consistently overestimated play. In conclusion,accelerometer measurements at 1 Hz (in 1-s intervals)and at the vertical axis cannot be used alone to exactlyquantify absolute levels of locomotor play in the homepen. However, counts of peak accelerations can providea rough estimate of inter-individual differences in playevents, and discriminant analysis can be used as a proxyfor one-zero sampling of inter-individual differences inlocomotor play.