Two
Ways a
Dog Learns
Consequences
& Associations
A.
Bindoff
If we
expect
our dogs
to understand
us, surely,
we must
first
attempt
to understand
our dogs!
Dogs learn
by the
consequences
of their
actions.
"If I
drop that
ball at
his feet,
he will
often
throw
it for
me" or
"Last
time I
stole
a sausage
from the
barbecue,
I burnt
my tongue.
Better
not do
that again."
Dogs also
have emotions.
"Those
fireworks
scare
me!" or
"Someone
is at
the door,
YIPEEE!!!"
Both of
those
concepts,
learning
by consequences
and the
experience
of emotion,
are something
we can
relate
to. We
might
learn
as children
that eating
all our
vegetables
gets us
dessert,
leaving
them on
the plate
means
missing
out! We
might
associate
a certain
song with
a special
time in
our lives,
bringing
back good
feelings.
Learning
by consequences
Learning
by the
consequences
of our
actions
is called
"Operant
Conditioning".
When something
has a
rewarding
consequence,
we learn
to repeat
it. When
something
doesn't
have any
rewarding
consequences,
we tend
not to
repeat
it. When
something
has a
punishing
consequence,
we learn
to avoid
the situation,
possibly
altogether.
Learning
by association
This is
called
"Classical
Conditioning"
or "Pavlovian
Conditioning"
after
Ivan Pavlov
who discovered
that when
a stimulus
(such
as a ringing
bell)
is paired
with an
event
(such
as food
being
given),
an association
is made.
For a
dog, the
ringing
bell would
have a
pleasant
association,
the dog
has learned
that food
will follow.
Here we
are getting
into the
realms
of emotion!
Things
that you
associate
with good
things
will also
tend to
give you
good feelings.
Things
that you
associate
with bad
things
will tend
to give
you bad
feelings.
Operant
and Classical
conditioning
describe
different
mechanisms
for learning,
but the
two go
hand in
hand.
How
Strong
Emotions
Affect
Learning
It has
been observed
that when
an animal
- dog,
human
or otherwise,
is affected
by strong
emotions,
rational
learning
can be
affected.
Imagine
the ridiculous
and terrifying
scenario
of trying
to learn
how to
program
your VCR
while
a crazed
murderer
is smashing
your door
down with
an axe.
You just
want to
get out
of there,
you don't
care if
you miss
the final
episode
of 'Inspector
Rex'!
(Ok, you
still
care...)
Imagine
teaching
a class
full of
excited
kids their
'multiplication
tables'
while
they are
running
around
playing
with each
other,
laughing
and joking.
Hopeless,
no good
teacher
would
try to
do that
without
settling
them down
first,
even if
it meant
doing
something
different
first.
It is
the same
with dogs.
There
is no
point
trying
to get
rational
learning
from a
frightened,
anxious,
or just
plain
over-the-top
excitable
dog without
addressing
the underlying
issues
first.
Emotional
issues
must be
addressed
before
consequences
even matter.
There
is a very
good reason
for this,
and it
has to
do with
the way
the brain
works.
To ensure
survival,
both human
and dog
brains
have a
simple
rule -
"do what
you have
to do
to survive".
When a
dog perceives
a threat,
the part
of the
brain
responsible
for survival
takes
precedence
over the
part of
the brain
responsible
for rational
learning.
As an
example,
imagine
you have
a dog
who is
aggressing
towards
another
dog. Using
the consequence
of punishment
(collar
correction,
verbal
etc) will
have very
little
effect
in actually
punishing
(reducing)
the behaviour.
It may,
however,
contribute
to the
cause
of that
dog's
aggression
by reinforcing
his unpleasant
association
with other
dogs!
Let's
use the
same example
but this
time we
'reward'
the dog
with food
when he
shows
aggression
towards
another
dog. The
act of
aggression
is unlikely
to be
reinforced
in this
case,
because
that part
of the
brain
is taking
a back-seat.
The 'learning
by association'
part of
the brain,
however,
is getting
a dose
of something
pleasant
in the
presence
of the
other
dog. It
is starting
to make
a 'good'
association
with the
other
dog, instead
of a 'fearful'
association.
It is
hard to
guess
how much
"Classical
Conditioning"
and how
much "Operant
Conditioning"
is taking
place
until
after
the event.
It all
seems
rather
paradoxical
at first,
but hopefully
I have
explained
it well
enough
that it
makes
sense!
Of course,
in the
above
example
it would
be far
better
to give
the dog
food before
he aggresses
and try
and stay
far enough
away to
avoid
a rehearsal
of aggression
altogether.
Then you
can reward
the dog
for non-aggressive
behaviour
too. You
get the
benefit
of reinforcing
good behaviour,
and making
a positive
association
all at
once -
what a
great
deal!
What
happens
next?
At some
point,
if we
have been
successful,
the dog
will be
in a more
suitable
frame
of mind
for rational
learning.
The dog
who is
nervous
around
other
dogs will
not be
so nervous
any more,
and will
be able
to complete
a short
'sit-stay'
in the
obedience
class.
Eventually
he will
be able
to complete
a long
'drop-stay'
with handler
out of
sight.
The dog
who 'doesn't
like men'
will be
able to
accept
a brief
and heavily
rewarded
'stand
for exam'
from a
man, and
eventually,
the full
examination
from teeth
to tail!
©
Aidan
Bindoff
2004
May be
reproduced
with permission.
abindoff
@ ozemail.com.au
