Leaders of
the pack
In the 80s we had Barbara Woodhouse
and Train Dogs the Woodhouse Way. In
this show Barbara would check, choke
and shout at dogs for disobeying and
‘making fun’ of their owners. Barbara’s
popular dog training show was replaced
in later years with Caesar Milan arguably, a more charismatic pack
leader - with the whitest teeth you have
ever seen. Caesar also checks, chokes
and shouts at dogs, this time for being
‘dominant’ and trying to challenge their
human pack leader. Given the similarity
of these two TV training methods, one
might think that not much has changed
in the last 30 years when it comes to
dog training. But it has, and a training
revolution is on the cards...
Now for the science bit...
To explain how far we have come and
the science behind the ‘force-free’
animal training movement, we have
to go right back to the 1890s and an
American psychologist called Edward
Lee Thorndike (‘Teddy’ to his friends).
Teddy was fascinated by animals and he
wanted to find out whether animals could
learn by observation. He set up ‘puzzle
boxes’ which had weighted buttons or
levers that opened doors. He placed
animals inside (mostly cats) and once
the doors were opened, allowed the
animals to run free. He also rewarded
them with food. He let some animals
watch other animals set themselves
free but found this made no significant
difference in how long it took to achieve
this. What he did learn however, was that
with every trial, the animals got quicker
at using the buttons or levers and setting
themselves free. This paved the way for
84
training as Teddy’s cats in boxes taught
us The Law of Effect. The Law of Effect
predicts that if an association is followed
by a “satisfying state of affairs” then it is
strengthened (so if we reward a specific
behaviour the animals does it more,
faster). If an association is followed by an
“annoying state of affairs” it is weakened
(if we punish an animal for a specific
behaviour they don’t do it as much).
at different schedules of reinforcement
(how often we reinforced specific
behaviours) and how that affected
learning and showed us that stopping
reinforcement or punishment will also
stop previously learnt behaviour (if a
dog learns that when he sits he gets a
biscuit - if you remove the biscuit, after
a while a dog will stop sitting). He called
this extinction.
And so came Ivan Pavlov who went
on to teach us that you could actually
cue behaviour with objects and sounds
as long as they predicted rewarding or
punishing outcomes: i.e. that if a bell
always meant food was given to a group
of dogs, that the bell (even without the
food) caused the dogs to salivate. The
bell was a secondary reinforcer. Of
course the dogs salivated for the food but they learnt to salivate for something
previously meaningless - a bell!
This was really useful but we only really
used the punishment part within the
‘dominance’ dog training model. That
was because trainers and behaviorists
felt that nearly all the behaviour dogs
showed was based upon the undesirable
ambition to become higher than their
master in the pack. We therefore wanted
to reduce this undesirable behaviour
and so partook in punishments as we
understand that punishment decreased
behaviour.
‘And there is every dog trainer’s best
friend, B.F. Skinner. Skinner furthered our
understanding by teaching us the effect
that different types of reinforcement
had on behaviour. He made a quadrant
of four ways to increase or decrease
specific behaviour: Positive Punishments
(something horrible is added like an
electric shock); Positive Reinforcement
(something brilliant is added like a mars
bar); Negative Reinforcement (something
horrid being taken way like someone
holding you under water then letting you
up for air); and Negative Punishment
(something brilliant being taken away
like the removal of your favorite top
out of your wardrobe). He found all
four of these reinforced affected future
behaviour outcomes. He also looked
However, meanwhile in the late 80s
there were some people doing things a
little differently. A new training method
was devised that worked solely on
marking any correct behaviour with a
predictor of food (like Pavlov’s bell), and
then giving the animal a reward. And so
it was that clicker training was born.
Clicker training was a method of training
popularised by Karen Pryor in the
late 80s in her book Don’t Shoot the
Dog. Initially used to train marine life
to do tricks in parks like Sea World,
the method uses a little metal box to
mark