Inga MacKellar
We live in a technology driven age where people’s expectations are demanding and instant solutions and responses seem mandatory. We expect fast, instant food; get annoyed when we get stuck in a traffic jam; become irritated at the supermarket checkout by the person who takes forever counting out their money; we get cross with the dog when it’s ‘dawdling along’, sniffing at every lamppost on its walk. I also challenge any computer user to deny that at some stage they have not become frustrated with it for being too slow! Countless consumer programmes encourage us to complain about poor service and goods and to generally be less tolerant. This demanding lifestyle will invariably have a toll on our pets’ lives – be it that we have less time for them or that we have high expectations and want them to respond to our wishes instantly.
There are many types of pet, but the dog remains one of the most popular and the species most frequently referred for behaviour problems (APBC). If we are to have a mutually enjoyable co-existence with our dogs, it is important that we make time for them, make an effort to understand them and think about how we interact with them.
I have found that many owners often resort to punishing their dogs for actions that they perceived to be inappropriate or unwanted, or perhaps for not obeying a specific command. In many instances, this was a result of a lack of understanding of dog behaviour and frustration. Due to the fact that dogs are frequently treated as humans (anthropomorphism), many owners use ‘human’ rearing and disciplining methods.
In a study of dog owners in the Netherlands, owners were asked how they would react if their dog displayed an innapropriate behaviour. The problem behaviours cited were categorized into eight situations: fearful, aggressive/dominant, ‘mating’, emotional, bored/lonely, eating, unruly and disobedient. It was found that in most of these situations owners demanded compliance and reacted by addressing the dog severely. Punishing the dog was specified as the first option to use when the dog misbehaves and many of the situations elicited feelings of irritation and anger in the owners together with disappointment and powerlessness (Ben-Michael). Owners frequently tell me ‘He knows that he’s done wrong as he looks guilty’ and often feel hurt and let down that their much loved and cosseted dog is ‘knowingly’ disobedient and defying them. On further investigation these owners have, invariably, punished their dog for what they regard as ‘bad’ behaviour. They would automatically smack a dog if it had urinated in the house, shout at it to stop jumping up or, thankfully more rarely, give it an electric shock from a special collar to stop it wandering out of the garden. One of the most common punishments is reprimanding the dog on returning to its owner after it has ‘run away’ on a walk. Repeated use of punishment has not resulted in any improvement in behaviour and, in some instances, had resulted in problem behaviour getting worse. In most of these cases, considerable improvements were made after the implementation of positive reinforcement coupled with negative punishment.
In order to understand how a dog might respond in a given situation, it is necessary to have some knowledge of how it learns. In the 1920’s a Russian scientist, Ivan Pavlov, found that he could make dogs salivate even though food was not always present. He achieved this by letting the dogs hear the sound of a metronome just before they were fed. Over time, the dogs would salivate to the sound of the metronome even if the food did not appear. He could keep this up indefinitely as long as he sometimes gave food after the sound (Pavlov). This discovery is the basis of classical conditioning i.e. a stimulus (e.g. the metronome), that has no apparent importance in itself, becomes significant because it predicts that something will follow (e.g. the food) and triggers an involuntary action (e.g. salivation). Classical conditioning can be used to trigger a response over which an animal has no control, including the process involved in emotional states. For example, if a dog gets excited about going for a walk, but is never taken on a lead, then the lead will have no significance and will not trigger excitement if it is produced. However, once a dog has been taken for a walk on a lead a few times, the sight or sound of it will trigger excitement in the dog, even if it is not taken for a walk.
Dogs also learn through operant conditioning. In this instance a dog will learn that if it performs an action, that action will have a particular result. Simply put, the dog learns ‘If I do this . . . this happens’. There are four possible outcomes: positive reinforcement, negative reinforcement, positive punishment and negative punishment. Where the outcome of the action is something pleasant or beneficial, the dog is more likely to repeat that behaviour – the outcome positively reinforces the action. Conversely, if the outcome is something unpleasant, the dog is less likely to perform the behaviour – the outcome is aversive – it could be said that the dog has experienced positive punishment. Negative punishment involves the withdrawal or absence of a previously rewarding experience when the action is performed, which will decrease the likelihood of the behaviour being performed again. Negative reinforcement involves the withdrawal or absence of an aversive experience when the action is performed, resulting in an increase in the likelihood of the behaviour being performed.
On the face of it the solution seems simple – if your dog does something that you don’t want it to do, make the outcome aversive and it will not do it again! But behaviour is not that simple! What was the dog’s motivation for the behaviour in the first place?
If, for example a dog is behaving aggressively towards a neighbour because it is frightened of him – reacting in an aversive manner, such as the owner shouting, hitting the dog or pulling on its collar, may make the situation worse as this may further increase the dog’s arousal. In some instances an owner’s response can also act as a reward for the dog by giving it attention. A dog’s fear may also be increased because something ‘nasty’ happens to it when the neighbour is present (being punished)! The next time the dog sees the neighbour it may be more frightened and react in a more aggressive way to try and scare off the neighbour . . . it gets reprimanded again . . . its fear further increases . . . and so the spiral continues.
Unfortunately, all the normal and usual human reactions to aggressive dogs tend to increase it. In situations where a dog shows signs of aggression, trying to calm and comfort it reinforces the aggressive reaction and fear if present. Punishment increases the existing fear and thus leads to more aggression (Jones-Baade). However, by employing positive reinforcement methods the dog can be taught that it is rewarded when it does not react to the neighbour and, over time will start to associate the presence of the neighbour with something pleasant happening (a process called desensitisation and counter-conditioning).
Let’s look at the common scenario of the dog that has ignored the recall command on a walk – when it eventually returns to the owner it is severely reprimanded. In essence, the dog has just been positively punished for returning to its owner! Dogs make an instant association with a particular action therefore, the dog does not understand that it is being punished for not responding to a command it heard ten minutes ago – it is just aware that it has approached its owner after having a nice time running around, sniffing in the forest, and the owner is immediately ‘aggressive’ towards it!! Because approaching the owner has been aversive – the dog will then be more wary of returning to the owner next time,. . . which makes the owner more frustrated and cross and results in the dog being told off more vehemently. A stage is eventually reached where the dog is reluctant to come back to the owner at all, so the owner resorts to walking the dog on the lead. If, rather than reacting with positive punishment, the owner had positively reinforced the dog with, say, a food reward when it had eventually returned, the dog would have learnt that coming back to its owner meant something pleasant and would, in future, come back more eagerly when called.
Some owners believe that a dog should be instantly obedient because it respects them and that giving it rewards is bribery. However, bribery is very different from positive reinforcement. Bribery entails the use of a lure to encourage a particular action – positive reinforcement entails the giving of a reward after the action has been performed – a subtle but important difference.
Most people immediately connect positive reinforcement, or reward training, with food. However, dogs are motivated by many things and what is rewarding depends on the drives, needs and temperament of the dog as well as the situation at the time. Obviously, where a dog is highly motivated by food, it makes sense to use food rewards. But if the dog has just eaten its dinner, it will probably be less motivated by food and a toy or verbal praise may be more appropriate. Owners are sometimes concerned that their dog will become overweight if they utilise food rewards. This need not be so – a simple way to overcome this is to measure out the dog’s food for the day and use part of the daily ration for positive reinforcement. Many dogs are now fed on a complete dry diet which can easily be carried in pockets – ready to positively reinforce the dog. It is, however, important to have a variety of treats so that if the dog behaves particularly well, a more highly prized reward can be given. That is, there should be scales of reward. The better the performance of the required behaviour, the more appetitive the rewards should be. These can vary from pieces of chicken, ham and cheese to raw carrots and home-made liver cake. Touch, eye contact and reacting to the dog, as well as toys etc, can also be added to positively reinforce a particular action.
The breed of dog will play a part in its motivation and it is important for owners to know the origins of the breed as this will affect the dog’s behaviour and general character. For example terriers were initially used (and still are in rural communities), for hunting. Most are extremely reactive and like squeaky toys, which they delight in shaking. Border Collies, bred for sheep herding, generally enjoy chasing and running whilst Cavalier King Charles Spaniels, which originated as companion dogs, particularly enjoy physical contact. There are numerous breed societies available to the dog owner and the greater the knowledge of the breed the easier it will be for the owner to find the right motivation. However, there will always be exceptions to the rule!! For example, whilst it is generally accepted that Labrador Retrievers are highly motivated by food – there are some that are not! It is vital that owners are aware of what motivates their dog in order to use the appropriate method of positive reinforcement.
An important factor to consider with positive reinforcement is the type of reinforcement schedule used i.e. how often the dog is rewarded. Whilst Gross (1992) defined five types of reinforcement schedule (continuous reinforcement, fixed interval, variable interval, fixed ratio and variable ratio), the two most commonly used in training dogs are continuous reinforcement and variable ratio:
• Continuous reinforcement – where each desired response is reinforced e.g. every time the dog sits, it gets a piece of kibble.
• Variable Ratio – reinforcement is given spasmodically, so the reward is unpredictable e.g. the dog is rewarded for sitting once, but the next time it is rewarded after sitting three times, then possibly twice etc. – there is no set pattern to when the reward is given.
As it is important that the dog learns what is required of it, continuous reinforcement is used initially to teach the dog a particular command or action. Once the owner is confident that the dog has learnt what is required and it is responding each time, the rewards can then be given on a variable ratio basis. Variable ratio can be likened to gambling – the dog responds extremely well in the hope that it will be rewarded with a ‘win’. If variable ratio coupled with a variance in the value of the reward, (i.e. sometimes the dog receives a piece of kibble and sometimes it receives a ‘jackpot’ – a handful of chicken), the dog’s response will become even more reliable. This is because, sometimes, the dog is being positively reinforced and other times it is being negatively punished (it is not receiving it’s expected reward) which leads to the dog becoming frustrated, trying harder and subsequently being positively rewarded for the increase in its behaviour. (This frustration effect is covered in more detail later in the chapter)
An extremely effective method of positive reinforcement is through the use of clicker-training whereby the dog forms an association of reward with the sound of the clicker through classical conditioning. Once the association is formed, this becomes a highly enjoyable and stimulating method of training for the dog as, in essence, it initially has to figure out for itself what it is being rewarded for. With a clicker it is easy to ‘click’ at the precise moment that the required behaviour is performed. Verbal or other signals can then be added as the dog starts to perform the action and it will make an association between the two. Clicker training also helps the owner to perfect their timing because they have to be vigilant and watch their dog in order to ‘click’ when it performs the required behaviour. There are a number of publications now available on this effective and fun method of positive reinforcement/negative punishment training.
Clients often tell me that they don’t have time to be constantly rewarding their dog. However, many owners are masters of positive reinforcement – but don’t realise it! Imagine the following scenarios with the family dog, Ben:
• Ben wants to play, but his owners are busy getting ready to go out for the night and take no notice of him. Ben happens to spot some training shoes lying on the ground, picks one up and starts to playfully throw it in the air. One of his owners instantly reacts and starts to chase him. Ben finds that this is great fun and runs around the house ‘ducking and diving’ with the owner chasing him in a frantic manner. The owner chases Ben because he is desperate to get his expensive shoe back. Ben views this interaction as a game and through repetition he could learn that it is a good way of getting attention. The owner positively reinforced the dog’s behaviour of running away with the training shoe.
• An owner is sitting watching TV and Ben approaches and lays his head on their lap. The owner absent-mindedly strokes Ben’s head. Eventually, the owner stops stroking his head and Ben nudges the owner with his nose and they immediately respond by stroking him again. The owner has positively reinforced the dog’s attention seeking behaviour.
• The owners are eating dinner and Ben approaches the table. One of the owners glances at and makes eye contact with him, which encourages Ben to sit and look ‘lovingly’ at them. Ben continues to try and catch his owner’s attention by placing his head on the owner’s knee (which he knows has worked in the past) at which point, another member of the family may comment, ‘Don’t be mean – give Ben a bit of your steak’ to which the owner responds ‘I’m not supposed to feed dogs from the table . . . am I Ben?’, looks at him and slips him a bit of his steak. Ben has been positively reinforced for begging at the table.
• The owner returns from work, walks into the house and Ben rushes up and starts to jump up at them in excitement. The owner is delighted by this effusive show of affection and interacts with Ben by greeting and patting him. The owner has positively reinforced Ben’s jumping behaviour.
All of these scenarios have involved inadvertent positive reinforcement on the part of the owner and made Ben more likely to behave in a particular manner when the situation and similar situations are repeated. Yet, more often than not, when Ben has been lying calmly or chewing a bone (rather than his owner’s training shoes!), he has been totally ignored. Many owners tend to interact with their dog when it is, from their perspective, ‘misbehaving’ or instigating interaction, but ignore it when it is behaving well!
None of the types of behaviour mentioned in the previous scenarios are necessarily a problem and many owners happily accept them. However, imagine a scenario where a young girl comes into the house and Ben bounces up to greet her, as he has done so for many years, and knocks her over. The girl cries, the owners are mortified and immediately punish Ben by shouting ‘Bad Boy!’ and giving him a smack. Ben is totally confused because he has behaved in his normal way, for which he has always been rewarded, but he is now being told off. Ben is unsure of the situation so he resorts to behaving in a manner that he has previously been rewarded for – he jumps up more vigorously!! The owners, who are still attending to the crying child get cross with Ben and banish him to the kitchen. When Ben is eventually let out, he approaches his owners, cowering, tail between his legs and ears flattened. The owners then ‘forgive him’ as they believe he is looking ‘guilty’ because, to them, he looks as if ‘he knows that he’s done wrong’! Ben does not know that he has done ‘wrong’ – he has been stressed by his owners’ behaviour, he anticipates more owner aggression and the resultant emotional state of fear is manifested in his body language. This situation could have been avoided if Ben had been accustomed to being positively reinforced for acting calmly when the family came home rather than for leaping about.
So why not use positive punishment? Surely the dog needs to know ‘who is boss’? There are an ever increasing number of gadgets and ‘quick fix’ solutions available for owners with problem dogs and a considerable number of these involve the use of positive punishment, some confrontational (directly involving the owner) and some remote (the dog does not make a connection of punishment with the owner). These range from check chains to collars that automatically squirt citronella spray when a dog barks to collars that deliver an electric shock. All of these methods can be misused either accidentally, or intentionally, and can result in pain or other distress for the dog.
Another possible repercussion of aversive devices can be the development of fearful behaviour towards an unrelated object or environment. For example, if a shock collar is activated to stop a dog barking and at that same instance a child walks past, the dog may associate pain with the presence of that child and start to display fear aggression towards the child next time it sees it. I have seen a case where an owner had installed a special fence to stop her boisterous Collie wandering off (This device works by giving the dog a shock through a special collar when it gets close to a boundary). During the consultation it emerged that when the other dog in the house, a timid Collie, had the special collar placed on him he wouldn’t even go into the garden! One could argue that the device had worked, but to such an extent that one of the dogs was now frightened of going outside at all.
Whilst accurately executed positive punishment, implemented by someone who is aware of all the possible dangers, can be effective in stopping unwanted behaviour, most people do not have the expertise to implement it without potentially detrimental effects. The timing of positive punishment is absolutely crucial and a delay of its use can have profound implications for the effectiveness of the technique (Solomon, Turner and Lessac) and the welfare of the dog. The average dog owner rarely ‘punishes’ their dog immediately the behaviour is displayed – a prime example being that of the dog that urinates in the house when it is left alone and is then punished some hours later when the owner returns home. The dog will not make any association between the punishment and its previous urination, but it may start to make an association between the owner’s aggression and their return home. The problem may well escalate as the dog displays increasingly stressed behaviour manifested in its body language when the owners return. If it finds that this behaviour stops or prevents their aggression it could learn to use it to avert its owner’s ‘aggressive’ behaviour towards it on future occasions. If a dog is extremely stressed a possible consequence is urination and the dog may start to urinate as soon as it sees the owner – which would exacerbate the situation further!
Another point to consider is how severe should the punishment be? A verbal reprimand? The use of aversive tasting substances such as hot mustard or chilli sauce smeared on items to stop a dog chewing? A smack with a rolled-up newspaper? Hitting it with a stick . . . or worse? Positive punishment is intended to be aversive, but it is clearly undesirable to inflict more pain than absolutely necessary. But what is necessary?
If the punishment is too mild, it may be ineffective, and the problem may then not be remedied easily by switching to a stronger punishment. In a study, pigeons were punished with an electric shock for pecking at a key that produced food (Azrin, Holz and Hake). When the shock was set at 80 volts the behaviour was totally suppressed. When the voltage was set at 60 volts, however, it had little effect, and when its intensity was then increased in gradual steps, the pigeons still continued to peck at the key for food even when the voltage had reached 300 volts! If punishment intensity is set at low levels initially and then increased gradually it may prove ineffective because the dog adapts to the gradual increasing intensity. The optimum level of punishment is thus the minimum level that is effective in suppressing the behaviour – but choosing that level is a skill that few people can claim to have. However, even if they have it, is it necessary or desirable? What are the implications for the dog’s welfare and its relationship with its owners?
Cognitive and linguistic capacities have been found to play a role in determining human reactions to punishment, but dogs do not have linguistic capacities! A study by Cheyne (1969) found that in a group of children who were given a number of toys and then punished for playing with a particular toy, those children who had an explanation given to them as to why they mustn’t play with the ‘forbidden’ toy were far less likely to play with it in a future situation when compared to those children who were just told ‘No’ and not given an explanation. Sears, Maccoby and Levin (1957) found that mothers who made extensive use of reasoning, reported punishment to be far more effective than those who used punishment alone. Whilst many of us tend to speak to our dogs on a daily basis, and they will respond to certain words such as ‘Sit’ or ‘Down’, they do not understand what we are saying. Therefore, in a situation where a dog is being reprimanded, a reasoned explanation of why they are being punished will have no effect whatsoever and the dog will just become confused and may develop emotional responses related to stress and anxiety in other aspects of its relationship. In extreme cases, a dog that feels particularly threatened may resort to defensive aggression towards the owner in an attempt to stop the physical punishment and in anticipation of it. This can then result in an extreme breakdown of the owner-pet relationship, and in some cases, the euthanasia of the dog.
It is important to remember that, as detailed previously, if a dog is displaying fear aggression, positive punishment can lead to an increase in that fear and an increase in the intensity of the behaviour.
If a dog is being consistently positively-punished it can also result in an increase in stress. Animals (and humans), that are placed in a situation where they have to deal with a threatening situation will experience a number of physiological changes, one of which is the increased production of adrenal gland hormones. This physiological response is useful as a short term response in an aversive situation as it results in increased blood flow to the muscles and provides quick mobilization of energy for vigorous movement, but it is harmful in the long term. Prolonged increases in the level of glucocorticoids (cortisol) can damage muscle tissue, inhibit growth, inhibit the inflammatory response and suppress the immune system (Carl son). Animal research suggests that stress can encourage the growth of malignant tumours and exacerbate allergies and auto-immune diseases. Equally, prolonged stress can impair the learning process (Dweck & Licht). Therefore, if a dog is continually placed in a situation where it is being positively punished by its owners, its physical as well as its cognitive well-being is compromised.
So far I have talked about positive reinforcement and positive punishment. There are, however, two other learning scenarios to consider – that of negative reinforcement and negative punishment. Liebermann (1993) defined negative reinforcement as an increase in the probability of a response due to the removal of an aversive stimulus, and negative punishment as a decrease in the likelihood of a response due to the removal of a reinforcing stimulus.
To illustrate these two scenarios, let us consider the situation of two children playing – a boy and a girl. The girl is very spoilt and used to getting her own way. She starts constantly pulling the boy’s hair, telling him that she will not let go until the boy gives her a sweet. Here we have a situation where negative reinforcement is being utilised – the pain (positive punishment) will continue until the boy reacts in the required manner – relinquishing a sweet. This particular method of learning, negative reinforcement, is, obviously, not one to be recommended as its welfare implications are related to positive punishment.
On another occasion the girl continually pesters the boy for a sweet because this has always worked with her parents! Her pleas, cries and tantrums are incessant, but the boy takes no notice, she becomes increasingly frustrated and her behaviour escalates, she stamps, cries and throws herself on the ground but all to no avail and the boy continues to ignore her. Eventually, exhausted, she stops her tantrum and sits down calmly at which point the boy gives her a sweet. Although her tantrums had previously worked, in this instance, she will have started to learn that tantrums will not result in obtaining a sweet (negative punishment), but sitting calmly will (positive reinforcement).
Negative punishment involves what is termed the frustration effect where the subject – be it a person or a dog – when faced with a situation where he/she has previously been rewarded for their behaviour, will get increasingly frustrated and will try harder but can become angry or depressed if the expected reward does not materialise. There appears to be a strong emotional component – think of the angry man kicking a vending machine when his expected can of Coke doesn’t materialise despite frantically pressing all the buttons or the frustration that you feel when you turn the car ignition on and the engine is dead! You repeatedly turn the key and possibly start shouting at the car before you accept that the engine will not start. Obviously in these two examples, in reality, the behaviour would not be extinguished as the vending machine will eventually be replenished with cans of Coke and the garage will fix your car so that when you turn the key the engine will start again. However, if, for example, the vending machine was never replenished again, the man would eventually learn that kicking or repeatedly putting money in that particular vending machine was ineffective. As he would not receive the ‘reward’ of the Coke can, he would eventually stop trying – his behaviour would have been extinguished.
Unlike positive punishment (where an action results in an aversive outcome), negative punishment entails the absence or withdrawal of an expected reward. This method can be utilised to stop unwanted behaviour that has previously been either consciously or inadvertently positively reinforced. Negative punishment can be a difficult concept to grasp, I believe, because the terminology is confusing. The words negative and punishment both conjure up images of something aversive, whereas the concept could be described more simply as non-reward or the ignoring of a behaviour that was previously responded to.
To illustrate this, let us return to our scenario of Ben, who instantly leaps up at everyone who enters the house. How can we stop that behaviour?
In many instances owners will resort to waving their arms about and shouting at the dog to ‘Get down’ or, possibly, hitting it. The first will inevitably prove to be ineffective, because Ben is stimulated by the increased interaction (positive reinforcement) and will continue to jump up. The second action, hitting the dog, will confuse him and he will re-double his efforts to display the behaviour that he has learnt is rewarded by his owner – jumping up. This scenario continues on a daily basis with the owners becoming ever more frustrated and the dog increasingly confused. So what can be done?
Employing negative punishment will initially result in frustration but eventually lead to extinction of the behaviour. All people entering the house must completely ignore Ben when he leaps up – including no eye contact, ie. they need to behave as if Ben doesn’t exist. Initially, Ben will be frustrated as he will not receive the response that he expects and will increase his efforts i.e. the behaviour will get worse. (It is at this stage when I am explaining the procedure to clients that they interject ‘ But we’ve tried ignoring him and it doesn’t work – he just gets worse!’ When I ask how long they persevered with the ‘ignoring regime’, they will invariably tell me that the dog became such a ‘pain’ that they couldn’t ignore him any more and they gave up.) It is vital that the owners do not succumb to interacting with him as otherwise Ben will learn that if he leaps up for long enough and hard enough, he will eventually get a response. However, by completely ignoring him every time he is excited no matter how long his excitable behaviour persists for, and positively reinforcing the calm behaviour by greeting him, Ben’s excitable behaviour will eventually be extinguished and calm behaviour, which is differentially reinforced will be more likely to occur.
The problem behaviour will have been solved through the use of negative punishment coupled with positive reinforcement, without the use of any positive punishment and without damaging the relationship between dog and owner. Obviously, if Ben had been positively reinforced for sitting calmly when family or visitors arrived at the house from the outset, the problem behaviour would never have developed in the first place!
Negative punishment techniques can also be combined with conditioned avoidance, where the dog learns that it will fail in a particular action if a certain stimulus is present. Conditioned avoidance can involve the use of training discs (five small cymbals on a key ring). Once the discs have been properly introduced, the dog will associate failure (non reward) with the sound of the discs and will stop whatever behaviour it is displaying when it hears the sound. For example, a dog jumping up to get food from a kitchen work surface can be made to jump down and ignore the food by the owner chinking the discs or throwing them to the ground. However, training discs can be inappropriately used and are not recommended.
Positive reinforcement (coupled with negative punishment) can be successfully used in most situations to overcome problem behaviours. This can sometimes be difficult to imagine when you have a dog on the end of a lead pulling its owner over in an effort to bite the next door neighbour!
Aggression is one of the most common behaviour problems in dogs and one that owners frequently react to in an equally aggressive manner and positively punishing their dog for. Quite often, if the aggression is directed towards another dog or person outside the home, the punishment given to the dog will be as a result of embarrassment and the owner feels that they have to be seen to tell their dog off for its behaviour. I always advise owners, difficult though it may seem, to concentrate on reacting to their dog in the correct manner, rather than fulfilling another person’s expectations. If a dog has behaved unsociably it is always important to apologise calmly and politely, but then concentrate on the dog and try to think why it behaved as it did.
Many owners are reluctant to muzzle their dog as a precaution if it displays aggressive behaviour. I would suggest, once again, this is primarily because they are embarrassed or worried about what other people think or because it does not look very nice! However – if an aggression problem exists, it is far better to ensure the safety of all by placing a basket muzzle on the dog. This will also enable the owner to be more relaxed because they will know that, should their dog behave aggressively, it will not be able to cause any serious injury. It is important, however, to gradually introduce a dog to wearing a muzzle rather than forcing one onto it distressing situations.
Many potentially serious situations can be easily avoided, such as sheep chasing. If an owner knows that this is problem behaviour, the simplest solution is to ensure that the dog is walked well away from any sheep and when necessary, placing it on a lead.
As a general rule I recommend the implementation of the following solutions, dependent on the scenarios involved.
• If the unwanted behaviour is not serious, but something that the owner wishes to modify, such as incessant attention seeking (barking, leaping up etc.), ignoring the behaviour (negative punishment) followed by positive reinforcement of the desired behaviour is normally effective. However, if the dog has an emotional need it may be necessary to provide an outlet for its attention seeking behaviour, such as the use of an interactive key – where the dog learns that picking up a specific object, such as a particular toy, will result in attention from the owner.
• If behaviour needs to be immediately stopped, the use of interruption, distraction and positive reinforcement can be implemented. For example, let us look at the scenario of a dog that is about to chase a cat (which is a very rewarding experience for the dog!). A distinctive sound, such as a squeaky toy, can be used to obtain the dog’s attention, before it starts to chase. If this is immediately followed by encouraging the dog, in an excited manner, to come to the owner and be positively reinforced with the squeaky toy and subsequent game, this will result in the dog concentrating on its owner rather than chasing the cat. Eventually, the dog will find it more rewarding to ignore the cat as it will have learnt that it receives stimulating interaction from its owner.
• Some dogs respond more readily to the use of a ‘leave’ command. It must be reintroduced in increasingly stimulating circumstances so that the dog learns that it is positively reinforced for its appropriate reaction to the signal of ‘leave’. For example the owner can progress from calling their dog away from a stationary toy, to one that is moving slightly to calling it away from one moving at speed. Eventually, when the dog is placed in a situation, where it may have previously chased a cat, for example, it will return to its owner once the signal ‘leave’ is given and, due to the implementation of positive reinforcement, its motivation to chase the cat will be reduced.
• Where a behaviour must be stopped so that positive reinforcement of an appropriate behaviour can occur the use of conditioned avoidance techniques can be extremely effective, as they can help owners deal with more simple behaviour problems.
Problem behaviours, particularly more serious ones, can be very complex and involve a number of contributory factors such as the history of the dog, the relationship with the owner, how long the behaviour has been displayed, how much stimulation the dog receives and so forth. It is vital to accurately assess the motivation behind the behaviour and this may result in recommending a number of methods of behaviour modification, such as the use of conditioned avoidance, an interactive key, house lines, etc. It is not possible to detail methods for specific problems in this short chapter and those that have been included here are only examples. Owners who are worried about their dog’s behaviour should seek professional behavioural help from an APBC member in order that the problem is properly understood and an appropriate behaviour modification plan can be recommended and properly implemented.
I will not deny that aversive methods can sometimes be effective in stopping a dog’s inappropriate behaviour. But in most instances where positive punishment is implemented, it is mis-timed and without awareness of the possible consequences. These may vary from increasing the unwanted behaviour, such as in fear aggression, to the dog forming an association of pain with an unrelated event, such as a child passing by. Frequent positive punishment also leads to stress, both for the owner and the dog, and a breakdown of the owner – dog relationship. Surely it is better to have a dog that responds to signals because it has learnt that compliance results in something pleasant, rather than a dog that responds because it is frightened and wants to avoid punishment?
| REINFORCEMENT AND PUNISHMENT | |
| POSITIVE REINFORCEMENT | POSITIVE PUNISHMENT |
A rewarding outcome for a particular action: egs. • When the dog sits calmly, it is stroked. • When the dog is not attention seeking it receives attention CONTINUOUS REINFORCEMENT LEADING TO A VARIABLE RATIO REINFORCEMENT SCHEDULE
STRENGTHENS OWNER- DOG RELATIONSHIP |
An aversive outcome for a particular action: eg. • The dog is jerked on the lead when it does not sit. POSSIBLE ASSOCIATION OF PAIN WITH UNRELATED STIMULUS TIMING AND INTENSITY OF PUNISHMENT MUST BE ACCURATELY IMPLEMENTED WELFARE IMPLICATIONS CAN COMPROMISE OWNER- DOG RELATIONSHIP |
| NEGATIVE REINFORCEMENT | NEGATIVE PUNISHMENT |
The absence or withdrawal of an aversive experience when the dog displays the required behaviour egs. • ‘jerking’ on lead stops when the dog sits. COMPROMISES OWNER- DOG RELATIONSHIP WELFARE IMPLICATIONS |
The withdrawal of a rewarding out-come where the action performed was previously positively reinforced: egs. • Ignoring the dog when it jumps up. • Ignoring the dog when it seeks attention. IMPLEMENTED ON A CONTINUOUS NON-REINFORCEMENT SCHEDULE |
Someone once said to me that they couldn’t stand by if they came across a group of boys throwing stones at a window, they would have to ‘sort them out’. Whilst they did not clarify what methods would be used ‘to sort’ the boys, I rather suspect they would have included some sort of confrontational positive punishment.
My aim, and that, no doubt, of my APBC colleagues, would be the same – to stop the boys – but our approach would be different. In the first instance, we would interrupt and distract them from the behaviour but then try to find out why they were behaving like this. Boredom? Frustration? They didn’t like the person living there? Whatever the reason . . . once ascertained, the core motivation could be assessed and the problem behaviour dealt with to obtain a long term resolution. Admittedly, instant positive punishment may well stop the boys’ behaviour on that day – but it would not have dealt with their motivation and if the motivation to perform the behaviour is stronger than the fear of the potential punishment the behaviour will continue. However, once the motivation is known the behaviour can be re-directed into an appropriate, more motivating, positively-reinforced action, thereby resulting in a cessation of the initial unwanted behaviour.
The majority of owners have dogs as pets. They are considered to be a part of the family, trusted and faithful companions. However, all too often, we totally ignore them when they are behaving well, be it lying calmly, interacting well with people and other dogs, or not chasing the cat. But we instantly respond and interact with them when they start to behave inappropriately. Dogs do not speak the same language as us, so we cannot explain things to them. They will only learn from our interaction with them. So it is important that we, as a supposedly superior species, through patience, understanding and positive reinforcement, nurture, develop and continue to maintain that most cherished and potentially rewarding of relationships with man’s best friend.