10/7/14

Novelty (or Practical Habituation)

I have been thinking a lot of late about novelty in dog training. More technically, I’ve been thinking about habituation (i.e. a type of non-associative learning) and how it works in the ‘real world’ for changing dog behaviour in simple ways.

 

Dogs can habituate to water.

Dogs can habituate to water.

When I was a kid, I grew up with a chow chow called Ted. Ted mostly lived in the backyard, but as a child, I one day decided that Ted was going to get a walk every day. And so I walked him every day for about a month (before moving onto the next project, as kids do). Ted started the month with enthusiastic jumping regarding the prospect of a walk. He also vocalised a little bit. By the end of the month, Ted had the lead put on with no fuss, no jumping, no noise, and soldiered on for the walk.

Sure, I could’ve implemented some kind of training regime. But, in reality, I didn’t. Ted started the month thinking walks were novel, and his behaviour stemmed from this novelty. At the end of the month, he was habituated to the walk. Previously, the outside world meant a lot to him and resulted in him getting aroused. By the end of the month, it meant close to nothing, and his arousal levels were far less.

 

Then there’s our foster dog Bandit. I picked him up from his surrendering family, one hour from my house, and drove him home. He drooled, paced, and stressed the whole way home. On ever subsequent car trip, Bandit’s behaviour got more mild. Recently, I drove him to a boarding facility about 20 minutes away, and he was laying, asleep, by the time we got there. No training went into this. Bandit just ‘got over it’ because he habituated to the car – it became less novel.

 

I find many outside dogs are often ‘over the top’ when they meet people, and I think this is a novelty thing, too. If dogs only see people on an occasional basis (i.e. when you go outside), of course they’re going to be excited to see you! If they were inside and saw you constantly, their responses are going to be more mild. Indeed, with most attention seeking behaviours (e.g. jumping up, head nuzzling, vocalising), these behaviours will decrease if the dog has sufficient attention to start with. If attention is given liberally, the resource becomes less important, and the dog’s behaviour changes.

 

I think the concept of novelty is often overlooked in dog training. Sometimes, dog behaviour will ‘get better’ simply because the novelty of something wears off.

Doing many varied things often can do more than maintaining socialisation – it can reduce novelty and so also decrease undesirable behaviour associated with that novelty.

09/3/14

A Link Between Desexing and Reactivity

ResearchBlogging.orgWhile desexing bitches is a common surgery, I was pleased to see Kim et al. take note that “the side effects of the operation, particularly any changes in behaviour, have been quantified in only few studies”.

That is to say, despite us commissioning vets to take the ovaries and uterus out of a great many bitches, we don’t really have much research about it. It’s a pretty scary state of play.

This particular study took 14 healthy German Shepherd bitches, between 5 and 10 months old. Half of these dogs were spayed, and the other half left entire. (The bitches were assigned to each group randomly, except for litter sisters, which were assigned opposite groups.)

After the spay, and having been given 4-5 months to recover from the surgery, the bitches were filmed in their kennel as a stranger and a dog approached. This footage was then scored based on how reactive the bitch was. A score of ’3′ indicated severe reactivity, and a score of ’0′ indicated no reactivity. The scorer was unaware of whether the bitch was spayed or not.

Photo courtesy of Rachel Willis.

Photo courtesy of Rachel Willis.

The reactivity of each bitch was recorded several times, and the reactivity of each bitch declined over the study. It’s likely that the bitches habituated to the novel stimuli. However, despite this affect, bitches in the ovary-hysterectomy group scored higher throughout the study.

Before generalising these results, there are some matters to consider:

  • Reactivity was only measured in a kennel setting, and how these dogs react in the ‘real world’ may be different. We can’t suppose that our pet dogs are going to respond in the same way as kennel dogs.
     
  • The reactivity of these dogs was not measured at the start of the study. While it is unlikely, perhaps more reactive dogs happened to fall into the treatment group by chance. Without a ‘before desexing’ score, we cannot be sure of this.
     
  • The authors make note that these results are for German Shepherd bitches aged 5-10 months old. We can’t assume dogs of all breeds and ages would respond this way to spays.
     
  • Further, the bitches in this study are working lined German Shepherds, which may be more reactive than the typical pet dog.

 

However, this study notes that other studies on bitch spays have shown that as a group spayed bitches are:

  • More likely to gain weight
  • More aggressive than prior to spay (if they were aggressive prior to spay)
  • More active
  • More likely to have urinary incontinence
  • More likely to be reactive after surgery

 

The authors recommend:

[V]eterinary practitioners should inform owners that a bitch may become more reactive after spaying either because they have lost the calming effects of progesterone or because elevated gonadotropins stimulate release of adrenal androgens.

 
Source:
Kim HH, Yeon SC, Houpt KA, Lee HC, Chang HH, & Lee HJ (2006). Effects of ovariohysterectomy on reactivity in German Shepherd dogs. Veterinary journal (London, England : 1997), 172 (1), 154-9 PMID: 16772140

Further reading:
Desexing: It’s bad for Vizslas too
Is desexing a cult?
Desexed dogs – 2.6 times less likely to bite!
Why would you NOT desex your dog??
Golden Retrievers: Cancer if you do, cancer if you don’t

07/1/14

Desexed dogs – 2.6 times less likely to bite!

Australian Veterinary Association makes this claim, in its PDF “Dangerous Dogs – a sensible solution“:

Entire (undesexed) dogs are 2.6 times more likely to bite than those that are spayed or neutered (desexed)

Are undesexed dogs really that risky?

I decided to read the three studies referenced individually.

 

avadesexing

 

 

Messam, LL, Kass, PH, Chomel, BB, Hart, LA 2008, ‘The human-canine environment: a risk factor for non-play bites?’, Veterinary Journal, 177(2); 205-15.

This study used data from 2003 (11 years ago) collected in Kingston, Jamaica and San Fransico, USA. Participants were recruited from vet clinic waiting rooms where they were presented with a questionnaire, set to determine the nature of their dog’s biting behaviour (and differentiating it from play biting).

When it came to comparing entire and gonadectomised dogs, this research suggests:

  • Intact dogs are more likely to bite than desexed dogs
  • Intact males are 1.68 times more likely to bite than desexed males
  • Intact males were 0.8 times more likely to bite than intact females
  • Spayed females were the ‘least bitey’

 

Guy, NC, Luescher, UA, Dohoo, Se, Spangler, E, Miller, JB, Dohoo, IR and Bate, LA 2001, ‘Demographic and aggressive characteristics of dogs in general veterinary caseload’, Applied Animal Behaviour Science, vol 74, iss 1. 

This research is based on data collected in Cannada in 1996 (18 years ago), targeting owners with a questionnaire waiting for vet appointments in three Canadian provinces. Their results indicate:

  • The lowest level of aggression (biting and growling) was reported in intact female dogs
  • Intact male dogs were twice as likely to have bitten as intact female dogs
  • Intact male dogs and neutered females incidents of biting was reported at a similar level

And to quote:

“Relative to intact female dogs, neutered male dogs of at least 1 year of age were at the highest risk for having previously shown biting behaviour, followed by neutered female dogs, and intact males… [O]ur results indicate that the behavioural outcomes of [neutering] are worthy of further investigation.”

 

Gershman, KA, Sacks, JJ & Wright, JC 1994, ‘Which dogs bite? A case-control study of risk factors’, Pediatrics, 93 (6 pt 1), 913-7. 

This study used data from 1991 (23 years ago) using 178 dog bites requiring medical treatment of a non-household-member in Denver, USA. Data was only used for dogs that had not bitten before. The study itself recognises this is a small sample size.

Their data concludes:

  • not-neutered dogs were 2.6 times more likely to bite
  • chained dogs were 2.8 times more likely to bite
  • dogs living with a baby were 3.5 times more likely to bite
  • male dogs were 6.2 times more likely to bite

 

So, are entire dogs 2.6 times more likely to bite?

If you are looking at the study in Denver, USA, in 1991 (23 years ago!) then, yes, their conclusions indicate that intact dogs are 2.6 times more likely to bite than desexed dogs.

But the other evidence referenced by the AVA does not make the exact same conclusions. The study conducted in 2003, using data from Jamaica and the USA, found that intact males were 1.68 times more likely to bite than castrated males, and 0.8 times as likely to bite as intact females. (This study is also old, with the data being collected 11 years ago.)

And that other article, with data from Canada in 1996 (18 years ago) makes pretty much the opposite conclusion. They found that neutered male dogs were the riskiest in terms of bites.

 

Questions to ask…

Why are we relying on data over a decade (or two decades) old? If aggression in entire dogs was a common phenomenon, surely we would have countless studies showing this problem.

Where is the Australian data-set?

Where is the study that controls for factors such as selectionsocialisation, and socio-economic factors?

 

What the AVA should really be saying is:

According to one study conducted in Denver, USA, 23 years ago, entire dogs were found to be 2.6 times more likely to deliver a bite (that required medical treatment) to a stranger than desexed dogs.

05/27/14

Preparing for Earthdog Season

It’s almost the time of the year for earthdog once again! Earthdog is a ‘winter sport’, spanning from approximately May to August, weather permitting.

Earthdog is a fun sport for dogs of ‘earthdog type’ (small terriers, dachshunds, and mixed breeds of), where these little dogs get to test out their natural instincts. The sport involves small tunnels being set up, so these dogs can have the chance to ‘go to ground’ and encounter ‘quarry’ at the end.  In South Australia, our quarry is typically a deceased rabbit, but judges have the discretion to use other quarry (like fluffy toys).  For many terrier and dachshund type dogs, this is the opportunity they’ve been waiting for, to put all their fire and enthusiasm into what they were bred to do!

 

Earthdog den setup at SACA Park, Kilburn, Adelaide, South Australia.

Earthdog den setup at SACA Park, Kilburn, Adelaide, South Australia.

 

There are two key parts of teaching earthdog: Teaching a dog to travel through the small tunnels, and teaching the dog to ‘work’ the quarry.

 

Tunnel Training

In earthdog tests, we use wooden liners to create tunnels. For those wanting to train at home, you can easily make tunnels out of cardboard. Tunnels should be 9 inches by 9 inches square. For the instinct test, the ‘easiest’ level for earthdog, they are expected to go through 3 metres of tunnel with one right angled turn. Lots of dogs have difficulties with corners, so do make sure you design your home made tunnels with a corner! You can make many segments to create 3 metres or more worth of tunnel.

The simplest way to get your dog to go through an earthdog tunnel is to lay a food trail and get your dog to gobble it up piece by piece, going through the tunnel as they do so.  Overtime you can make the food trail more sparse, and use a word to get your dog to go through the tunnel – a command like “Tunnel!” or “Get the bunnies!” are most suitable.

 

Quarry Training

In an earthdog test, once your dog gets to the end of the tunnel, your dog is expected to ‘work’ the quarry’. Working is normally digging or scratching, barking, or biting at the quarry. I would suggest you either invest in your own deceased bunny (most butchers can order one in for you – make sure they keep the fur on) or, if you’re faint hearted, invest in a particularly fluffy and life like toy. See if you can get your dog excited and tugging on the bunny or toy. Having a dog very committed to getting their quarry is an excellent start.

The next step is to get the dog to work even when they cannot get the quarry. Set your dog up behind a wooden or cardboard grill (again, you can make this out of cardboard boxes).  As you wave your quarry in front of the grill, your dog needs to work (bark, dig, scratch, or bite) at the grill from the opposite side.  Reward your dog by giving him his quarry for small inclinations to work at the start, but over time, build up what you expect from him.  At the most advanced level of earthdog, the dog needs to work for 60 seconds – so if you can get 60 seconds of continual work from him, you’ve got what you need!

 

Put it all together

If you have a dog that is going through tunnels, and is working quarry through a barrier, you then need to put the pieces together by putting that barrier at the end of the tunnel.

 

If you need help, the Earthdog Advisory Committee is running a training day on the 14th June 2014 at SACA Park (Kilburn, South Australia).  It is $2 per dog for you to come try, and get some training and help from members of our committee.  Only dogs of ‘earthdog type’ (small terriers, dachshunds, and crossbreeds of) are eligible to train.  RSVP is not necessary, but if you require more details contact Tegan Whalan on 0421 506 482 or email teganwhalan@gmail.com

03/2/14

The Power of Habit

When most dog training focusses on ‘rewarding the good’ and ‘punishing the bad’, the importance of habits (and habitual behaviour) is often overlooked.

Sure, a lot of dog behaviour is based on consequences. This is well understood.

But sometimes dogs do things because they ‘always have’. The regularity of performing this behaviour in itself drives further incidence of the behaviour.

It’s not necessarily that a dog learnt that it was appetitive to partake in a particular behaviour, but that it learnt that it could do that behaviour, and did that enough times that it became a habit.

HairlessHounds Photography

If you don’t want your dog to stand on the sofa, don’t let it become a habit! Photo courtesy of HairlessHounds Photography.

Sure, sometimes this habit behaviours start because of the consequences. For example, a lot of reactivity behaviour.  Initially, the dog was concerned about other dogs so barked at dogs when they got too close. This behaviour was reinforced, as the scary dog normally went away when that happened. However, while that may have been the dog’s thought process two years ago, after the dog has practiced barking at other dogs for a two year stretch, what was initially goal orientated behaviour became habitual behaviour – “I bark at other dogs because I saw them”.

 

Often, dogs perform behaviours because ‘they always have’, and there is nothing intrinsically appetitive about the behaviour or its consequences.

 

Take for example my girl Myrtle. At one stage, whenever I let her out the door, she would run barking to the fence – and she did this enough that it became a habit. Myrtle’s thought process wasn’t based on reinforcement or punishment. If you asked her why she ran to the fence, her answer would probably be about the antecedent (“I ran to the fence because the door opened.”).  It’s a very simple behaviour chain.

Habits can be fixed by concentrating on the antecedent.

While on one hand habits are hard to break, many are also easy to solve by simple management like solutions, which concentrate on the antecedent. In Myrtle’s case, if I put her into that yard through the gate instead of the door, she did not bark at all. After a week or two of putting her into the yard through the gate, she simply ceased to perform her run-and-bark-to-the-fence behaviour. The antecedent was removed for long enough that Myrtle got ‘out of practice’ when it came to this habit. She now can enter the yard through the door with no problems.

 

Basically, if your dog always barks when he sees the postman, or always jumps up when you come home from work, or always scratches the upholstery in the car, your best way to fix this is to just not let the antecedent happen. Remove the postman from view, don’t allow your dog access to you when you come home from work, put him in a crate so he doesn’t scratch the upholstery of a car.

Every time a dog performs a behaviour, he gets in practice and it could become a habit. You need to minimise opportunities for dogs to practice any type of behaviour you do not want to occur.

 

Instill Good Habits

Want your dog to rest quietly in their bed of an evening? Tether them near or crate them on ‘their spot’ for them to settle there. When you remove the physical restraints, the dog will have learnt to sleep in that spot simply because they haven’t had opportunities to sleep in other places.

Don’t want your dog to barge through the front door when you open it? Scatter treats as you open the door – your dog will never practice barging through the door. They’ve learnt to be slow and stay inside (where the treats rain from the sky), and so never get in the habit of barging through the door in the first place.

There are many more examples. Let your dog practice doing all the things you want. Set them up for success in practicing good habits. When you stuff up, the dog learns alternative and possibly less desirable behaviours – and it’s easy for these to become a bad habit.

 

Habits and toilet training

Through rescue, I have had a lot of dogs come in that are used to living as ‘outside dogs’, and have never been toilet trained. Almost all these dogs have been easily toilet trained with no or minimal accidents inside. Why is that? I would argue that they are in the habit of toileting outside – they may not even know it’s possible to empty their bladder or bowels inside because they’ve never had the opportunity. A good habit has been formed.

The first two weeks that you bring a new dog into your home is the best time to instil good habits, especially surrounding toilet training.  Many good habits can be formed by simply not allowing your dog or puppy to engage in bad toileting practices.

Ian Dunbar’s long term confinement area works on this principle. It prevents puppies from getting into the habit of toileting on carpet, tiles or floorboards by minimising opportunities for them to ‘get it wrong’.

However, there’s another upside to this – not only did you prevent the pup from learning the ‘bad habit’ of toileting on inside surfaces, they also learnt the ‘good habit’ of toileting on turf. It’s a win/win situation.

 

The moral of the story? Don’t let bad behaviours become habits!  Everytime a problem behaviour is practiced, it becomes part of the dogs’ behavioural repertoire.  Use management to break patterns of behaviour so they don’t become a bad habit.

Equally as important is to make sure your dog gets into good habits! Maximise the opportunities for your dog to practice behaviours you want.

 

Further reading:

When Management Succeeds

Ouch! Lead work