off site links
Rethinking the way we look at the relationship between dog and handler allows the concept of the handler being the Alpha dog to be put aside. A new approach is offered that works to build the bond between the handler and the dog, a bond built on need and trust rather than hierarchy and obedience. This leads on to strategies for building this bond and in so doing creates a powerful training methodology.
The concept of 'Alpha dogs', 'domination' and its application and the ability to make it work when the dog is at a distance has never sat well with me; likewise 'behaviourist training' denies a dog any ability to think for itself; this does not fit with what I have observed. However the dog training world seems to have split between these two major ways of thinking. Is there a 'third way'? A Dr of Psychology specialising in the effects of pets as therapy when asked the question 'what are the differences between the relationship of a pet handler to their 'pet' and a working handler to their 'dog' came up with a clue. A pet dog's relations are interpreted using 'attachment theory' whilst a working dog's relations can be studied using 'dominance theories'.
So what is attachment theory?. The concept was first postulated by child psychologist John Bowlby and elaborated by Mary Ainsworth. The fundamental tenant being that a child in a balanced relationship will 'seek proximity' to its caregiver. Countless tests have since been undertaken using this theory to prove the veracity of the ideas. They have also been repeated on the relationship between handler and dog and dog to handler (Topal. Kurdek). Would a dog that seeks proximity to its handler be an achievable dream? The main areas of the theory that may prove interesting to dog handlers apart from the concept of sought proximity are the ideas of 'secure base' where the 'attachment figure' is seen as a dependable source of comfort and one who resolves any problems associated with exploring the world. 'Safe haven' where the attachment figure is sought out for contact, assurance and or safety when distressed and 'proximity maintenance' staying close for the sheer pleasure of being near. The other side of these being 'separation distress' when the attachment figure is missing or out of reach. Grandin documented the similarity of mental processing power between an autistic child with impaired connections between the brain's frontal lobes and a dog who has underdeveloped frontal lobes thus giving clues that may allow our understanding of the learning patterns of autistic and ADHD children and dogs. Further our main emotion controlling brain structures are the same as those of the dog (Psychoeducation.org). Whilst a child is not a dog and the suggested similarities may be coincidental the conclusions might prove of use and interest as well as translate to the world of the dog handler.
One of the most basic lessons currently taught a dog is to come to its name, without this the conversation between Alice and the gnat springs to mind. We 'teach' the dog 'obedience', to come when it is called. What if we could turn the tables and have the dog naturally 'seeking proximity' to us? What, if anything, could we teach it so that it wanted to be close? That it became stressed if separated? Babies and for that matter puppies develop a bond with the mother from birth – the care giver provides food, warmth and security; enough reason to stay close if you are insignificant and helpless. However what would one do as a new 'parent' of an older child in a blended family? Could an adult build a need for proximity into the younger person under these circumstances? Soldiers develop strong bonds after being involved in a fire fight. Are there any clues offered by the relationships developed between hostages and their captors? More importantly is any of this going to be useful in dog training?
The look for answers started with research into 'mother and toddler'
parenting skills. Touch, contact, feeding, warmth, security, bathing,
ritual. Advice on building relationships where there was no history
of relationship came up with 'receive eye contact, give affection
in the face of apparent rejection, respond every time, create and
hold behaviour boundaries'. Denial of all these create stress, ranging
from mild to acute. Bonds of attachment in military and hostage groups
seemed catalysed by shared experience and more particularly dire
stress – threat of loss of life and escape from this. With
hostages such a bond is formed, that occasionally the hostage will
apparently freely join the captors – Patty Hurst being an example.
It is these in extremis cases that pointed to a way forward, allowing
'Attachment goes to the one who relieves the stress'.
Stress is a body's reaction to a change requiring a physical, mental
or emotional adjustment or response. Stress can come from any situation
that provokes feelings of frustration, anger, nervousness, or anxiousness.
Selye working in the early 70's developed a model for dividing stress
into 'distress' and 'eustress'. Eustress being where the use of physical
or mental stress is used to enhance function. Perhaps there could
also be a division between 'passive' and 'active' stress? Passive
being stress caused by state eg cold or hunger whilst active stress
is created by something more transient like a shout or a touch. Could
stress be manipulated to encourage and build attachment bonds? Food,
shelter and warmth are a given – indeed I would suggest a fundamental
right of any child or human owned dog. Denial of any of these to
create stress and allow reinstatement should not be an option. (Is
this the initial reason lost and found dogs respond so readily to
their saviours). The continual provision of these basic needs via
one provider will no doubt help create a baseline level of attachment
to that person, as such a kennel hand or lab assistant providing
food, water and exercise for basic needs would fit in this category.
An owner or handler should offer more than this. The initial training
of birds of prey relies solely on this. For babies – contact
and ritual – doing the same things the same way – being
predictable creates a sense of security just as it does for those
with special learning needs (Grandin); for a dog is it any different?
The same ritual and daily patterns of behaviour – and expectation
of a behaviour help define the world and create security by predictability.
Safety and security are core issues with children within changing families; these children keep their guard up and can be distant and distrustful. The advice given on self help web sites is to create clear boundaries of behaviour and respond to infringements in a clear and consistent way – the child then knows what to expect and also knows that whatever happens the care giver can be counted upon to be there – should it be any different for a dog? Limits and boundaries define the world and thus make it less scary and allows the development of behaviours linked to consequence. To be allowed, for example, to be on the sofa one minute and not the next must be quite disorientating. If a weed is seen as a plant in the wrong place then a bad behaviour could just be a particular behaviour at the wrong time. It will depend upon the handlers attitude – are they seeking 'obedience' ore are they looking for a 'bond'? Correcting a child with an attachment problem will only make it feel more unsafe and worse the action may reinforce the behaviour as it has been seen to get a response. Is it any different for a dog?
Lacking speech, early social relationships with babies are built using eye contact (Challoner). In adult life how often does one avoid eye contact to avoid engaging in conversation? How often when a dog looks to a person does that person respond with voice, or touch or not as the case may be? Babies need touch and eye contact to thrive - perhaps dogs do to? If a handler does not respond in some way to the dog's 'look' then the dog will cease to look or in the words of operant conditioners 'extinction' will take place. As an aside very young babies will also smile to a head nodding in an affirmative manner (Challoner).
Non threatening methods of creating attachment can be seen to include touch, security, food, eye contact, boundaries, ritual and predictability amongst others. However earlier it was mooted that relief of stress was a powerful bonding promoter – the direness of the stress and the strength of the relief creating a proportionate bond. Escaping the threat of death is likely to be the most powerful. The Alpha roll and that ultimate near death experience historically touted - 'helicoptering' 1 being extreme examples; both are considered to work, from the arguments above it is possible to see why. Confusion may well reign when the aggressor and the reliever are the same person; Dr Jeykell and Mr Hyde have been around for a long time. Could putting in place controlled stressors offer a useful method to manipulate the desire to be close. Behaviourist training methods, of which clicker is the most well known, relies on reward for an action, it is a method for shaping a behaviour. You could clicker train a dog to return when called – but is the dog returning because it wants to be with the handler or because it gets involved in a series of behaviours that culminate in its return? The training is undertaken by purely positive means and no 'aversive' methods are used. (I propose an aversive is anything the dog does not wish for at that particular time. If it was not hungry then offering yet more food would become an aversive.) Aversive has connotations of 'hurt' and stress is often associated by many with the idea of punishment. If instead one were to think think of 'pressure' instead of 'stress' or 'aversive' then a new lexicon may be available to the trainer. A dog that is interrupted from doing what it wants with a view to improving its performance or behaviour is being subjected to a stress or eulistic pressure. A training method that interrupts, informs and then rewards would thus prove doubly powerful – the dog would learn and attachment would grow.
Looking briefly at the idea of the 'retrieve'. A dog is classically
taught to 'retrieve' on 'command' a concept bound up with the words
'discipline' and 'obedience' - but eating in the wild is quite dangerous,
an awareness of proximate competitors and predators alike is needed,
what if retuning to the safety of the 'attachment figure' allows
the dog to eat safely? A 'retrieve' could now be seen as the dog
wants to bring itself and its quarry to a safe place because it is
to its advantage, not because it is told to do so. No need to worry
about the dog running off with it, running off on its own is not
now part of its language.
Other training theories can also be explained with the above concepts. Premack for instance. Here the dog is wanting something and learns it will not get it until it has fulfilled the given task. In a state of tension by its own desire perhaps but under pressure none the less, the handler giving / allowing it the self reward. Here working dog handlers have a slight advantage as the dogs are usually highly selected for particular habits, the training being the refinement of these traits. The use of an electric collar as a distance control tool becomes a mere punishment whilst the increased drive associated with force fetch fits well with the above argument as the high levels of stress, in this case caused by pain, is relieved by the handler on completion of a task, learning and strong bonding are achieved
Make a little thought experiment. Assume you are trying to learn the piano. Which teacher would you prefer and under which regime do you think you would learn fastest? A. Every time you make a mistake your teacher raps you over the back of the hand with a ruler. B. Only when you do something right will your teacher tell you you are right. Or C. Every time you make an error you are stopped, the error explained and then after completing correctly you are praised. I would suggest C is the most powerful as not only do you end up knowing what is right but you also know what is wrong and why AND you feel good about it. Which style do you adopt when dog training? Which would the dog prefer?
Applying the principles outlined above it can be seen a very powerful cycle of learning can be created. For a dog to learn a new skill at the most optimum rate it must be taught at a rate that suits the dog, deviations from the learning path must be stopped quickly and the error/s must be corrected even if the size of the task is reduced. Success must be recognised and rewarded. It is considered dogs don't have the ability to generalise so these skills need to be practised under a variety of conditions. Dogs in a state of slightly elevated stress will, as do humans, respond more strongly.
The above idea could be summed up in a four letter acronym PPPP.
Pressure, Prompt: Praise; Practice.
Pressure – stop the dog from continuing the mistake and in so doing mildly stress the dog. The dog after all would rather carry on with what it was doing. The way it is stopped affects the level of stress.
Prompt – help the dog into the required behaviour by getting it to try again, if necessary break the behaviour into smaller sections.
Praise – reward with praise the good result using both physical and verbal means.
Practice – do the exercise within as many different scenarios as possible.
Throughout the dog's life maintain the behaviours by regular practice.
A stress, when relieved by a care giver will enhance an attachment bond. The greater the bond, the more secure and safe the relationship. The more secure the relationship the greater the desire for continued proximity.
Strength of attachment can be manipulated by always responding to the dog's desire to communicate, be this visual or physical; going out of ones way to deliver verbal and physical praise. It can be enhanced by creating and maintaining behavioural boundaries; in the world of Alpha perhaps these could be seen as 'gang rules' that are being adhered to in an effort to stay accepted? It can be built by following rituals; for instance 'manners' at a doorway or behaviour when being fed. Predictable behaviours on the part of the handler encourage the sense of security. Finally it can be enhanced by indulging in joint activities that involve new learning of new skills and by the repetition and maintaining of old ones; physical and mental interaction can take place to the emotional benefit of both parties.
Improved bonding creates a state where the dog performs because it 'needs to' rather than 'must do'. The dog no longer a subservient to a theoretical domineering alpha but a willing partner within an effective dyad.
Learning can be maximised when a balance is drawn between stress and relief. Particularly when the learner is in a slightly elevated state of eustress.
It is therefore argued that the difference between a 'pet' dog and a 'working' dog is only one of degree. Dogs displaying great 'obedience' also display a greater than average 'bond' with the corollary also holding true that a stronger bond will produce greater apparent 'obedience'.
Thus the building and maintenance of attachment bonds takes away the need for hierarchies and takes away the need to control by domination.
© Guy Bagshaw. MSc. March 2010.
Sources and bibliography.
Caroll. Alice in Wonderland http://www.alice-in-wonderland.net/books/2chpt3.html What's the use of their having names the Gnat said, `if they won't answer to them?' `No use to them,' said Alice; `but it's useful to the people who name them, I suppose. If not, why do things have names at all?' Accessed 01 March 2010
Challoner http://www.scribd.com/doc/9378121/Baby-Smile-Revised-1208 accessed 01 March 2010
Grandin Animals in Translation: Using the Mysteries of Autism to Decode Animal Behaviour. 2007.
Kurdek http://spr.sagepub.com/cgi/content/abstract/25/2/247 accessed 01 March 2010
Psycoeducation.org. http://www.psycheducation.org/emotion/brain%20pix.htm accessed 03 March 2010
Topal. Attachment Behavior in Dogs (Canis familiaris): A New Application of Ainsworth's (1969) Strange Situation Test
Page views: .3098 .