The problem with punishment…April 10th 2014
As a society, we focus quite a lot on punishing unwanted behaviours. Whether it’s the use of a Naughty Step, a speeding fine, or even prison – the intention is to decrease the probability of a behaviour recurring in the future. Humans are complex creatures. We have the ability to learn right from wrong. We are able to be educated in why our actions may be unfavourable. We have morals. However, animals are completely unable to think like us. Their brains have a largely reduced cerebral cortex – the part of our brain that is responsible for higher levels of thought, language, human consciousness and reasoning.
Punishment in humans is often delivered sometime after the unwanted behaviour – speeding fines or prison sentences, for example. We are able to understand that our behaviour at a certain time was undesirable and that there is a consequence to that behaviour. But animals are unable to associate a consequence that is delayed from the action. They live in the here and now. Unless the timing of the reward or punishment is given within ONE SECOND of the behaviour, they are unable to link their behaviour with the consequence.
Therefore, any delay in delivering a punishment means that whatever your animal is doing the second you do deliver the punishment will be associated with something aversive.
For example, your dog toilets in the house while you are out. You come home and stand in the poo as your dog runs up to greet you. You shout at it for toileting in the house. You have successfully punished your dog’s greeting behaviour. Even if you attempt to re-form the association by showing your dog the poo and telling it off, the dog will only associate the presence of poo with being told off – they are unable to associate their action of doing the poo with being told off.
Another example – you have a hidden electric fence around your garden and your dog wears a collar that gives it a shock if it attempts to cross the boundary to prevent it running onto the road. The intention is that the dog is punished whenever it attempts to cross the boundary. However, the dog attempts to cross the boundary at the same time as a cyclist passes and the dog gets a shock. The dog may now associate cyclists with electric shocks. The likelihood is that this dog will now be fearful of cyclists.
Most problematic behaviours, including aggression, develop because the animal learns to show an effective response to a perceived threat. The Fight or Flight response is a physiological reaction that occurs in response to a perceived harmful event. If your pet is punished for showing an aggressive or avoidance response, it will simply perceive the situation to be scarier than it thought it already was. This means that in the long run, the aggression will escalate as the animal becomes more fearful, or the avoidance response may intensify to loss of bowel control, for example.
For punishment to be effective, it has to be at an extremely intense level, otherwise removal of the punisher actually results in an increase of the initial unwanted behaviour, before returning to the original level before punishment was introduced. The correct intensity is difficult to define because it is subjective – what may be almost insignificant to one animal could be distressing to another, depending on species, breed, individual nature (i.e. sensitivity and pain threshold), previous experiences, and motivation at the time. The punishment must be aversive enough to produce such physiological and behavioural responses so the animal will not engage in the unwanted behaviour again. Inevitably this will initiate anxious and fearful long-term responses.
Unpredictability from the animal’s point of view in receiving punishment causes chronic stress which can effect the immune response. It also causes confusion about what behaviour is required. Imagine that you needed to learn a new behaviour as a new employee, but in order to teach you this behaviour, your boss only shouted at you when you did the wrong thing. You might try a whole range of different possible responses, but may never identify the exact thing that they wanted you to do. Where owners rely mainly on punishment for inappropriate behaviours, it is very difficult for an animal to work out what it is supposed to do. As would also happen to you in your workplace, animals will tend to either end up becoming very frustrated and show a behavioural consequence of this emotion, such as aggression, or give up entirely and stop trying any behaviours at all.
5) Root of problem
Punishing a behaviour does not alter the motivation for the unwanted behaviour, resulting in control of the symptoms rather than treatment of the cause and preventing an animal from exhibiting its normal behaviour pattern (contrary to Section 9 of the Animal Welfare Act 2006). As motivations behind the unwanted behaviours have not been removed, they may be expressed by other undesirable behaviours – the suppressed motivation may be re-routed into other undesirable behaviours.
This is analogous to a dog with diarrhoea messing up an owner’s cream carpet. Treating the symptom might involve putting a cork up the dog’s bottom to prevent the expression of the diarrhoea. But this isn’t dealing with the cause of the diarrhoea – is it a bacterial infection, a blocked intestine, a virus? The cork might appear to be working as the carpet is staying clean, but diarrhoea is gradually building up behind the cork until one day it cannot hold anymore and there is a massive explosion of diarrhoea all over the carpet. This is the same ‘ticking-time-bomb’ that is seen behaviourally where a dog is being treated symptomatically, rather than dealing with the cause of the problem.
Punishment can cause an increase in anxiety and suppression of other behavioural responses resulting from the animal being unsure of which behaviour is likely to be punished. The animal may become behaviourally inhibited, fearful, frustrated or aggressive. This is also known as Learned Helplessness, which literally means that the animal has learned that there is no point in trying to express any varieties of behaviours due to fear of being punished. This decreases the animal’s ability to learn.
Since training techniques are widely used that do not require the use of punishment, there is no need to use techniques which impact negatively on the welfare of animals. Of course unintended associations, due to poor trainer timing, or the chance association with another stimulus, occur as frequently with reward based training as it does with punishment techniques. However, the long term consequences of these ‘reward mistakes’ are much less serious than when punishers are associated with unintended stimuli. Avoidance responses to things that are perceived as aversive are likely to be long lasting and resistant to change compared to those occurring as a result of positive reinforcement.
Be aware of any language that is used to support or mask the use of punishment such as “interrupting behaviour” or “being pack leader”. Remember that if the intention is to decrease the expression of a behaviour, punishment is being used. The fallout varies depending on the individual’s sensitivity and the severity of the aversive used, but whether it’s a rattle can or a prong collar, the risks are there. Contact a reward-based force-free behaviourist for more information and to help you reward alternative behaviours.