EDUCAN and Dognition, towards a new technology of behavior. Cognitive technology, of course!

Publicado el 19 de March de 2015

As dog trainers we are behavior technologists. Hence, we need to have a technology that translates the raw knowledge acquired by basic scientists into tools which are applicable to the education of our canine learners on a daily basis.

However, the technology available in the dog world was developed within a dated framework which is now under strain and needs to be overcome. Despite being effective it precludes us from exploiting the new discoveries and from offering services of a higher quality.

The science of behavior has changed its focus in recent years. It used to be limited to observable conducts whereas now it pays prominent attention to the study of the cognitive processes underlying such behaviors.

A further change concerns the fact that behavior analysis used to focus on general learning processes, those which are common to a majority of animal species. In contrast it has now started paying attention to the learning capacities peculiar to each species and even to each individual within a given species.

Finally, since dogs had not awaken the interest of researchers focusing on behavior, their way of acting used to be inferred on the basis of comparisons with other species, either rats and pigeons studied in laboratories or wolves observed in a natural environment. Nevertheless, in recent years dogs have been exhaustively studied with regard to their cognitive skills, both individual and social, as well as in relation to their communicative abilities and their way of interacting with humans.

At the end of the 1990’s EDUCAN started to develop a new technology of behavior designing protocols and intervention guidelines that took advantage of this wealth of new scientific knowledge available. Since then we have being making a sustained effort and, of course, my new book , “Your Dog Thinks and Loves You” further progresses in the same direction.

At that time those scientists working in the field were changing the way we understood dogs but did not show too much interest regarding the practical application of their discoveries. Today this has changed too and Dognition is the stepping stone in this line of research.

Dognition is a website cofounded by Brian Hare who is the author of “The Genius of Dogs” and the Director of the Duke Canine Cognition Center. Within its panel of experts Dognition counts with some of the leading researchers on canine cognition. Its objective is to offer owners custom results about their dog’s cognition. This is done by proposing games that owners can play with their dogs in the comfort of their own homes.

Dognition is thus developing technology for cognitive assessment, an effort that dovetails nicely with EDUCAN’s mission: the development of technology that allows the provision of cognitive training and education.

To better understand the importance of Dognition’s technological contributions, consider the example below. My colleague Javier Moral, Dean of Studies at EDUCAN, and I teach our training courses. However, our way to prepare the materials for the class differs widely. Javier’s intelligence, on the one hand, is very visual and so there are plenty of colors, arrows, overall views and other graphic representations in his notes. On the other hand, mine are totally different because I focus on words to learn. Thus, I summarize, create overviews and phrase the contents that I have to explain to students in alternative ways. Doubtless, if one of us had to prepare for the class with the notes of the other, it would be an ordeal and our performance would certainly be inferior. However, neither of us is more intelligent than the other. Simply put, our cognitive skills are different and, here is the bottom line, the working methodology needs to be in consonance with such peculiarities to make the utmost of the cognitive capacities of each one of us.

Likewise, we are aware that our children do not learn in the same manner. Hence in order to help them develop their full potential we identify activities where they excel and wisely combine them with others that contribute to develop weaker areas. We should do the same with our dogs.

However, the currently available training technology only offers (offered ) resources that serve to exploit general cognitive skills, those which are common to many species. Such strategy may provide quality tools when we want to teach circus skills or merely mechanical and repetitive routines. However, such tools are barely sufficient to provide an integral education for our dogs and at any rate they are suboptimal. This is particularly evident with regard to individuals belonging to breeds that diverge from the average standard on one side or the other. Good trainers are capable of deviating from the general rules to adjust to the needs of each individual learner. However, when they engage in this process trainers lack any technical criterion or, for that matter, any objective tools for the assessment of progress. In sum, they are solely relying on their talent which is clearly unsatisfactory.

This is the reason why Dognition and EDUCAN have come together in a project that fills both parties with hope. Its aim is to develop a new behavior technology that, in the first place, evaluates objectively and clearly how each dog learns, and then, to propose protocols and technical tools that suit each custom assessment.

And, since I know my followers well and I feel that you are about to ask it, I will tell you before you do. Yes, we are working on the translation of our websites and the rest of the information to the other party’s language.

This will not only lead to a new understanding of how to train dogs but it will also change our mind frame with regard to how we live together with dogs. Through training we will be able to help dogs blossom, live full and integrated lives, thus ultimately becoming happier.

A new and brighter era is dawning in the training and education of dogs world.

Share Button

About wolves and dogs

Publicado el

In the world of training there are some questions that reemerge from time to time giving rise to enthusiastic opinions on both sides. One of these topics is the reliance on comparisons between dogs and wild wolves.

The controversy is currently on fire in Biology and I intend to explain its state of affairs in the present article so that readers are able to ground their personal opinions on a more scientific basis. I will attempt to be brief, though I have to recognize that I am seldom successful in that regard.

Not so long ago wolves and dogs were deemed to be two different species from the taxonomic stand. Taxonomy groups animals into families, genus, species and so on depending on how close to one another they are thought to be. In order to assess this evolutionary proximity, morphological/anatomic similarities and differences (i.e., form and structure) were typically taken into consideration. Such classification method required some guess work and heavily depended on the analysts’ sagacity to interpret the observations made. By using these yardsticks, dogs (canis familiaris) were separated from wolves (canis lupus).

The classification methodology just described has changed due to developments in the study of the genome. DNA studies have caused reconsideration of many different species and substantial reorganization of their taxonomy. Conclusive data support the conclusion that dogs belong to the lupus species. Hence the most relevant organization in the world dealing with taxonomy (i.e., Integrated Taxonomic Information System –ITIS- and Mammals Species of the World) currently classify dogs as canis lupus familiaris, a subspecies of wolves.

Does this mean that dogs are tantamount to wolves and that we can draw a strict parallelism between them? The answer is no.

Ethology has proven that ecology determines behavior. Thus, for instance, Iberian wolves (canis lupus signatus) live in small packs composed of an adult couple and one or two offsprings. There are seven individuals at most in these packs, what leads to simple and relatively primitive social interactions compared to, for instance, arctic wolves (canis lupus arctos) whose packs can reach thirty individuals. The size of the pack determines crucial aspects like home range and prey size.

Ecology has an important influence on hunting techniques. Iberian wolves practice stalking and hunting raids, depending on the type of prey they are after, whereas other lupus subspecies do not make use of such techniques.

Let me quote from Signatus.org:

“Usually when they attack flocks of domestic animals wolves stalk in a coordinated manner. When confronted with the obstacle of shepherd dogs, one of the wolves will let himself or herself be seen in order to lure the dogs’ attention. When the prey consists of rabbits, one or several wolves will beat the area while the rest observe, usually close to the access of the hutch where the rabbits will flee when they are pursued.”

Also available at: faunaiberica.org and animalesenextincion.es

In sum, the behavior of wolves is far from uniform. Furthermore, in relation to dogs we ought to take account of the domestication process. After the experiments carried out by Belíayev, we know domestication has changed dogs’ way of learning as well as their relationship with the environment.

Belíayev worked with silver foxes, a species used in the fur industry. By selecting for tameness the Russians wanted to obtain specimens who would be easier to manipulate, in order to cut costs down. After a few generations they obtained foxes who could be easily manipulated and who were friendly with humans. It was also found that tamed animals had improved skills for learning through operant conditioning, higher social plasticity, lower instinctive rigidity and a decreased cognitive confrontation of problems (Beliáyev 1969, 1979, 1981, 1982). On a different note, the story had a happy end for the wolves since due to a change in the color and patterns of the fur they were considered useless for the industry.

These differences regarding tameness and living environments lead all scientists to agree that dogs and wolves are two different etho-species. In simpler words, they differ in their behavior. In fact, due to the existence of such outstanding behavioral differences more and more ethologists are arguing in favor of a different way of classifying species on the basis of whether they have followed a different course of evolution, as it is the case with wolves and dogs, as well as with Right whales in the North Atlantic and the Pacific regions.

Taking these criteria into consideration the ICZN and the ITIS currently admit the term canis familiaris as a valid synonim of canis lupus familiaris for the purpose of scientific works, articles and communications.

Summing up, today we know that dogs are a subspecies of the wolf, given their genetic proximity, but that they are a different etho-species because of their ecology (constantly lying on a sofa in our living room while our children cuddle them) and the process of domestication has had a dramatic impact on their conduct, setting their own course of evolution, one that differs from any of the other members of the lupus genus. Therefore, it is correct to make comparisons between dogs and wolves to obtain data, but a strict parallelism is not possible since it would neglect prominent differences.

As it is usually the case, it’s neither black nor white.

Share Button

Benefits of stress (II)

Publicado el

WARNING: This is an extremely long article, one of those my webmaster bitterly complains about!

In a previous article I analyzed the general benefits of working with stress. As Fernando Silva commented then, if we are unable to teach our own dogs to work with little doses of stress and to manage them correctly, our hands will be tied for training other people’s dogs in commercial works, behavior modifications or, for that matter, to train high quality conducts for dog sports.

Bottom line: If we cannot handle the low level of stress caused by confining our dogs in the car for a couple of hours, as Mr. Silva explains, or the stress arising from a trip to a new location, as it happens to those who take part in sport competitions of any kind, or the stress deriving from the implementation of any of the available handling / safety protocols designed for the treatment of fearful dogs or some types of aggressive dogs, or even the stress caused by the learning of a difficult behavior, we will be unable to train most of the times. Thus we will have to renounce to work precisely with those dogs who would benefit the most through an improved quality of life in the medium term.

In sum, it is imperative to teach dogs to manage stress. As a matter of fact this learning produces two main benefits:

  • Dogs will progressively become less stressed out when confronted with the same stimuli or with environments that formerly resulted in high levels of stress.
  • Dogs will no longer appear to be insecure, unstable or nervous when something affects them. Their attitude will change progressively as they become capable of confronting problems in a confident manner.

The first point mentioned above contributes to improving the excessively emotional reactions so characteristic of sensitive dogs (e.g., Border collies or Malinois) when they are challenged with changes in the scenario. Very often these dogs exhibit worry, excessive surprise and even fear in the face of certain changes to the environment. I have witnessed how dogs impeccably trained with fully respectful methods reacted with fear in such circumstances. Facts that novel trainers misinterpreted as meaning that the dog had been mistreated and harshly trained. This was not the case. Simply put, these dogs can be compared to brilliant scientists who are unable to present their work publicly because the crowded atmosphere inhibits their communication skills and make them appear as intellectually clumsy. Thus we need to train their stress management abilities. In order to get there the most common mistake we should avoid is overprotection. As a matter of fact, trainers tend to avoid presenting stressful situations instead of teaching dogs the tools to sort out such situations.

The second point referred to above is even more important because it determines whether dogs will learn to enjoy themselves when confronted with levels of stress sensibly graduated. This effect can be compared to the case of persons growing in the face of adversity. We all agree that there are few feelings that beat the satisfaction stemming from successfully overcoming a difficult task that we perceived as a problem (e.g., the training of a particularly complex exercise, taking part in a tough competition and so on). This implies a dramatically important change in frame of mind for our dogs, from considering a situation as a source of worry to seeing it as a chance to have fun applying the skills they master. The shift depends on the acquisition of mastery in stress management. I think most of the great trainers I have come across do not emphasize their successes as much as how their dogs kept working and growing in the face of adversity, for there are few things as moving as taking part in a competition and, in the most difficult situation, noticing how your dog increases in implication and tenacity.

As a matter of fact, the benefits deriving from good stress management are so important that some researchers have devised ways to replicate stressful experiences while taking the associated risks away, so as to improve performance and learning punctually. These stress “simulators” are based on three main yardsticks:

  • Arousal levels: Stress always implies an increase in the degree of physical activation. Thus if we manage to increase the level of arousal over the regular thresholds we will be able to rip improvements in attention, reaction times as well as in the skills to discriminate relevant information. The best part is that this can be achieved without any of the risks associated with actual stressful events. It suffices to start by exercising dogs a bit to speed their physiology up. This simple intervention substantially improves performance, self-satisfaction and welfare.
  • Novelties in the environment: The introduction of anything new under the sky triggers a minimum amount of adaptive stress. By presenting novelties to dogs in an intentional and calculated manner we can improve their capacity to adapt, their attention skills, their sustained concentration and all of the benefits already commented in the first part of this article. And without risks. Thus researchers found that automatic improvements in the ability to analyze problems, better results and a decrease of distractions followed by simply changing the workplace of various workers.
  • A break from routine: Routines calm and eliminate stress. When researchers changed the routines of several workers, thereby increasing their stress levels, a very interesting effect emerged. At the beginning all workers claimed that the change would be detrimental to their work and their performance, as it was bothering to change their customary way of doing things. However, after a period working out of the routine, the results achieved in terms of performance, time and quality of the work showed improvements. Even more surprising, the workers’ feelings of satisfaction had also increased. This can be a temporary effect aimed at eliminating the possibility of suffering residual stress. After a change of routine the return to the normal situation prevents the stress from building up residually and having negative effects.

By using these three principles we will set good foundations for stress management, something that will allow us to reap the benefits without incurring the losses of stress.

But watch out! These three simulators are so effective –remember the close link between stress and amusement- that those who rely on them on a frequent basis can become addicted (do not forget that stress leads to the production of endorphins) and may lose the motivation to work without stress. This explains cases of workaholism, persons who need a permanently high level of arousal, novelties and new challenges to feel good. They are on drugs!

It also explains why so many Agility dogs seem to show extremely high levels of stress during the events. Recently a friend of mine who competes in Agility commented the case of her Border collie with me. She did not understand why the stress levels were so high in the field. He was a couch potato at home and walked and played in the countryside several days per week, as a normal dog. The social structure at home was also normal, he got along well with the other dogs, and my friend had always trained and treated him with respect. The reason for this behavior is that Agility is an unintentional but brutal stress simulator: high level of physical activation, environmental changes (competitors often go to different fields in a matter of weeks) and changes in routines (each field differs from the rest). This is why Agility tends to generate addiction. In these cases the solution is not to prevent the dog from activating, something that cannot realistically be achieved, but to teach the dog to manage the stress from the beginning so that it does not start building up in the course of different competitions and training sessions. That kind of stress is similar to that exhibited by people who lose their heads for a hobby, as it is more closely related to excitement in anticipation of the amusing experience than to any negative experience!

How to manage stress in these peculiar exciting environments will be the topic of the last part of this trilogy. I hope you are enjoying this saga about the benefits of stress for dog training and their general quality of life, always provided it is adequately managed.

 

Share Button

The benefits of stress

Publicado el

Last year a modest soccer team, Alcorcón, gave Real Madrid a proper ticking-off. Alcorcón won 4-0 and a new idiom was coined in Spanish language to celebrate the event: “el Alcorconazo”. Analysts agreed that one of the main reasons for this unusual defeat was that Real Madrid was too relaxed, whereas their rivals did their best. In fact, people blamed Real Madrid players for not being sufficiently stressed, unlike their rivals, and ready to confront the competition.

It is typical that whenever we speak about stress in relation to dog training, or more generally in relation to dogs, we emphasize its negative consequences. However, we regularly forget to mention the numerous benefits stemming from this adaption mechanism. Due to this biased view stress is becoming one of those “evils”, full of harmful consequences, something necessary to avoid at any price.

However, research has shown the many benefits of stress when it is correctly managed. In fact, if stress was not beneficial it would not have developed as an adaptive mechanism in the first place. It exists because it works and it helps in the assessment of many different situations and in dealing with them.

Nowadays we know that stress is not only convenient but essential in order to succeed in sports and to achieve artistic creation, that it enhances enjoyment and eliminates the dependency from external reinforcers, and that it is an indispensable element for personal growth in each of us.

We should take into account that stress is an individual and internal process. Its causality can be modified, thus there is no fix relationship between cause and effect: what stresses some dogs, may leave others unaffected. To put it in more simple words: Stress is not something that happens TO the dog, but something that happens IN the dog. It is important to bear this consideration in mind while training and not to consider those factors generating stress as if they were fixed. Stressors are dynamic and that is why one of the worst strategies to deal with them is to systematically avoid all stressful situations. This way the dog will progressively become more and more sensitive, and hence the stress response will progressively be triggered by stimuli of a lesser magnitude.

But then, how should dog guardians deal with stress? Given that it is an internal process which can be modified, the most useful and effective strategy consists of teaching dogs to manage it so that they will not trigger the stress response in relation to stimuli which do not represent an actual danger. This progress has a great potential for alleviating the situation of dogs with phobias and those who are particularly sensitive.

Once the dog has learnt how to manage the stress, we will not just be able to avoid its detrimental effects, but also to exploit its many benefits.

Stress has several benefits for dog training provided it is correctly managed:

  • Strong implication and compromise with the work. The dog enjoys the activities to the maximum.
  • Self-reinforcement, due to the correct activation of stress mechanisms the dog experiences the activity as reinforcing in itself, what makes the use of external reinforcers superfluous.
  • The dog has more fun because enjoyment is directly linked with stress. Most of the activities that we find enjoyable give rise to stress too (eustress), and this relationship holds also in the opposite sense: we enjoy those stimuli that trigger eustress in us.
  • Reduced latency (i.e., increased speed), the responses offered under stress are faster and more assertive.
  • Attention training, stress contributes to focusing all the attention on important issues. As a consequence there will be automatic improvements in relation to:
    • Concentration.
    • Skills to ignore trivial stimuli unimportant for the task.
  • Improvements in resilience, defined as the capacity to resist stressful situations without suffering negative effects and to overcome such situations positively. In other words, it seems that it is necessary to train with stress in order to avoid its negative effects.
  • Dogs develop the feeling of controlling the situation at issue, and hence the negative elements that they may encounter are managed proactively and perceived as something that they can overcome. As a result they become more self-confident.

However, in order to exploit these benefits without falling into the negative side, it is crucial to be very strict in fulfilling the following requirements:

  • Provide a clear and quick response to the stressful event, the solution to the situation should be immediate and easy to reach for the dogs, either because they know it perfectly or because they can easily deduct it.
  • Provide only short activation periods. If we extend the working session for too long two results are likely, either the beneficial effects will disappear or the dog will become “addicted” to working under stress and we will not be able to do without it to maintain the quality of the training. In both cases we will face the problems we already know stress causes.
  • Provide enough time and leisure activities for the dog to recover. Stress is designed to provide its benefits when it is activated for brief periods of time and occasionally. Therefore, it is very important to allow the dog go back to normal after the activation period. One simple way to do this is to let dogs practice their favorite activities (e.g., running in the countryside off-leash).
  • Ensure the good physical condition of the dog so that the additional arousal does not give rise to tensions or muscular problems that prevent the dog from enjoying it. Stress is physically exhausting. Therefore, no benefit will be achieved if the dog is sick or in poor physical condition.
  • Activate your dog only occasionally, not in every single training session. It is quite easy to be tempted to forget this recommendation after witnessing the many and important benefits deriving from stress, both for the training and for the general quality of life. A skilful trainer may obtain high performance from dogs at the price of making them addicted to stress. However, any benefits for the quality of life of the dog will completely vanish and the stress will show its detrimental side.
  • Autonomy, the results of combining stress with a strict training style where even the smallest mistakes are immediately corrected are devastating. For the dog to feel control over the situation and the stress to have benefits it is of paramount importance that the responses are perceived with a certain degree of leeway, not as something exact and that cannot be changed.

The benefits of stress are becoming so popular that some interesting models for simulating stress are being proposed. This would allow us to reach the benefits without taking the associated risks. I had included the explanation of these models (as well as its tentative application to dogs) in the original draft of this article but my webmaster reckons that my articles should not be longer than two pages and you have already read three. Hence, I will publish this as a second part in a few weeks. Hope you enjoy it!

I promise you that I will write about our work at the Zoo soon. However, I owed this article about stress to Eliseo Rodríguez

Share Button

Negative punishment, a further step forward

Publicado el

Negative punishment is one of the most widespread and advertised conditioning processes. It consists of the removal of a positive stimulus contingent with the emergence of a conduct that we wish to eliminate.

It is important to clarify from the outstart that negative punishment is not a sort of panacea to get rid off every kind of conduct. It is only useful with regard to operant behaviors, that is, those behaviors which have no self-reinforcement processes attached so that their emergence depends on the association with other reinforcers. Thus nobody should harbor hopes of getting rid off a behavior like digging in the garden by letting the dog doing it at will!

Doubtless it is very important to know the scope of application of a process. However, it is equally important to know the cases where that process is not effective. Few things have been as deleterious to the public image of negative punishment as its indiscriminate prescription also to sort out emotional problems and problems associated with self-reinforcing behaviors. Anyhow, the reputation of this process among professionals is excellent. The reason is that it enables us to consistently eliminate many conducts without generating conflicts or resort to aversive stimulation.

There are several ways of applying negative punishment but the most popular one consists of the suppression of the positive stimulus acting as a reinforcer for the targeted conduct. As a rule of thumb we advise clients not to reinforce dogs when they offer the unwanted behavior. In fact this is a mixture of negative punishment and extinction that will only yield slow results. We can speed up the behavior change if we organize things in such a way that dogs are being positively stimulated before engaging in the problematic conduct so that the pleasant stimulation can cease as soon as the problematic conduct begins.

For instance, suppose that we want to prevent the dog from jumping on persons to greet them. A simple way of doing this is to caress and pamper your dog while sitting on the floor. As soon as the dog jumps on you, you will end up the session by standing up. This intervention requires more preparation but it produces faster and more consistent results than following the typical advice of “ignoring dogs when they jump on you and pampering them when they stay with all four paws on the floor”. The reason is that with the latter advice dogs learn that they should jump once in a while to trigger the owner’s sequence of behaviors. It happens very often with regular owners, not with seasoned professionals hopefully, and this slows down and even blocks any progress. By organizing things in such a way that the positive stimulus is always present before the inadequate conduct emerges we will clean up the dog’s head, as well as the owner’s! This way a crisp and fast improvement will be achieved.

A more rare application of negative punishment may be carried out through satiation. Satiation is an application of negative punishment and it consists of keeping the positive stimulus active until it is no longer positive, either because of its prolonged administration or its intensity. Very often this is the most entertaining, effective and easiest way to eliminate certain operant behaviors. I sincerely think that my clever use of negative punishment through satiation is responsible for many of the private customers I have taken away from my competitors.

Whenever I had to go to an interview with potential customers and their dog kept on jumping on people I knew that no other firm would cause a better impression than ours. Why? It is simple. Customers would be confronted with three different strategies from our competitors. As we will see, even if all of them are effective when they are applied well, they are less amusing and clients do not appreciate them as much.

  1. Negative correction by the trainer like blocking the dog with the knee, bothering the dog in the hind legs or stepping on the leash to prevent the dog from jumping. Most of the clients do not like this strategy and, in fact, it may provoke a strong impression on many dogs. Furthermore, it requires skill.
  2. Punishing through suppression, which means to ignore dogs actively until they stop jumping. Once the bad behavior stops, the trainer will reinforce the dog. In addition to the problems I have already pointed out with this kind of intervention, some customers do not enjoy watching how their dogs keep on insisting. This phenomenon is due to the shape of the extinction curve and the over-arousal caused by the change of attitude from a social partner. Moreover, on that first session you can forget about achieving satisfactory results, judging them from the client’s point of view.
  3. Counterconditioning the dog with other conduct like sitting down. This is not always easy for clients on their first training day. They will be overloaded with information about what to ask their dogs for, what to teach them and what to assess.

So when I arrived with satiation in my toolbox, I said to my clients: Ok, your dogs want love and affection, let us give them that! So I told them to hug their dogs, and caress them enthusiastically. A few minutes later the dog would say: hey, that’s enough! But of course, we would keep on loving them!

The owners had fun doing this, it felt easy even on the first session and it produced faster results than any of the alternative strategies. I insist that the other strategies also lead to excellent results if applied correctly, but they require more time and clients like them less at the beginning.

I remember how funny it was for a client whose border collie was a sofa squatter not to let him walk down so that he would give up the habit of using it! Of course do not try this with lazy dogs, you may need some thirty hours for the intervention to work. It is very important that you adapt your techniques to each individual, as you wouldn’t try to fit a square peg in a round hole.

I really hope that you have enjoyed this article because I have unveiled one of my top commercial secrets to write it!

Share Button

About luring

Publicado el

In a previous post Richard Ibarburu commented about the excessive use of luring as an example of positive reinforcement based strategy that could potentially hinder the emergence of didactical choices. Richard’s wise comment made me realize that nowadays dog trainers, including me, always refer to luring to emphasize its drawbacks rather than its virtues.

Not long ago “Security Dogs”, a well-known dog training center, kindly invited me to a workshop they organized with Fernando Silva. At a certain moment Fernando made use of luring to induce a dog to perform a certain conduct. Two of the students looked at each other and one of them said: “luring!” Almost immediately a mock of deception followed in their faces. On top of that, the teacher himself explained almost apologetically to the audience: “This is too much luring for my taste but sometimes you have to do it”. It is my belief that at that particular moment and with the work Fernando was doing no other technique would have been better and more effective than luring. My claim makes sense if we take into account the high level of dog training that Fernando always exhibits.

So the question remains, is it really luring that bad? How did we reach the point where we all make bitter remarks as soon as we refer to this training technique? To answer these questions we have to analyze what is luring about as well as its effects on learning.

Luring is a training technique that consists of guiding dogs to perform a behavior directing their attention with a primary reinforcer, usually food. The continuous presence of the primary reinforcer contrasts with other techniques. Two important aspects derive from this definition:

  • The dog’s attention focuses on the primary reinforcer whereas other present stimuli lose salience.
  • Dogs feel that they are being continuously reinforced. In other words, reinforcement is not circumscribed to the point when they actually obtain the reinforcer.

From these considerations stem some important consequences regarding training. Such consequences should not be taken as good or bad in themselves. The right strategy is to assess in each case whether luring is playing in favor or against our training goals.

One of the main effects of using luring is that the dog will concentrate on the food to the detriment of other environmental stimuli. In other words, the dog may not perceive the clicker or even our commands, what can delay or worsen the learning of associations. On the opposite side of the scale, luring can be useful to help inexperienced dogs disconnect from the environment and learn to concentrate on their work. The latter effect may be of much use in private commercial training sessions in order to be able to work with dogs in the park where they play!

It is true that the timing is worse as the dog feels that he is being rewarded all the time. This may prevent us from giving full value to the dog’s higher quality behaviors. However, this effect can be useful to promote the establishment of a positive emotional association of the dog with the work. In some cases this is far more important. Luring is also useful to manipulate dogs with whom we have no previous bond, as it is the case in commercial training sessions.

In order to gain a broader perspective on this subject I am going to point out the benefits and drawbacks associated with the use of luring. In this way you will be able to understand when it is advantageous to rely on luring depending on three types of considerations: the specimen, the training stage and the type of work you are carrying out.

Benefits:

  • It helps training positive focus in dogs who have difficulties in concentrating for a prolonged period of time.
  • It promotes concentration in difficult environments and situations, a feature that as we have already seen can be useful to work in the park where the dog regularly plays and walks, or also when the training field is crowded with other dogs. Some trainers boast of working without resorting to luring, but then it turns out that their dogs are unable to work if they are not all by themselves and in complete silence. Wouldn’t it be more practical to plan for focus training during a couple of days relying on luring rather than to keep on organizing this sort of “mystical” training sessions?
  • It relegates certain stimuli when we do not want the dog to perceive them. This may help dogs in overcoming small lacks of confidence. In aggression cases, it may also allow to deviate the dog’s attention from other dogs or persons. At the same time we will be achieving a positive emotional state and, as dogs feel that they are being rewarded all the time, we are also reinforcing them for not offering aggressive behavior. This may work as a good kick start to counter-conditioning processes in the frame of behavior therapies.
  • It helps to train motor patterns as opposed to final conducts. In certain behaviors related to sports our aim is not so much the final position the dog reaches but how such position is reached. Thus I am not just interested in teaching the dog to lie down but in the dog performing a given motor pattern that ends in the lie-down position. The difference between final conducts and motor patterns is of the utmost importance for sport training. In this regard, to attain the desirable motor patterns from the beginning and to fix them in the dog’s muscular memory, luring is simply irreplaceable.
  • It would be difficult, if not impossible, to induce certain conducts without luring.
  • It is very useful to mitigate hesitations and lacks of confidence. Some dogs may want to work and to keep progressing but if they are sensitive they can become loaded with tension. In relation to hesitant conducts, greater consistency is achieved by using luring. Subsequently, on this more solid foundation, one may return to working with other training techniques.

Drawbacks:

  • Poor perception of the environment, including conditioned reinforcers like the clicker, the commands, and even the handler! This will pass its toll on the quality of the work which will be slower and less clear in relation to the parameters mentioned.
  • It promotes maximum concentration on selfish behavior engines. Since dogs are following the lure, they will fail to perceive the need to synchronize with their guides or to pay attention to the handler’s indications. Affection is also dramatically relegated. In sum, the coordination and activation of social engines is absent.
  • Poor perception of signals and subtle stimuli. Dogs are so absolutely focused on the primary reinforcer that they will be blind to other indications we want to introduce to fine-tune or to progress in our work.
  • It is “hard” to switch to other ways of working the behavior. Transitions are one of the main problems with luring and many dogs experience severe problems with them. However, once again, we can exploit this weak point to our benefit. This is what we do in our training proposal by introducing the so called “counter luring” step. It consists of providing dogs contradictory information, the lure may tell him to sit while the handler says “lie down”. This way we train our dogs to give priority to the latter so that they become more pro-social and less focused on individual reinforcers.
  • There is no activation of problem solving abilities. Putting “the carrot in front of the donkey” (and I am quoting from a comment to a previous article) does not engage problem solving skills at all.
  • Self-reinforcement is also completely absent from the picture because the dog is reinforced continuously and problem solving is not engaged.
  • Learning resulting from luring is rough and unrefined. Since dogs’ attention is totally focused on the reinforcer, they will only learn the most evident parts of the trained behavior. Subtle steps ahead or modifications to the main conduct will go unnoticed.
  • Dogs do not act proactively, they merely react. Dogs depend totally on us and other external stimuli to perform the trained conduct.
  • In this way, the autonomy of dogs is undermined. Hence they will be unable to work without strong support and will not take the initiative to offer the behaviors.

Overall, we should take into account our training objectives and our training possibilities to assess whether relying on luring may be of help or not. Sometimes luring may act as a false friend. Beyond its apparent simplicity it may hinder the achievement of the results we are aiming for. Keeping its pros and cons in mind will help you to take the right decision to use it or not. This way you will be able to proudly say “yes, I also use luring”.

Share Button

Not only dogs: Now also sea lions, parrots and dolphins

Publicado el

Hi everyone,

In a previous post I hinted that we were on the verge of closing a very important agreement for EDUCAN. Today I can finally announce that the project is already on its feet!

We have signed an agreement with ZOOS IBÉRICOS, an entity that belongs to Parques Reunidos which is the company that owns more zoos and aquariums all over the world. The purpose of the agreement is to implement our cognitive-emotional training methodology with various species. We will start with sea lions, parrots and dolphins at the Madrid Zoo.

Since I am thrilled with the project I would easily exceed the limits of the post. To prevent it, I will just quote the section of the agreement with ZOOS IBÉRICOS referring to this partnership. I have edited the document just adding bold characters to emphasize what I believe are the most relevant pieces of information. This will also please my “webmaster of the universe” who always insists on me to use bold characters to emphasize important passages in blog articles.

OBJECT OF THE AGREEMENT

The main object of the partnership between the two parties to the present agreement is to design, implement and evaluate the results of the application of new training protocols for various species. For this purpose the main yardsticks will be the three pillars of the cognitive-emotional training methodology: (1) the exploitation of the cognitive skills of each species, (2) the assessment of their emotional state, and (3) the understanding of their unique social behavior patterns.

This main objective can be broken down in two different operational aims:

1. Improved animal welfare:

The new protocols will pursue the improvement in the quality of life and working conditions of the animals. Regarding the choice of behaviors, both parties agree to give priority to animal welfare over other considerations like the appeal to the public and how spectacular the chosen conducts seem to be.

According to recent ethological findings animals generate conduct differently depending on whether they are in an environment where they have to fight for survival (i.e., scenarios characterized by predatory risks, resource scarcity or the need of an active defense of the territory) or in a welfare environment (i.e., scenarios where there are plenty of resources and no foreseeable risks for the animal).

Most of the current training techniques, including many of those relying on positive reinforcement processes, are based on the survival environment paradigm, what worsens the results even with animals who enjoy an optimal quality of life.

One of the main technical and ethical purposes of this partnership is to develop working protocols based on the way animals generate conduct in welfare environments. This means that animals will not work driven by the need to ensure their survival but solely by the motivation to improve their physical, emotional and social welfare.

2. Research:

The aim is to apply state-of-the-art knowledge on ethological issues to animal training, as well as to evaluate the results deriving from this implementation.

The present project is pioneer in:

a. Exploiting the cognitive processes known in each species for the purpose of training animals. Until now operant conditioning has been the main working tool for animal trainers. Operant conditioning has the advantage of being applicable to all of the species typically involved in animal training programs. However, cognitive ethology has proven that different species can have various mental processes like problem solving, learning by imitation (as shown in Pepperberg’s rival model in relation to psittacidae). […] “[B]y taking advantage of these specific capabilities the quality of the work will increase. Since many of the referred cognitive processes are self-reinforcing, their use drastically reduces the need to rely on external sources of reinforcement and helps attaining more consistent conducts while improving the emotional state of the animals. In other words, animals enjoy offering the behaviors and see them as goals in themselves rather than as mere gates of access to food (i.e., the normal state of affairs when primary extrinsic reinforcers are relied upon). Cognitive processes have the further advantages of allowing self-assessment to animals and improving their intrinsic reinforcement capabilities.

A further advantage of the new methodology is that training sessions will work as environmental enrichment interventions thereby improving animal welfare. In addition, fewer sessions will be needed to maintain the behaviors, the improvements will be achieved faster and the quality of the work will be more solid.

b. Evaluating and fine-tuning the emotional state of animals during training sessions and shows, decreasing their levels of distress and improving stress management by trained animals, promoting positive emotional states in them and developing reliable instruments for tracking emotional welfare in trained animals.

c. Developing protocols to take advantage of those social processes characteristic of the species, like affective bonding, so that the interaction between animals and trainers in the course of the working sessions and exhibitions becomes a desirable and self-reinforcing objective. Nowadays we know that affection is an important drive in many species of social mammals. Notwithstanding the fact that many trainers have exploited these aspects intuitively for a long time, this working methodology has not been tested in a systematic and scientific manner. Thus one of the objectives of this partnership is to create protocols for those species covered by the project.”

…and a last excerpt from this agreement: “one of the tasks entrusted to EDUCAN consists of:

– Designing protocols, schemes and working techniques for different species.”

Overall, it constitutes a VERY exciting project which is going to demand lots of work from us (probably we will have to restrict some commercial activities to be able to sleep once in a while). However, it is a huge step towards the long due paradigm update in relation to animal training.

I am very grateful to Miguel Bueno Brinkmann, biologist and head curator of birds and sea mammals at the Madrid Zoo, and Pablo Roy, head trainer for sea lions at the same institution, for their interest and invaluable help in making this project come true.

I AM DEFINITELY THRILLED

Share Button

To suceed or not to fail?

Publicado el

Let me explain to you which are the main categories I use to classify animal trainers. There are of course different training schools, different modalities, and various kinds of stimuli that different trainers choose to apply or not to apply. These are all important considerations and yet they are less important than the distinction between the following two categories: (1) those who train their dogs not to fail, and (2) those who train their dogs to succeed.

I want to clarify from the outstart that I am persuaded that dogs should be trained to succeed, and not merely to avoid mistakes. For some years I followed the opposite path and this is perhaps the reason why I am convinced that my current position is the right one. We all know that nobody can be as convinced as a converted.

The vast majority of trainers sponsor the competing view, errorless training. They tend to think that errors are dangerous and hence you hear sentences like “don’t let your dogs make mistakes if you don’t want them to learn incorrectly”. This is also linked to the fact that we feel uneasy watching how our dogs make mistakes in the execution of the behaviors we are teaching them.

Thus preventing the dog from making any mistake is the priority for many trainers. This means that our energy and attention are focused in the task of monitoring errors and choosing techniques to prevent them, block them or correct them. This way dogs finally succeed because of the many things they do not do, not because of the things they are actively trying. Training sessions are devoted to create “secure situations”, to close doors, to limit options…

There are several problems with this way of training which, by the way, is not circumscribed to trainers using aversive methods (many trainers who adhere to positive reinforcement based methodologies, even cognitive trainers, sponsor the errorless training view). Here is a list:

First, after a period of time working under these guidelines dogs, who are not silly at all, realize the nature of our goal. Their voluntary attention is then focused on locating the behaviors they should not do. This way they become more effective in not doing than in doing. This is the corollary of the general work scheme we are implementing and we ought to be comfortable with that by-product.

Second, errorless training is exhausting from the emotional perspective. Just think about one of those movies where a young and promising sport star is pressed by the trainer/father/agent who takes great care in underlining every error that is being made. Of course the demanding figures are full of true love and good intentions for their pupil, but as this sort of torture goes on and on, the rising star’s original motivation is killed. Then, the future star decides to send everyone to hell and runs away to Idaho with his or her better half to settle a ranch.

Our dogs cannot run away to Idaho to settle a ranch even when they get fed up of training with us. Therefore, we must be very careful and take seriously our responsibility to prevent the dog from getting emotionally overloaded. This does not mean that one should not demand effort and implication from the learner. However, we should realize that by emphasizing errors we will be undermining our learners with insecurities and draining their emotional energy. Imagine how you would feel if you had to go to a job where your boss would remind you EVERYDAY of the mistakes you make or would tell you to be very careful not to make any. That’s not the way to build up self-confidence, implication and work-team.

Third, even if at the beginning it is easier to train the dog not to fail (learning not to do is faster than learning to do), we will soon realize that there are so many possible mistakes that it would be impossible to cover all of them in a person’s lifespan. In contrast, there is only one successful conduct in each case. Hence it is much more comfortable and expedient for both dogs and trainers to focus on that.

Summing up, in my opinion it is very important to build the dog’s head to succeed. For this purpose one should allow initial mistakes and inform the dog about them without worrying. Errors are a necessary part of active learning. We should devote our attention to the right answers so that the dogs’ attention is also directed to them.

Later on, when dogs consistently work to achieve success and their mental scheme is already built up, we will be able to give more importance to those mistakes that arise since this would not create other problems.

Share Button

Training in welfare environments (I)

Publicado el

In a previous article I commented about welfare or comfort signals which can be used to evaluate how dogs are feeling. My topic today regards one of the main pillars of our work: the different learning and conduct generation that dogs exhibit depending on whether they are in a survival environment or in a welfare environment.

Given the importance of the topic for the design of training techniques, this will only be an introductory article. Actually the methodologies for teaching and learning vary dramatically between both scenarios. An adequate understanding of the underlying relationships will enable us to make a systematic treatment of the mechanisms that guarantee that dogs enjoy working, as well as to replicate the training, without excessively relying on the talent of either the dog or the trainer.

First, I need to clarify that working in one type of environment or the other does not determine the quality of the work. In both scenarios we can rely on either positive or negative stimuli to promote learning. Then, what is the difference? What varies is the way these stimuli are used.

  • In survival environments animals have no guaranteed access to resources (e.g., food and water) and their safety is not taken for granted either.
  • In welfare environments animals have abundant resources as well as guaranteed security.

As noticed earlier, some species –including dogs and human beings- have a different way of learning and generating behavior depending on whether they are in one kind of environment or the other.

It is important to realize that in welfare scenarios the aim of dogs is to have a good time, increase their welfare, whereas in survival scenarios the aim is to obtain resources crucial for survival and security. The behavior differs and so the learning does.

It is crucial to understand this bifurcation because probably most of us live in a welfare scenario. Let me put an example. Almost certainly some of you have been looking for a particular item, even visiting different shops for days in order to find it. It does not matter what the item in question was, that damned Charlie Patton Yazoo L-1020 record, a particular book or some other thing. Probably when you returned home you found an empty fridge. You probably thought, “I’m exhausted, I don’t feel like going to the supermarket at the moment. I’ll do it tomorrow.” Summing up, we are capable of generating an enormous amount of conduct to get something which is not an actual need and yet we refuse to walk round the corner to buy food, that is, a primary resource crucial for survival. Isn’t it weird? No, it is not. The explanation is simple.

We know that we are not going to starve to death. In fact, we can probably obtain food whenever we want to. Once a primary resource is guaranteed its value as a motivator decreases. In contrast the self-reinforcement value and the extra-quality of life represented by our desired item gains a greater value as a motivator. This is one of the first differences between welfare and survival scenarios. In a welfare environment any thing or event that is self-reinforcing effectively work as positive stimulus, whereas primary stimuli lose effectiveness. In survival scenarios the opposite holds.

A second difference between the two types of scenario regards the way animals manage negative stimuli. For a negative stimulus to trigger conduct in a survival environment, it has to be clearly perceived by the learner. Moreover, if it is to work, it has to give rise to a negative emotional state that warns the animal about a possible risk. In contrast, in a welfare scenario it is much more effective to rely on low intensity negative stimuli. These are capable of interfering with the enjoyment of pleasant activities but without overruling the positive emotional state stemming from the latter. As a consequence, dogs will generate a great amount of conduct to offset the interference to keep enjoying the pleasant activity. This is the reason why it is so effective to introduce minor bothers in shaping sessions (e.g., attaching a post-it or a rubber band to the dog). This strategy also helps dogs to confront problems and teaches them to manage stress.

It is important to know that we cannot jump from one kind of environment to the other at will. We have to choose one of the two categories for each kind of work and be coherent henceforth. If you plan to rely exclusively on food as a reinforcer, it is better to propose a survival environment, because then food will have greater value and the learning obtained will be more consistent and of a higher quality. On the other hand, if your dog has a great time solving problems and experiences training as an aim in itself, pick up a welfare environment.

Share Button

Teaching the dog to go away

Publicado el

Doubtless the most important behavior we can teach to our dogs in order to improve their quality of life is to come when called. A reliable recall, one that works even if the dog is distracted in other activities, allows handlers to grant dogs free range so that they can explore the environment.

In this article I will not tackle the issue of how to train a good recall. My topic for today regards some of the risks incurred when training a bomb proofed recall, as well as the way I work with puppies before starting recall training to avoid these side effects.

Many years ago Jaime Parejo and I were in close contact. Jaime is, by the way, a terrific trainer, a re-known specialist in search and rescue (S&R) and a person I deeply appreciate. At that moment he was writing the book about the “chest method” (“método arcón”) that would gain him recognition all over the world. We used to speak lots about his work of course. I remember Jaime would use the term “yo-yo effect” to refer to the tendency of some dogs to come back to their handlers once they had reached a certain distance threshold. He had realized that this problem derived from excessive recall work and that it was very harmful for S&R dogs as it severely limited their working autonomy.

Jaime hit the bull’s eye with this observation, a big share of the population of trainers’ dogs were not able to go away beyond a certain distance. I have been always more interested in the identification of the reasons why dogs behave in a certain way, and in taking advantage of these tendencies, than in designing specific techniques. Hence I analyzed the problem from this perspective. I realized that, until then, and mostly in an unconscious way, I considered such effect to be beneficial. After all it is the dog who takes the trouble to stay close to the handler. However, after reflecting on Jaime’s words, my opinion changed and so did my way of training recalls and, in general, my way of taking care of dogs.

On top of my affection for him, Jaime will always have my recognition for helping me gaining this insight. My dogs have lived happier lives since then.

Dogs who exhibit this “yo-yo” tendency do not enjoy their country explorations as much as they could. The physical exercise they practice is also limited and so is the outlet from stress that they enjoy. After all, they cannot give free reign to their innate motor patterns.

This is a common problem with dogs whose guardians are professional trainers and, as I have said, it used to be a problem with my dogs. These dogs are always paying attention to their handlers and they need that the latter provide them instructions to be able to amuse themselves. If they run after a rabbit by any chance they will come back overwhelmed because when the chase is over they will find themselves far from their handler and this causes them bad feelings. The problem with some other dogs is the opposite: they have the time of their lives when they are in countryside and come back when they feel like it more or less. However, I will tackle this second category of problems in a future article, not today.

To achieve welfare with active breeds it is crucial that they are able to run freely and explore vast areas. The notion of health as the absence of diseases has long being overcome: To achieve welfare the focus should not be on avoiding stress and anxiety. For dogs to blossom, it is far more important that they can live full and happy dog lives.

Nowadays the first thing I teach to my puppies is to go away rather than the recall, so that they can fully enjoy their walks. Puppies absorb every bit. Puppyhood is a stage when it is very hard for us trainers not to make our dogs too dependent on us. It is difficult for us just to walk them without any training. And if we give up to the temptation, then it will be very tough for the puppies to be at ease unless they are engaged in some kind of activity with us.

I have to recognize that for me it is very easy to train the behavior of going away: the back door of my garden opens directly to the countryside, plus my adult dogs already master this conduct and the puppy will willingly accompany them in their adventures. During this stage I am very careful not to make anything that promotes dependency: I don’t hide so that the dog does not have to worry about where I am, I don’t reward the puppies every time they approach me, I just walk and let the puppies realize how wonderful and interesting the world is, full of different smells, sounds etc. I want them to experience how good it feels when they run, jump and learn to use their bodies. I want them to behave as dogs with my adult dogs, that they learn that in the countryside the most amusing partners are other dogs, not me.

It almost looks like the reverse image of the guidelines to build the recall one would read in many books. And yet I am very proud of my recall work. As a matter of fact it is the behavior my colleagues ask me about more often when they see how consistently my dogs respond. My dogs do not expect any other reward than naked social reinforcement and as soon as I release them, they run away and start exploring once again without keeping an eye on me in case there is a new recall in the pipeline. However, if I called them once again, they would come without the slightest hesitation.

Of course I recognize that the privileged conditions where I train play a prominent role in making this outcome possible. I can train in a secure area. I know that my adult dogs will come when called and that the puppy will follow them.

As I said earlier, I have always been attached to the idea that if we know how a mechanism works (even a negative one) we will be able to exploit it to our advantage. Thus, notwithstanding I make every effort to prevent my dogs from acquiring the “yo-yo effect” on a permanent basis, I apply a technique that allows me to engage this mechanism at will so as to limit the range of movement dogs enjoy temporarily. It consists of recalling them many times in a row at the beginning of the particular walk, ten or twelve times will suffice. I recall them every time they reach the distance I need. This way the dogs know that they should not go beyond that limit in this particular walk and I can adjust the distance to unusual environmental conditions. However, I never start practicing this kind of work until the dog has learnt to come when called without being too dependent on me.

If you empower your puppies to go away and to behave expansively, you will improve the quality of your walks with them. Not only will your puppies have more fun behaving as real DOGS (with capital letters), but you will also grin and amuse yourself contemplating how they blossom. Take my word: dogs deprived of their autonomy could never be as happy.

Share Button