Positive reinforcement is much more than just “training with treats”. That’s why there is a guide to efficient, animal welfare-compliant training in the following – based on the questions that I am asked most frequently in this regard.
What is positive reinforcement?
The recipient, not the sender, decides what is rewarding. Even if you like to whisper your dog when he has done something right, animals do not always find touching pleasant when they are concentrated. In principle, a reward serves to reinforce desired behavior so that it occurs more often as a result – this is why it is also referred to as an “amplifier”.
Learning Theory: Operant Conditioning – Quadrants
In the case of positive reinforcement, “positive” does not mean “good” as often expected, but something is added in training, after which behavior due to an improvement in the animal’s condition will occur more often in the future. In short: The dog learns that certain behaviors (unlike others) pay off and shows them more often in succession.
This is a sustainable, efficient and science-based approach to dog training, which is also low-risk and is also recommended by the BMG.
But does my dog become a command robot who only does things for the reward?
If you consider the following points, that shouldn’t happen. Common side effects of training through positive reinforcement are happier, more cooperative dogs, who gain self-confidence and solution strategies, and build a more secure bond and stronger relationship with their owners. In addition, most owners also gain confidence in their own handling skills and their four-legged friends. Some clients have even reported that changing their perspective to focus on desired behavior also has a positive effect on other areas of life.
Doesn’t the dog learn this without reward or conditioning?
Possible, but almost certainly not as sustainable. Of course, dogs also learn through habituation or when they have negative experiences. However, this can quickly backfire.
Psychologist Kirsten Cordes, one of my trainers, based on the well-known quote from Watzlawick, said: “You cannot not condition”. Behavior has consequences for the dog. If, for example, he wants to distance himself from other people, barks at them loudly and then backs away, he has achieved his goal. In simple terms, conditioning means linking stimuli with reactions – and since this is not consciously controlled, it cannot be avoided. What we can do, however, is to take advantage of this.
Doesn’t positive reinforcement take much longer than other methods?
Studies that compared modern with outdated methods found that positive reinforcement not only brought about more sustainable, but also quicker successes in reaching a training goal. While punishments can work immediately, they have a tail of possible negative consequences. You can read more about the possible consequences of outdated training approaches here.
How do I know if I confirm my dog or just fed him a biscuit?
If behavior occurs more often afterwards, positive reinforcement has taken place – if not, then you have simply given your dog something to snack on.
The tricky thing is that you can of course (unwanted) reward undesirable behavior. For example, if your dog nudges or jumps at you to get your attention and you talk to him (also complaining), then the behavior has brought him to his goal. If you want to wean him off, for example, you simply have to give him more attention for not poking or not jumping – ideally before the unwanted behavior occurs at all.
The best thing is to make a list of all sorts of gutzis, activities, games, etc. that your dog particularly likes. The more difficult an exercise is for him, the better the reward should be. This way you can always reinforce your dog in a way that suits the situation and needs, which will increase his willingness to cooperate and his learning success in the long term.
Do I always have to have cookies with me now?
Yes and no, because you have created a list with the various things that your dog thinks are great. You can at least fall back on that theoretically if he did a great job. Of course, this should be appreciated, especially if it takes place in everyday life. After all, your dog should remember that such behavior is particularly profitable!
However, you must be aware that alternative amplifiers such as playing, touch, etc. must first be learned (= trained) as such. While they offer you great opportunities to make the training exciting, of course that means your own training time again.
For targeted exercise sessions, you should unpack the treats as a positive reinforcement. Experts such as the internationally recognized animal trainer Ken Ramirez advise you to use food rewards 80 percent of the time during training.
With appropriate training progress, the rewards can then slowly become “inferior” (instead of the favorite treat after an exercise, there is now only a “normal” one as soon as it is no longer so challenging for the dog) and variable (= there is no longer any successful one Execute a reward, but sometimes just praise).
Do not be too stingy with the rewards – trainers are mostly more successful because they have much higher reward rates than normal dog owners. In fact, it is more important to reward the dog for desired behavior than that it is always a particularly tasty or large gutzi.
Yes, but isn’t that bribery?
Bribery means offering something beneficial to someone else as compensation for doing something for you. The main difference to training with positive reinforcement is that the dog must perform the desired behavior with the latter before he gets anything for it. Because in modern training you switch to variable reward rates as soon as possible, the dog does not expect “compensation” for every performance later.
Please don’t forget, your dog does an incredible amount for you every day – he adapts to your everyday life, your language, your requirements as best he can. Target rewards, your dog will thank you.
What is the best reward?
As already mentioned, your dog decides which reward he thinks is the best – and this can vary from situation to situation. On a hot day, being allowed to throw yourself into the next river is a great thing – probably less in the middle of winter.
Since we always include an emotional component in training and it is better to reward as often as possible for rapid progress, I like to work with feed rewards, especially with basic obedience and behavior modification. In principle, I choose my amplifiers so that they make it as easy as possible for the dog to implement the desired one. For trick dogging and moving exercises, where we need greater enthusiasm and more movement from the dog, positive reinforcement is used rather as a game reward than to teach him a quiet seat, in which he should later have no ants in the rear.
Why should we train with treats?
Food is essential for survival. For this reason alone, most dogs like to work for a wide variety of feed rewards – and it is much easier for us humans to ensure that the dog has really been rewarded. Of course, as mentioned above, there are also alternatives – but these often require a higher level of engagement with the dog and its environment as well as its learning behavior.
Treats serve as a stress indicator in training. If a dog has stress (whether good or bad), it may be that he can no longer take in food, spit it out again or takes it particularly “snappy”. This contains important information about his state of arousal – and the more excited your dog is, the less targeted learning takes place. This means that your dog no longer necessarily learn what it should learn.
For example, if your dog refuses treats at a certain distance from a certain trigger, which he otherwise likes very much, then this tells you that he needs a greater distance from this stimulus for successful training.
Does your food-motivated dog eat certain treats much coarser or more hectically than others? Then these may be more distracting than promoting concentration. Make it easier for him to stay calm by eating a few degrees less popular feed. If Digga leaves Prosciutto on the street, for example, he then gets a piece of deer meat. If he learns a new exercise to think about, I leave the deer in the goodies rest safe and sometimes even resort to dry bread. Snapping treats can also be due to negative stress. In this case, your dog usually needs more distance from a particular trigger.
What should be in the treat bag?
Everything your dog likes and doesn’t make him sick. I like to work with a mixture of different treats with different consistencies. With household scissors or a knife, I cut chicken fillets, natural casings, duck fillets, lambs, beef sausages and various other things into small pieces – as already indicated above, it doesn’t have to be a giant bite to be rewarding. Then there are often a few soft trainees, a few homemade dog biscuits and pieces of fruit or vegetables. The fact that the dog does not know exactly what he is getting for an exercise makes the training more exciting for him.
An ideal treat is so small that your dog chews a maximum of once and is therefore quickly ready for the next repetition. A customer with a small dog puppy, for example, worked with 2x2mm pieces of Gouda or sausage slices – 100 reward opportunities from a large ham leaf!
Do I have to use a treat bag?
If you get to the dog biscuits without a lot of fumbling or rascal, you can of course also work without a treat bag. The food bag only serves as an aid that enables faster handling, better timing and rustle-free rewards. Remember: Desired behavior should be rewarded within two (2!) Seconds of its occurrence so that it can be linked as such. If you work with Marker, you may get an extra second because the trained dog will remember what you thought was good – but you should always be nimble!
How is the training going?
You always get some kind of homework from me, which of course I also explain to you. Some things are built up in separate practice units.
For example, if you have received three different exercises as homework, you simply divide them up for the day – e.g. attention training in the morning, hand target at noon and crockery handle in the evening. If there are opportunities in everyday life to briefly incorporate an exercise, you can of course also use it.
With some exercises it is more important than with others that they are built up well before the real application – please stick to the instructions you get from me and do not experiment without consulting. I am always available to you by phone or email.
In general, it is important to differentiate between training and everyday life. In training, we want the dog to make the desired connections between its caregiver, its environment and its behavior as quickly as possible. That is why planning, time and money flow into training. In everyday life, we can often just deal with situations in such a way that no undesirable learning experiences take place. This means that we also discuss management options with each other so that you know how to avoid such situations.
What is management?
We call management measures that prevent undesirable behavior without any intervention – e.g. sticking a frosted glass film on the balcony door in front of which the dog always barks, changing the doorbell, securing the unavailable dog with a towline, in good time before dog encounters to change the street side etc.
For you, it also means looking ahead and consciously noticing possible trigger stimuli and experiencing them with your dog, if possible, only in a stress-free intensity or proximity.
Management should complement the training and cannot replace it. It ensures that your dog can no longer consolidate the unwanted behavior and bridges the time until certain exercises are trained so far that you can use them in everyday life. How quickly this happens depends on how well you manage and how successfully you practice.
How do I start a training session?
If you want to train specifically, of course you have to prepare certain things beforehand – reward, any aids, etc. – and consider what you want to practice and at what level you were last successful. Then you speak to your dog and ask him to come to you with a phrase like “let work” or the like. Starting a training session with such a signal helps the dog prepare for learning – a bit like the school bell for you back then, ideally only more positively linked.
How does practicing work?
For your initial practice sessions, it is best to choose a dog with a low distraction and familiar environment, such as your own living room or garden. A training session should last between two and five minutes. You should do at least eight to ten repetitions of a training step. If your dog is successful at least eight times out of ten at a training level, you can proceed to the next training step. Depending on the level of concentration and motivation of your dog (and you!), You will end the exercise session after two to five minutes.
The next unit of this exercise begins with a few repetitions of the last step you successfully completed. For example, if you reached step 3 of the hand target on Monday and did three repetitions, on Tuesday you start with five repetitions of step 2. If your dog is successful, you can go back to step 3 – if it is not, you are working on a success rate of at least 80 percent go before you proceed to step 3.
How difficult an exercise is for your dog to do depends on motivation, environment and level of distraction. If you want your dog to be able to practice the same behavior later on, you must of course gradually incorporate these factors into your training. When you start to vary the environment and distraction, you should make it as easy as possible for the dog to succeed even under the changed conditions – so go back at least one, if not two, training steps in your requirements.
How and when do I end the training?
A training session of an exercise should last between two and five minutes. How long you want to practice depends on your and your dog’s motivation and ability to concentrate. The guideline value of two to five minutes is rather at the lower end, so that all clients can also find time to practice with their dog. If you are both enthusiastic about it, of course there is nothing to be said against exercising longer. But don’t forget to take a break after at least 15 minutes and let the dog drink.
Please note the following during training: The training should in any case be stopped as long as it is fun and the dog is successful – after all, it should learn to cooperate joyfully.
Whether you finish the training with a phrase or not depends on your own dog. If he has a hard time coming down after a training session, it may be just fine to teach him an “end of session” signal like “finished work” to make it easier for him to relax.
So that your dog shuts down faster after practicing, you can give it something to chew, take a sniff walk, offer a massage or otherwise do a nice quiet activity.
What do I do if my dog makes a mistake?
A recent study shows that dogs who are warned of their mistakes by a neutral tone also make more mistakes and learn more slowly. For training, this means that you shouldn’t correct your dog according to the Dogs in the City motto if you want to move ahead quickly in the plan.
What you can do instead is this: Be neutral for two to three seconds – no “no”, no “na-a”, no redirecting the dog, no turning away – just be there without rewarding the dog. After the few seconds have elapsed, you give him the opportunity to earn a reward for behavior that you are certain he can master (which may be a completely different behavior than the one in which he made a mistake). When you return to practicing the behavior in which your dog made the mistake, you start again at the last successful training step.
What do I do if my dog does something he shouldn’t do?
Quite simply: tell him what to do instead – provided he has learned beforehand how to perform this behavior under the given circumstances. In addition, it is advisable to change the situation or manage it as discussed above so that the dog cannot perform the undesirable behavior and to practice at the next opportunity how the dog should behave in such situations instead.