Training in a Laboratory Environment: Methods, Effectiveness and Welfare Implications for Two Species of Primate

Abstract

The use of Positive Reinforcement Training (PRT) for co-operation during routine husbandry and laboratory procedures is widely advocated as a means of promoting the welfare of nonhuman primates. However, while research originating in US zoos provide qualitative descriptions of how PRT may be used in the training of a wide variety of species, quantitative data and evidence to support the view that PRT reduces stress predominately comes from laboratory studies of primates whose training may have used other methods. Despite official guidelines, training is rarely carried out in the UK and the educational and wider organisational structures concerning training, present in the US are largely absent. The techniques used in the UK were assessed through detailed observations recorded when four stump-tailed macaques were trained to co-operate during venipuncture. Data recorded during training sessions showed that although food rewards were given, their delivery was slow and inconsistent. A certain amount of coercion was used which violates a principle of PRT which states that co-operation should be voluntary. The macaques showed increasing resistance to the process and a mild but detrimental effect on the subsequent behaviour of the study animals. When training resumed 18 months later there were considerable improvements in the techniques used. The macaques showed a greater willingness to participate and there were no significant changes in their behaviour when training days were compared to those when training did not take place. The behaviour of the macaques during venipuncture was judged to be arising from engineered compliance rather than voluntary co-operation. However, it was concluded that the technique observed, if carried out correctly, was a reasonable compromise between forced restraint and voluntary co-operation given the paucity of evidence showing the effectiveness of PRT for invasive procedures. However, it was also concluded that the use of coercion should be recognised and provide a focus for future refinement. The effectiveness and welfare implications ofPRT was assessed through the training of common marmosets to target and allow in-homecage weighing and to provide urine samples. It was found that the trained animals perfonned reliably and that time invested in training could be recouped through faster data collection. Following a period of training or increased positive contact with humans, observations of marmoset behaviour showed a decrease in stress related behaviours and an increase in allogrooming supporting the view that improved relations with humans had a beneficial effect. Following exposure to a mild stressor, trained marmosets showed no elevation in levels of urinary cortisol or stress related behaviours. Untrained animals showed increased levels of locomoting and selfscratching following exposure to the same stressor. It was concluded that PRT successfully reduced the stress associated with the presence of, and manipulation by, humans. Final recommendations were that training can promote the welfare of nonhuman primates and should be used in UK laboratories to a greater extent than is currently the case. However, the lack of educational opportunities for animal trainers in the UK needs to be addressed. It was also recommended that in light of the growing evidence showing the benefits that can arise from training and good relations with humans, the zero-handling policy practiced in many UK zoos should be reassessed.