What better way to celebrate my tenth blog post (it’s still early days for Kat Welfare Matters!) than by taking part in a blog party organised by the wonderful Zazie of Companion Animal Psychology?
The blog party theme is something I wholeheartedly support: ‘training for rewards’. Reward-based training, or positive reinforcement as it is also known, is a form of ‘operant conditioning’. This is just a fancy way of saying it is a training method that involves an animal learning to do something because of the consequences its behaviour has. When animals behave in a certain way they are ‘operating on’ their environment. The positive or negative consequences of that behaviour affects the frequency with which the animal will perform that behaviour in the future. In other words, if the animal does something (e.g. the cat sits) and the consequence of this is positive (e.g. a yummy treat that the cat loves) it results in them doing that more in the future. We can add commands as well, but the basic idea is that the animal does something and we provide the ‘consequence’ of their behaviour.
There is a lot of great evidence for the benefits of reward-based training. I strongly suggest you check out the list of reasons to go positive at Companion Animal Psychology. It makes sense that rewarding animals for a behaviour is more likely to result in better trained animals than punishing them. The escape/avoidance approach to training that was developed in the 1940s relies on negative reinforcement and punishment; animals escape the aversive consequence by changing behaviour. In contrast, positive reinforcement is a ‘modern’ approach to learning and training, as it provides specific information to the animal about the exact behaviour that is required. In doing so, it enhances the human-animal bond.
Linking reward-based training and animal welfare science using the concept of positive affective engagement
Training for rewards emphasises the idea of giving an animal something they find rewarding or enjoyable. These rewards may be different for different animals, but should result in the animal experiencing positives emotions. This links directly with the concept of positive experiences, or affects, that is being increasingly emphasised in animal welfare science.
While animal welfare scientists used to concentrate their efforts on the negative experiences of animals, we are looking more and more at the potential for positive experiences. Back when animal welfare science began, people were worried about how badly animals were being treated and their potential negative experiences – especially animals confined in intensive production systems. But now that we have learnt more about how to keep animals and what they might need to minimise their negative experiences, we are able to look at providing them with opportunities for positive experiences e.g. play and/or exploration. Ensuring animals are receiving good nutrition (food and water), that their environment is suitable, and that they are healthy typically only moves them from negative/poor welfare towards neutral. None of these ‘survival-related’ provisions  are enough to result in overall positive welfare. However, when we start to look at how animals are behaving and provide them with opportunities to behave in ways they find rewarding, we can start to move them towards an overall positive welfare state.
Rewarding animals for a behaviour that we want them to do and that they enjoy could result in heightened states of positive welfare. Hence asking a dog to ‘sit’ becomes ‘playing sit’; the dog starts to associate the command and their response with a positive reward (e.g. food and/or your happy voice). We know that play is a positive experience for most animals, so why would we not want to make training positive for them?
‘Positive affective engagement’ is the experience animals may have when they respond to motivations to engage in rewarding behaviours. The concept includes all experiences that are positive . In reward-based training, the concept could include the state of engagement that results from an animal’s goal-directed, energised performance of a learned task that it enjoys. The enjoyment may result from the task itself and/or be due to the associated reward. A 2015 review by Mellor  emphasises the scientific support (via ‘affective neuroscience’) for the existence of positive experiences and their assessment using behaviours we observe in animals. In other words, training for rewards is not only a very effective training method, but there is good evidence to suggest that it could improve the lives (or welfare) of our animals.
- Stafford, K., ed. The welfare of dogs. Animal Welfare, ed. C. Phillips. Vol. 4. 2006, Springer: Dordrecht, The Netherlands.
- Mellor, D.J. and N.J. Beausoleil, Extending the ‘Five Domains’ model for animal welfare assessment to incorporate positive welfare states. Animal Welfare, 2015. 24(3): p. 241-253.
- Mellor, D.J., Enhancing animal welfare by creating opportunities for positive affective engagement. New Zealand Veterinary Journal, 2015. 63(1): p. 3-8.
- Mellor, D.J., Positive animal welfare states and encouraging environment-focused and animal-to-animal interactive behaviours. New Zealand Veterinary Journal, 2015. 63(1): p. 9-16.