|Institution:||University of Saskatchewan|
|Keywords:||cattle; calves; latrine; training; learning|
|Full text PDF:||http://hdl.handle.net/10388/ETD-2014-12-1855|
This thesis explored the feasibility of training cattle to eliminate in specific areas of a barn and investigated cattle’s ability to generalise knowledge between different locations. In Chapter 2, all incidences of urination and defecation were recorded by group-housed female Holstein calves across 144 h. There were substantial differences between individual calves in the mean daily frequency. Calves urinated and defecated most frequently during daylight hours when they are more active and the location of voiding was likely related to the amount of time areas were occupied. In Chapter 3, calves were trained to urinate in a specific location via classical or operant conditioning. Classically conditioned calves were held in a stall for a set time and given no punishment or reinforcement upon urination, whereas calves in the operant treatment were immediately rewarded for urination in the stall. Classically conditioned calves did not urinate more than controls. Calves trained using operant conditioning had a higher frequency of urinations in the stall than their controls but did not seem to generalise this association; failing to urinate more than controls when tested again, 5 months later, in a new location (Chapter 4). The use of visual cues may be an effective way of helping cattle to generalise previously learned associations to a new location or context. Two experiments were conducted (Chapter 5) to investigate whether prior exposure to colour cues improves calves’ performance in a Y maze colour discrimination task. In Experiment 1, either both side and colour or colour alone predicted the location of milk reward in a Y maze. Our results suggest that calves overlook colour in the presence of more salient cues, such as location. In the second experiment, calves were first classical conditioned to associate coloured signs with presence of absence of milk (colours were randomised for controls) before testing in a Y maze discrimination task. Nine out of ten classically conditioned calves, but no control calves, achieved the learning criteria. Classical conditioning can be used to rapidly train cattle to follow colour cues and generalise these associations to new contexts or locations.