Tag Archives: training

03/11/13

Teaching Distance Drop (with the food placement method)

It’s a little-known fact that dogs will naturally migrate to where food or rewards occur.  Many trainers don’t take advantage of this tendency, which is unusual, especially considering it’s so simple and easy to achieve quick results with appropriate food placement.

Taking advantage of food placement is particularly relevant when teaching distance behaviours. If you want your dog to perform behaviours at a distance, then your rewards should also take place at a distance.

While I plan to blog in more detail about food placement in training at a later date, I have created a video which illustrates the process in teaching a distance drop with the food placement method alone, using my girl Myrtle.

This is a brief summary of the method: Continue reading →

11/23/12

General Training Bits (Dunbar)

This post is part of the series in response to Dunbar’s 2012 Australian seminars. See index.

There are a lot of little training bits that Dunbar mentioned that don’t really go in any other post… Here they are for you:

  • In the ‘big wide world’, dog owners should always carry treats for classical conditioning. Always. For the life of the dog. You can perhaps relax a bit after the dog is 3 years old, but it certainly should go beyond 3-4 months of ‘puppyhood’.
  • We should aim to remove food lures (but this is not the same as classical conditioning, where we need food all the time).
  • A dog that will play fetch or tug are more reliable off lead, as they often seek people to engage in play.
  • Dunbar stated that he did not support using behaviour-ancedecent-consequence (that is, BAC) in training as he finds it ineffective.  That is, as a dog sits, you say “sit”, and hope they associate the word with the behaviour.  (In my personal experience, I have captured behaviours using this method and put them on cue in very few repetitions… I find it quite effective.)
  • Sit and lie down are great solutions to almost any problem – as long as you’re still there.
  • Fixes for humping:
    • Tell the dog to do something else (e.g. sit, drop, fetch).
    • Cue the dog to “off” and withdraw from the environment if the dog continues to hump.
    • Along with ‘put your problem on cue’, put humping on cue! (Jean Donaldson did just that.)
    • If your dog really likes humping, give your dog a ‘humpy cushion’ to reward them for good behaviour.
  • For teaching a dog to ‘take stuff’, then associate the cue with them taking good treats from your hand. The dog will form a habit, and will automatically take less-good-stuff when presented.
  • Tugging can be vamped as a secondary reinforcer.
  • Punishment is insufficient.  Punishes inhibits behaviour only.  Training is not just stopping undesirable behaviours, but also quickly getting back on track into more appropriate behaviours.
  • “Sorry behaviour” exists. Dunbar used the example of young horses who, when kicked out from a herd for a short time, return in an apologetic way.  This can be a way to strengthen relationships.
  • Dunbar described ‘back chaining’ as “moving to a position of strength”, and used (people) learning poems as an example. That is, if a person was to learn the last bit of a poem first, and work their way back, they’re likely to learn it quicker.
  • Dunbar advocated teaching dogs that a gruff or loud tone of voice means “better treats”, which helps to protect against dogs running away or acting unfavourably in situations when their owner’s voice becomes tense.
  • Suggested putting running fast on cue, and using it in recalls and dog sports (like flyball).
  • As a safety behaviour, teaching dogs “not my daddy” when people go to open their crate: anti-theft training.
  • Training is:
    • Changing the frequency of behaviours, and
    • Putting behaviours or absence of behaviours on cue.

     

  • Dunbar advocated using a stern voice (as a punisher) to teach dog boundaries – particularly, not to go out the front door and not to go out the front gate.
  • Dunbar suggested that we are ‘hung up’ on etiology (work out why the dog is doing what it’s doing) instead of actually working to fix the program.  In Dunbar’s words, “Just get on with it and train the dog!”
  • Mental exercise tires a dog quicker than physical exercise.  Nosework is the ultimate in mental exercise.
  • For dogs that greet people in a problematic way, they should be taught ‘shush’ and ‘sit’, and people coming into the house should be schooled to cue these behaviours.
    • Furthermore, these dogs can be taught that people arriving is a cue for quiet.
    • In Dunbar’s opinion, ignoring and back turning doesn’t stop jumping up – but saying “sit” often does.
  • Dunbar’s summary (para-phrased) of dog training ‘these days’:  There is a lot more food, and a lot more classical conditioning.  Dogs are getting friendly and safer around people.  But, dog-dog aggression is increasing because of lack of off leash training.  Lots of people who begin clicker and luring training keep these tools forever.
  • Though lure-reward training or clicker training is a good place to start training behaviours, these behaviours need to be phased out.
  • He said, “If we phased out our reward tools, we’d blow punishment out of the water.”
  • As you move away from a dog, their comprehension of the cue decreases.
  • Flooding is only okay when the rewards are justifiable. Puppies can be flooded.  Dogs that are human aggressive should not be flooded.
  • Rewards drive behaviour!
  • He likes the simplistic law of effect from Thorndike’s: Rewards increase frequency of behaviour, punishments decrease frequency.  Really, Dunbar argues, this is all there is to training.

Continue reading →

10/23/12

Dog training doesn’t happen in a laboratory!

This post is part of the series in response to Dunbar’s 2012 Australian seminars. See index.

Along with Dunbar’s criticisms of the four quadrants of operant conditioning, he also criticised learning theory for being “mostly irrelevant” to pet dog training.  ‘These days’, learning theory is common knowledge for most dog trainers, but Dunbar considers it to be mostly irrelevant in the ‘real world’ of dog training.

 

Outside of the laboratory is a whole wide world of training environments and possible rewards. So why are we so caught up on learning theory?

Much of learning theory has been established by computer-use of reinforcements and punishments.  To Dunbar, this means the findings of learning theory, as delivered a lab, is only relevant to lab settings.  In a laboratory, the subjects are normally rats or pigeons, computers control the training, and the animals are contained.  In the real world of dog training, humans are not computers (they are inconsistent), dogs are more complex than rats and pigeons, dogs escape from people (aren’t contained), and dogs bite!

But humans have an advantage: Humans have voice and can moderate their tone to reward and punish.  Computers cannot use verbal rewards or punishments, and so research on verbal feedback is almost entirely neglected.  Dunbar encourages verbal feedback to train recalls, and claims it is easy to do.  He believes that verbals are more expressive than clicks, jerks and shocks.  Verbals can describe how desirable behaviour was and also an appropriate alternative behaviour.

Punishment may be effective in a laboratory, but (to quote his handout) “people are inconsistent and so the dog quickly learns those times when he will not be punished, i.e., when the owner is physically-absent (dog at home alone), physically-present but functionally absent (dog off leash), or physically-present but mentally absent (owner day-dreaming or making a phone call).”  On top of this, owners normally have bad timing, and dogs learn to be separated from their owners to avoid punishments.  (See also: Dunbar’s thoughts on punishment.)  Dunbar described people as “screwed before we start” if we seek to replicate laboratory settings in real-world dog training. Continue reading →

08/29/12

Put your Problem on Cue (Dunbar)

This post is part of the series in response to Dunbar’s 2012 Australian seminars. See index.

Dunbar advocates putting the 8 big behavioural problems on cue, and then training the opposite.  The idea is that you can cue the ‘opposite’ (non-problematic) behaviour when the dog is displaying the problem behaviour.  The problem behaviour should be taught first, as he thinks dogs are more likely to display ‘the most recently taught’ behaviour.  These 8 behaviour problems, and their opposites, are:

 

Large white and brindle wire haired cross breed sleeping

“Settle down” – a useful behaviour to cue dogs to perform when they’re jazzed up or over enthused.

1. Jazz Up / Settle Down

Often dogs can be over excited, over stimulated, or generally ‘worked up’ and this can be problematic for owners.  For this reason it is useful to have a ‘settle down’ cue, but Dunbar of course suggests that you teach the opposite, too – a ‘jazz up’ cue.  You could turn this into a class game where the winner is the person who settles down their dog the fastest, or meets a 3 second deadline.  Teaching a dog to ‘jazz up’ is also easy, and often inspires and motivates class members to train.

‘Settle down’ is useful when trying to prevent problematic behaviours, such as excitement at the front door, or fence-fighting behaviour.  ‘Jazz up’ could also, potentially, be useful reward in the obedience ring.  Diane Baumann, in her traditional training book Beyond Basic Obedience, encourages owners to have an exciting cue (like ‘jazz up’) to mean an exercise is finished.

 

2. Woof / Shush Continue reading →

06/17/12

Reward Training Techniques (Dunbar)

This post is part of the series in response to Dunbar’s 2012 Australian seminars. See index.

Dunbar described five reward training techniques:

 

 

Lure Reward Training
He called these ‘techniques that cause the behaviour’ and the ‘Plan A’ of dog training – that is, it should be the first option when teaching a dog a behaviour.  More about this method is outlined in my lure reward training post.

 

All or none reward training
Dunbar created ‘all or none’ reward training after thinking about dogs in shelter situations.  These dogs need to default to good behaviour, or just be ‘good’ without any verbal cues.  In all or none reward training, you just wait for the animal to do what you want, and reward it.  For example, if you have a dog on leash and wait long enough, they’ll eventually sit.  The term ‘all or none’ comes from the behaviour: He’s either sitting, or he’s not.  Dunbar advocates this way for inattentive or ‘crazy’ dogs, and suggests it should be the ‘Plan B’ in dog training.

 

Black and white working cross breed runs with a tennis ball in mouth.

Life rewards: Running, playing fetch. Much better than any boring treat!

 

Shaping (often with clickers) Continue reading →