This is an interesting read but these days, I feel, somewhat of a blind alley of an argument designed to tie people in knots! The writer is absolutely right that when we are with animals we can never be sure what is altering their behaviour - we can only see whether it increases or decreases in frequency. He does make one assumption though that is a tad worrying for someone who is a dog trainer - and that is that animals work for food because they are hungry and that giving them food in an effort to increase the frequency of behaviour could be negatively reinforcing if it removes hunger. He should read up on the research on Contra-Freeloading! Animals that are not hungry will work for food even when there is free food available in plentiful supply. You could say the same about scratching an itchy animal. Is the scratching going to reinforce the behaviour it immediately follows because it relieves an uncomfortable itch or because the animal finds it pleasurable (assuming it does - not all horses like to be touched - the same as humans)? Can people who are not achy and stiff still enjoy a massage because it makes them feel good even if they have no aches and pains to relieve? Does giving the dog a toy to play with involve adding something - the toy - or removing something - the frustration at having the toy withheld as incentive to perform the behaviour? The same is true for negative reinforcement and positive punishment. In order to form a behaviour, traditional, classical and natural horse training methods all involve applying (adding) an aversive stimulus to get behaviour to happen, so that they can remove that aversive to reinforce the behaviour. So arguably we could say that this involves positively punishing one behaviour to get another. We punish walk to get trot, we then leave the horse alone while he is trotting (we relieve him from the aversive applied to cause the trotting) and then we add an aversive or warn that we will do so again to get him to stop. Forming behaviour in those systems involves applying aversives or conditioned aversives (things the animal has learned to fear as they predict the onset of an aversive for non-response) to cause behaviour to happen and that inevitably involves stopping another behaviour. Which is why it is actually a pointless argument and one that most advanced animal trainers have moved on from. These days the most modern animal trainers are more interested in the newer sciences - such as Affective Neuroscience - than they are behaviourism. Jaak Panksepp is the man to really read on this - his very accessible work has shown that the brain processes stimuli and events in two main ways - the mammalian brain either perceives things as rewards (appetitives - things we actively seek more of) or as punishments (aversives - things we actively try to escape or avoid). Anything that involves an experience that is aversive (the animal experiences a stimulus or event that it would rather not - a stimulus or event that it values when it is over or a situation or event that if finds frustrating, that makes it angry or fearful or that causes escape or avoidance behaviour (such as is employed by these common horse training systems) will produce an emotional response in the brain that is unpleasant. It feels bad. When the experience is over the animal feels relief, which is not the same emotion as is experienced on attaining something appetitive and can be measured - and has been by Jaak Panksepp and others working in this field of Affective Neuroscience - the neuroscience of emotion. Any experience that the animal perceives has appetitive value - a stimulus or event that when it sees it or anticipates it, it wants more of and will work to gain or acquire, produces feelings of excitement and anticipation that can be measured in real time in the brain in mammals. These days we are more interested in how the animal feels and in avoiding using procedures in training that produce bad feelings. Bad feelings can be created by any use of aversives to form or reinforce or punish behaviour by their addition or removal. Bad feelings can also be created if aversives are used to drive or keep animals away from food or other resources they need for survival or if food is used to tease animals. Bad feelings can also be experienced if food is withheld. Animals who are repeatedly driven away from food using aversives can come to fear feeding time itself and show calming signals in that situation. It is never appropriate to judge a method of training by the standards of those who do it badly, but without doubt the use of aversives in training - whether to form behaviour so that we can reinforce it, or to interrupt unwanted patterns of behaviour, does not make animals feel good, regardless of whether we consider that to be positive punishment or negative reinforcement. Training hungry animals is not good practice either if you want relaxation and calmness about the training. While there are folks out there who want to tie us all in knots with the interconnectedness of the quadrants of operant conditioning, it is much simpler if we focus on asking ourselves whether the animal is working first and foremost to avoid something he does not like (which makes him feel bad) or to gain something he does (which makes him feel good). If we look at the dog - imagine the kind of emotional reaction you get from a dog who loves going for walks when the owner picks up the dog lead and contrast that with the kind of emotional response you get from a dog that hates baths when it hears the bathwater running. He will get excited at the dog lead because he values what follows and wants more of it. He will get upset when he hears the bath running, and may try to hide, because he values baths most when they are over and would rather avoid having them altogether. All mammals are the same. All we have to ask ourselves about whether we are training ethically, is, Is my animal behaving to avoid me applying, or to avoid experiencing something he does not like and would wish to escape? Or is he behaving to gain something that he is anticipating with calm excitement? One feels bad, the other feels good. Thats really all we need to be thinking about!
Posted on: Wed, 24 Sep 2014 13:58:42 +0000