Dogs in Need of Space

December 3rd, 2011 § 2 Comments

Excellent blog post here, called Dogs in Need of Space.

the dog-human bond

November 2nd, 2011 § Leave a Comment

Being a good-enough puppy parent

I’ve long felt that manipulating rewards and punishments doesn’t explain everything which is going on between people and their dogs.  But I could never quite put into words why I thought this.  It just seemed that some things operated in another dimension.

Gradually, a combination of learning more about what we know of the human-dog relationship from reading about research; an awareness of the animal-assisted therapy literature (which features the dog-human relationship greatly); and a knowledge of attachment theory, has come together and I’ve started to hash out some thoughts about this ‘missing dimension’ of feelings and connections between us and our furry pals.  Here are some of my first, rough, thoughts…

To start with, dogs and humans have been interacting with each other for up to 100,000 years.

A jaw fragment from a domestic dog, 14,000 years old

The result of this, is that canine and human minds are curiously attuned.  Here are a few recent studies which illustrate this phenomenon:

1)  Dogs’ understanding of human pointing gestures (Lakatos et al 2009):  A researcher kneels down, with an upturned cup on her left and another on her right.  She points to one of these cups.  Opposite her is someone holding a dog.  The dog is released when the researcher points.  The dog, without having received any training, goes to the cup the researcher has pointed at.

Research has found that even little puppies can ‘read’ human pointing gestures in this way (rather than looking at the finger itself, as many other species do).  The fact that puppies can do this, indicates that they are born with this ability – the behaviour is innate.

Dogs perform as well as a 2-year old child on this test, and they outperform chimps (who often flunk pointing gesture tests).  They cannot perform as well as 3-year old children.

It seems logical that this ability has evolved in dogs as a result of our two species working so closely together for up to 100,000 years:  Herding sheep, hunting and guarding together, it was advantageous to both of our species, for dogs to be able to understand human gestures.

2)  Dogs’ ability to perform social referencing (Merola et al 2011):  When a strange object suddenly emerges from a cardboard box, dogs are likely to show fear if the experimenter shows a fearful face, and likely to be accepting of the object if the experimenter shows a happy face.

That is:  They ‘reference’ the response of their human companion, in order to help them determine whether something novel is a threat or not.  Dogs are, here, ‘reading’ the social responses of another species, extracting meaning from these social responses.

Such social referencing experiments are much usually directed at human toddlers – this is a standard child-development test.

See from 3.45-8.10 mins, on this video (may not be viewable in some countries, unfortunately):

3)  Humans’ ability to understand dogs’ barks (Molnár et al 2010):  When played recorded barks, humans can reliably identify the contexts in which those barks would have been made – the emotional ‘meaning’ behind them.  Humans are able to identify five barks with different emotional ‘meaning’.  It seems that it is not only dogs which can understand the emotions of humans, but the reverse is true as well.

See from 14.50-21.19 in the same video, above.

4)  The capacity of dogs and humans to regulate each other’s emotions (Odendaal & Meintjes 2003):  In this study, owners and their dogs were left alone together, and owners were told to pat, stroke, cuddle and talk to their dogs – as they usually would during a ‘petting session’.  After 20 minutes, the stress hormone (cortisol) levels in both dogs and humans were greatly reduced and the feel-good hormone levels (of oxytocin, dopamine and endorphins) were increased.

As similar results were seen in both humans and dogs, this is a mutually beneficial exchange, wherein each helps the other to reduce stress and anxiety.  These results suggest that dogs and humans can ‘regulate’ each other’s emotions when they feel bonded or emotionally attached to each other.
We have long been aware of such emotional regulation taking place within the human species:  When a human baby cries, the baby’s caregiver comforts him or her, so that the stress and fear (resulting in elevated cortisol levels) which the baby had been experiencing, are reduced.  But examples of cross-species emotional regulation are extremely rare.

The oxytocin levels in the dogs were actually five times greater after the petting session with people than before.  Oxytocin, as a hormone, is associated with attachment to others; with what we humans experience as love.  Oxytocin levels rise greatly during experiences of bonding and affiliation, and the quintupling of oxytocin in this experiment suggests that humans and dogs can experience each other as attachment figures.

There are a great many studies from the field of animal-assisted therapy, which echo the findings here – supporting the idea that humans and dogs can form mutually beneficial relationships, experiencing each other as attachment objects.

5)  Dogs prefer a human attachment figure over another dog (Tuber et al 1996):  In this study, eight 7-9 year old mutts lived as littermate pairs in kennels all their lives.  They had been fully socialised to people and were looked after by one human carer.  The pairs had not been apart even for a minute in the previous two years and hardly ever during their life time.

Despite this, when one of each pair was taken away for four hours, the remaining dogs’ behaviour did not alter.  The stress hormones (cortisol) in their blood did not change.  That is, they were not stressed by the absence of their sibling.

When they were taken to an unfamiliar kennel, they were upset and their cortisol levels went up over 50%.  Yet this proved true, whether they were with their littermate or not:  The attachment between the siblings provided no support and no emotional regulation to reduce their stress.
However – crucially – if the human carer sat with the dog in the new kennel, the dog would stay near the carer and seek comfort.  Remarkably, the proximity of the human carer alleviated the dog’s stress completely:  If the carer were there, their cortisol levels remained almost normal.  That is:  The presence of a human caregiver was better able to reduce stress in these dogs than another dog could – even if that dog were a sibling, and even if the siblings had lived together all their lives and saw their human carer much less frequently.

This research suggests that, outside of puppyhood, dogs are unlikely to experience another dog as an attachment figure.  Yet they continue to experience humans as attachment figures.  Dogs remain ever-ready to attach to humans, throughout their lives:  Research shows that just a few minutes of friendly attention from one person on two consecutive days, is enough to make adult dogs in rescue kennels desperate to remain with that person; they howl, jump at the door and scratch when that person leaves on the second day.

6)  Human and dog tuggy games are not competitive, but are played by the dog for the sheer pleasure of social exchange with a human (Rooney et al 2000):  When dogs play tug with other dogs, they generally want to compete with each other for possession of the toy.  If one dog lets go of the toy, the other dog is likely to guard that toy from him or her.  If a second tug toy is provided to the two dogs, they will often take a tuggy each and play by themselves.

When playing tug with a human, dogs are much more likely to relinquish the toy, apparently to keep the game going.  They do not want possession of the toy, but the pleasure of social exchange with the human.  And they completely ignore the presence of a second tug toy; they continue to bring the one toy to the human and to encourage the human to play tug.

 


Attachment

Much of this research illustrates the way in which canine and human minds are curiously attuned.  When we start talking about minds which are attuned, and when we start touching on the regulation of each other’s feelings, we are moving into the realm of attachment theory.

John Bowlby

Attachment theory was the work of John Bowlby, a psychoanalyst.  Although attachment theory is used these days to talk about human relationships, curiously it all began with a cross-species relationship, since Bowlby was hugely influenced by the experiments of Konrad Lorenz with geese (1935).

Lorenz found that, if baby goslings first saw Lorenz on hatching, instead of the mother goose, the goslings would adopt Lorenz as the parent figure and follow him everywhere instead:  They became ‘attached’ to Lorenz:

Although (luckily) human children don’t adopt the first moving object they see as ‘mum’, Bowlby believed that children come into the world biologically pre-programmed to form similarly strong emotional attachments with others – again, as a survival mechanism.  We might think of it as a baby’s need to be near a caregiver or parent, in order to feel safe.  Being separated from this caregiver is frightening for a baby:  If the baby ends up alone, he might die of hunger or exposure – or, in the wild, end up becoming someone else’s lunch.  Defenceless and helpless, he can’t exist in the world if he’s not looked after by a caring adult.

It is the role of the caring adult, or caregiver, to soothe the baby when she becomes distressed – to make the feelings of fear and anxiety go away, thereby reducing levels of the harmful stress hormone, cortisol, in the blood.  The baby and the caregiver both experience increases in oxytocin at these moments of affection and bonding – in fact, attachment might really be understood as the giving and receiving of love.

As the baby grows up, he begins to be increasingly interested in the world, and becomes an increasingly independent toddler.  But he doesn’t stop needing his caregiver’s presence:  Toddlers begin to explore further away from their caregivers, but they still need to return at moments of stress or fear, to ‘touch base’ and reassure themselves that they are safe and that their caregiver is still reach-able.

Here is a an example of a securely attached child, from a classic experiment by Mary Main into attachment styles.  Note that a securely attached child will be distressed, and likely cry, on separation:

I believe the basic tenets of attachment theory remain very similar, for dogs.

Attached by more than the lead between you

John Bradshaw’s excellent book In Defence of Dogs suggests that dogs’ minds categorise humans differently to the manner in which they categorise other dogs.  This is borne out by the finding that dogs do not play tug with humans, in the same manner they play it with other dogs; they prefer to play in a socially cop-operative way with us.  It is also borne out by the finding that adult dogs appear to be biologically pre-programmed to prefer a human attachment figure even to a member of their own species.

Indeed, it might be that dogs are born not only with a biological predisposition to attach to their canine mother as puppies, but also with a biological predisposition to attach to a member of the human species, throughout their adult life.

Humans and dogs have hunted, herded and guarded together for up to 100,000 years.  The propensity of dogs to remain attached to their human caregivers throughout adulthood enabled new heights of co-operation and understanding to be reached between our species.  This was, undoubtedly, of benefit to both in our shared activities.

And they called it, puppy love…

The results of recent research into human/canine attachment don’t appear to have filtered down to dog training itself yet.  Most puppy training books talk about praising and rewarding your puppy when she does the right thing, and ignoring or preventing unwanted behaviour.  In other words, about the giving (and removing) of rewards.

This way of thinking about things is called ‘learning theory’ or ‘operant conditioning’.  Learning theory is very useful and an excellent way to teach your puppy all sorts of behaviours, from sit and down to recall and complicated tricks.

However, to wholly and entirely see your puppy (or any animal) in terms of learning theory is to treat her a bit like a computer or an input-output, operant-conditioning machine.

Skinner box

I believe there are aspects of the dog-human exchange which can’t be reduced to learning theory.  These elements are emotional and relational and they operate in the dimension of attachment behaviour.  Usually, we don’t need to think about these things, because they are happening in the background and they tick along just fine without us knowing about them.

But sometimes the advice which is given from the perspective of learning theory conflicts very strongly with the suggestions we might make from an attachment perspective.  To show you what I mean, here’s an example.  Here is some advice from Jean Donaldson (1996), a fabulous writer on the subject of learning theory, on getting your puppy used to being left alone:

First, if it’s a new dog or puppy in your house, set the precedent right away… right off the bat, leave baby dog or newly adopted dog alone for brief durations, over and over… It is absolutely normal that puppies and dogs will distress vocalise when you leave them alone… wait for a lull before going to them, at whatever cost.  Resist thoughts like ‘what if he needs to go out’ or some other vital communication from dog to human, and worry about the potent behaviour modifying influence of your arrival chez dog.  (49-50)

The thinking, here, in terms of learning theory, is that returning to the puppy will reward what the puppy is doing (making noise).  So the owner should not return until the puppy is silent.  This makes perfect sense, in terms of learning theory.

Yet, from an attachment perspective:  This is an immature, helpless and bewildered baby animal, in an unfamiliar environment.  He has just been taken away from his mother, his main attachment figure to this point, and his siblings.

On his own for probably the first time in his life, he feels abandoned and, so, in fear of dying.  (Any immature animal, truly abandoned, most certainly would die.)  With his noise, the puppy is attempting to get a caregiver to respond and help him with the unbearable fear he is experiencing.

This is terrible, but in his protests there is hope:  There is hope that someone is listening, that someone might yet respond.  When might he fall silent?  When he has given up all such hope.  Psychologically, this is a far worse place.  In humans, we see the effects of truly non-responsive caregivers in the orphanages of developing countries, where children silently rock themselves – knowing that there is no point making noise, because no one will come.  It has been proven that the resulting unregulated high cortisol levels wreak havoc on the developing brain, over time, leading to all kinds of psychological difficulties.  Yet this is exactly what learning theory would lead us to, with puppies.

Currently, separation anxiety is the single biggest problem behaviour in adult dogs, with 15% of all dogs suffering from some form of separation distress.  Separation anxiety is a major reason why dogs are given up to rescue and rehoming centres, because of the destruction and noise they create.  What is ‘separation anxiety’?  It is an acute panic, on the part of the dog, who believes he has been abandoned – left behind forever, to die.  When he sees you leave, he does not believe you are coming back.

A dog with separation anxiety does not feel ‘securely attached’.  He does not trust you, his caregiver, to be reach-able when he needs you – instead he believes that you will be inaccessible, far away from him.  It can easily be seen how the conventional advice from a learning theory perspective – ‘do not return to a noise-making puppy’ – can cause a puppy to associate being left with escalating panic and fear.

Is it any wonder that separation anxiety is such a great problem for today’s dogs, when the predominant approach to puppyhood pays little heed to a puppy’s emotional needs?

Conceiving of the dog-human relationship by way of attachment often results in startlingly different solutions than does a blanket application of learning theory.  We need to pay much more attention to attachment around issues of separation (temporary or permanent) of the dog from the caregiver, throughout its life. This includes:  Leaving the breeder/mother; the first night; night-times in general; alone-training; kennelling; hospitalisation (of dog, at vet’s); rehoming.

Animals do grieve and are aware of the absence of attachment figures, and an awareness of this could have great implications for all kinds of policies:  From vets permitting human visitors to pets who have been hospitalised, through to puppies being kept next to the new owner’s bed during the first night.  And fewer dogs growing up to have separation anxiety, because their owners were educated into an awareness of their needs, instead of a hardcore behaviourist approach.

We also need to pay attention to attachment during the socialisation period, when the puppy is using the owner as a ‘secure base’ – retreating to gain confidence and to ‘refuel’, and then going forth into new territory, to encounter new things.

We can but hope…

References

Donaldson, J.  (1996)  The Culture Clash.  James & Kenneth, London.

Lakatos, G., Soproni, K.,  Dóka, A. & Miklósi, Á.  (2009)  A comparative approach to dogs’ and human infants’ understanding of various forms of pointing gestures.  Animal Cognition, 12(4), 621-31.

Lorenz, K. (1935). Der Kumpan in der Umwelt des Vogels. Der Artgenosse als auslösendes Moment sozialer Verhaltensweisen. Journal für Ornithologie 83, 137–215, 289–413.

Merola, I., Prato-Previde, E., Marshall-Pescini, S.  (2011)  Social referencing in dog-owner dyads?  Animal Cognition.  [Aug 28 2011 epub ahead of print]

Molnár, C., Pongrácz, P., & Miklósi, Á.  (2010)  Seeing with ears:  Sightless humans’ perception of dog bark provides a test for structural rules in vocal communication.  Quarterly Journal of Experimental Psychology, 63, 1004-1013.

Odendaal, J., Meintjes, R.  (2003)  Neurophysiological correlates of affiliative behaviour between humans and dogs.  Veterinary Journal, 165, 296-301.

Rooney, N., Bradshaw, J.  & Robinson, I.  (2000)  A comparison of dog-dog and dog-human play behaviour.  Applied Animal Behaviour Science, 88, 235-48.

Tuber, D, et al.  (1996)  Behavioural and glucocorticoid responses of adult domestic dogs to companionship and social separation.  Journal of Comparative Psychology, 110, 103-8.

Where Am I?

You are currently browsing entries tagged with canine behaviour at g a l o d y.

Follow

Get every new post delivered to your Inbox.