•    

    Positive Reinforcement

    Dog looking up

    There is a saying that the only thing that dog trainers agree on is that they disagree with one another. In general I would tend to go along with this observation. This is probably because there are few industry standards regarding a dog trainer’s qualifications, education, experience and acceptable best practice and methods.

    The term ‘positive trainer’ is kind of mythic in itself – there is no specific definition and anyone can describe themselves this way and still use equipment or methods that employ fear, pain, intimidation or shock; this makes choosing a trainer exceptionally difficult for concerned pet owners.
    The following is to us, what is meant by the term positive training.

    Please understand that even a trainer that is not using positive methods such as choke chains, shock collars, yelling or ‘alpha rolls’ is still using operant conditioning whether they know it or not.
    So lets clear up any myths or misconceptions about Pet Central’s Pawsitive training:
    We want to improve the dog’s attitude toward his owners, training and the world around him so we use positive reinforcement (R+) more than any other quadrant. This involves the dog getting access to something he finds pleasant so as to encourage him to repeat a behaviour.

    Different dogs are motivated by different things such as food (in fact all living things are motivated by food – we all have to eat right?!), toys & games (even adult dogs enjoy playing different doggy games with other dogs, people or toys), social contact (dogs crave social attention from their nearest and dearest which usually includes several species) and behavioural outlets particularly chasing moving things (dogs are related to hunting animals and their desire to chase moving furries is still strong in most dogs).

    If you can work out which things are most motivating to your dog in different situations training becomes really easy – remember you control your dog’s access to these nice things so use them in training; have him earn it!

    We try to develop training programs that involve putting the dog in a situation in which he can earn a reward – this is always our first port-of-call.

    Making sure that the dog can earn a reward means that we set the dog up for success by carefully managing his environment. The more often a dog practices any behaviour (desired or not) the better he gets at it so the first step in training is to eliminate the dog’s opportunity to practice an undesired behaviour.

    In its place, we train an incompatible, highly rewarding behaviour. This is the backbone of good training – teach the dog to do something else that is heavily reinforced so that he can’t and won’t want to do the other behaviour

    Happy looking dog on a lead

    Training, just like life, cannot be without aversives however. Aversives are things that animals will work to avoid. In pawsitive training we minimise the use and strength of aversives.
    A dog may find a regular leash and collar aversive if it prevents him from sniffing that tree – no jerking, yelling or punishing need be involved. So what do positive trainers do in those situations?

    They will train the dog that the alternative that we would prefer is not aversive but absolutely great! The dog for example is taught that sticking with the owner is far more rewarding than the smelly tree. We can use sniffing the tree as a reward for good leash manners though!

    Along with a primary use of R+ we also extensively use Classical Conditioning. This is a form of learning that helps to change an animal’s involuntary responses. We like this to train conditioned emotional responses (CER), which is useful to teach an animal that a once neutral or scary thing now means something good, is coming his way.

    As we work with many dogs that suffer behavioural issues grounded in fear this is very important and we use it in programs known as desensitisation and counter-conditioning (D+C) programs. This teaches an animal that something scary e.g. a stranger means a treat is coming but we only work at the dog’s safe distance.

    We never attempt to work with a dog that is above threshold – this is not only ineffective as the dog will be unable to learn like this but is also very strongly aversive and distressing.

    Many trainers, particularly TV trainers and their followers, rely heavily on flooding which means to expose the dog to a scary thing at full strength, right up close and restrain the dog so that he cannot escape or aggress. The dog eventually stops responding due to a phenomenon known as ‘learned helplessness’ – this is why TV training appears to be such quick work, the dog no longer feels it has any control over its environment and all learning has ceased so it no longer responds. We do not use flooding as a way of training dogs nor do we condone its use.

    We may occasionally use negative punishment (P-) although only where R+ is not yet possible. P- means to withdraw access to something pleasant to reduce the chances of a behaviour being repeated (you might turn the TV off if the kids are being noisy, this is negative punishment). Negative punishment is most commonly employed in leash training, greetings and teaching calm behaviour.
    While training loose leash walking (LLW) we will stop dead when the dog puts any pressure on the leash, we might even use ‘penalty yards’ by taking a couple of steps back and waiting for a loose leash. As soon as we have a loose leash the dog gets to move on until the next time he puts pressure on the leash.

    To minimise our use of P- and maximise R+ we will first practice this exercise in low distraction areas where the dog is more likely to watch his owner and stay close so that the dog earns lots of clicks and treats.

    We might also use P- as part of training programs to train calm behaviours and to greet in a civilised way. We might withdraw attention or access to a treat until the dog is calmer. Again to minimise the use of P- we teach an incompatible behaviour first such as a sit so that we can reward the dog for that, thus using R+.

    Now that you know what we do, there is also a list of things that we don’t/won’t do:
    We don’t yell or tell dogs off nor do we use equipment that is designed to be aversive such as (but not limited to) choke chains, prong collars, tightening collars, shock or spray equipment etc.

    We do not use any physical punishers such as leash jerks, slaps, kicks, pokes, forced positions, scruffing, compulsion and such. We never put a dog in a situation that the individual finds intimidating, frightening, painful or strongly aversive.
    Positive training programs are developed with these guidelines in mind and above all else to rely on methods grounded in R+, that are dog friendly, scientifically derived, effective, motivational and fun.

    Our decision to leave out positive punishment (P+) and negative reinforcement (R-) is based on many factors including the problem that humans are very bad at using punishers.

    For them to be effective they must be severe (at first), bang on time and fit the crime.
    Society is addicted to the use of aversives such as slapping kids, physical fouls in sport and then yellow cards (!), prison sentences and on and on – the reason these are commonly used over and over is because they are not effective. If they were then they would only be needed a small number of times. Repeat offender stats tell us that that is not so.
    Well its the same in dog training – punishers are almost always employed as a result of trainer error.

    We set our dogs up for failure by leaving our favourite, most expensive shoes within reach for chewing, by leaving our adolescent dog the run of the back garden with access to our roses for digging, by allowing our dogs off leash even though we know we don’t have a 100% reliable recall, the list goes on…

    Punishment is highly reinforcing for trainers too! At first punishment suppresses ALL behaviour, as the dog may not know exactly which behaviour is punished. Because the timing of punishment must be sooooo exact (which is difficult) continued use of punishers are likely to cause more long term and widespread suppression – and now we have a dog that is afraid to offer any behaviour and is probably not too fond of his trainer. But because of this suppression the trainer believes that their punishing ways have worked and so are more likely to repeat.

    APDTI positive trainers much prefer to have dogs that offer behaviours and we can select the right ones, let the dog know and have a dog that just loves training (and his trainer!).

    But don’t dogs need to know when they have misbehaved & don’t we have to show them who’s boss?
    Glad you asked, this is one of the most predominant myths and misconceptions in dog training. Life offers plenty of aversives without us adding unnecessary ones too to their lives. My dog knows which behaviours I want rather than the ones I don’t want. Most of the time there is only one thing we want our dogs to do (e.g. lie down quietly on your mat) but there are an infinite number of ones I would rather he didn’t do. So which system is going to be more efficient? You’ve got it; catch your dog doing the right thing!
    If my dog misbehaves I have a very careful look at the situation: was he over aroused? Did I put him into a situation that was too much for him? Do I need to go back to the drawing board training that behaviour?

    What did I do that caused him to mess up? Do I need to proof that exercise a little better?
    By the time I have gone through all of these options it is too late to use a punisher and I will usually have found the answer: the buck stops with me!

    There are situations that we might get into where I need to stop or interrupt my dog’s behaviour. This is where I will use what’s called an instructive reprimand as an interrupter. This will only be possible if I have a super proofed behaviour that I can use in all sorts of situations. The one I use for my dog is ‘wait’ to cause him to stop in his tracks, particularly useful during squirrel chasing I can tell ya!

    Golden Retriever sat on grass

    Probably the best one is to teach your pet dog to sit – there are so many things he cannot do if he is sitting! The beauty of this is that the sit behaviour should have been heavily reinforced in the past so that being asked to sit should have nice associations for your dog so its not really a reprimand at all.

    Start now by teaching your dog to sit before access to anything such as doorways, leash on or off, sniff that other dog, play with your pal, throw the ball, hop on the sofa, get out of the car, ask your dog to sit several times during play sessions and then let him resume playing… Soon you will have a dog that asks for permission to do these things and a great way of controlling him if things get out of hand.

    Please don’t be fooled into thinking that positive means permissive – I can assure you it does not!
    Let me say this again: you do not need to show your dog who is boss, dominate him nor put him in his place. Please be your dog’s guide to the human world, don’t try to trample on him.

    Positive Peeve:

    One of the biggest misunderstandings associated with positive training is the use of food and one of my biggest peeves in relation to positive training is that some trainers don’t know how to fade training aids so they tar us all with a bad reputation.
    Food is a primary reinforcer; this means that animals learn to alter their behaviour to gain access to food very quickly. Using food is one of the short cuts in dog training. But it is by no means the only or most effective reinforcer.
    Dogs will work for all sorts of reinforcers and the strength of that reinforcer to an individual dog varies greatly from dog to dog and from situation to situation.

    Dogs are pretty highly motivated by food but there are some situations where food is trumped by higher value reinforcers such as social contact with other dogs and the chance to chase squirrels.
    Because of this I use food to build behaviours strongly in low distraction situations but I quickly change to using access to these high value reinforcers in other situations.

    There are a couple of different ways of teaching a dog a behaviour using operant conditioning – which behaviour, which dog and which situation will dictate which method to use.

    We can capture the behaviour – this means to wait for the dog to offer the behaviour or put the dog in a situation where he is likely to offer the behaviour and reward him whenever he does. This only works for behaviours that dogs offer naturally and regularly or those that are easy to elicit.

    We can lure the behaviour – this means to use a treat, toy or target to move the dog into position. This is easy and quick but it can take some time and skill to wean the dog off the lure.

    We can shape the behaviour – this means to reward successive approximations of the final behaviour until the dog can perform the full behaviour. This is most useful for multi-stage, complex behaviours.

    But when I am initially teaching a new behaviour, regardless of which method I choose, in a low distraction environment, I tend to use food.

    Many people of course say that their dog is not food motivated – well biologically your dog is. To have your dog delighted about working for food we make a change to the dog’s routine. Instead of feeding your dog from a bowl in one sitting we measure out the dog’s daily ration of kibble and leave it in little strategically placed pots in key areas of the house where you will always have rewards within reach.

    The dog now has to earn his dinner with good behaviour. This will greatly speed up training.
    We also use higher value foods for difficult behaviours, really challenging situations, classical conditioning and for really good behaviour.

    However, the use of food is not as important as the ‘no-longer-use-of-food’! If food is used for too long or used inappropriately it becomes difficult to get reliability (in dog and trainer) without the use of food – the reward becomes a bribe.

    Once we get the behaviour we begin on the road to weaning food out of the equation. Food can be replaced by less food, then by toys, then by praise and cuddles and then by the act of carrying out the behaviour.

    Food can be used occasionally there after but should no longer be relied upon to get a dog to perform.
    I should stress here that if a response is taught or maintained by classical conditioning in which food was/is used then food must be continued for the training to work and keep on working.

    Remember that training is a life long activity so food and other reinforcers are likely to be re-introduced several times over throughout the dog’s training. Access to reinforcers should be controlled all the time by the owner and remember that reinforcers are anything your dog wants access too.

    German Shepherd playing with a rubber toy

    Make a list of your dog’s favourite things and use these for training and living behaviours.
    Regardless of which training tools we use (and of course we only use ones associated with positive training!) they are just that: tools. All training tools require responsible use. If we cannot achieve the same behaviour without the tool than no or next to no training has gone on.

    So using positive training methods properly will and does work very efficiently and effectively.

     

    Positive Philosophy

    So you can see that ‘positive training’ is more a philosophy that employs hundreds of different methods that can be applied to varying situations. ALL of the methods that we choose are selected because they are dog friendly, motivational (as in teaching the dog to want to do something), effective and scientifically derived – not based on opinion or myth! They are our only limitations.

    So when I hear people say that such a methods isn’t any good and that there is more than one method – I agree insomuch as other ideas fit the criteria for our philosophy (i.e. dog friendly, motivational, effective and scientifically derived).

    For interested trainers there is a wealth of scientifically derived information available in relation to dog behaviour. Saying that however, there is still much to be done and still much to be learned – that’s what makes the dog training field so exciting and why it is so frustrating when we see people still use old, disproven, traditional and all too often damaging methods.
    So if you like the sound of this, choose pawsitive – your dog will thank you!

    Anne Rogers – Pet Central Pawsitive Petcare