A CONSIDERATION OF TRAINING
METHODS
By Deborah Jones, Ph.D. 2001
(Originally Published for Clean
Run)
Over the past ten years or so, there has been
a major shift in dog training techniques and methods. Older, traditional
types of training often relied on the use of physical force or
aversives (unpleasant events) in order to get the dog to perform
or to comply with commands. For many, many years this was the
commonly accepted way to train dogs. However, we now have another
option that is become more and more popular. Newer methods are
primarily based on the use of positive reinforcement as a training
tool. As trainers and owners become more and more educated in
the use of operant and classical conditioning, the use of aversives
and force is losing favor. We are in a time of transition in dog
training, which can lead to confusion and stress (for the trainers).
On the one hand, established and successful trainers are saying
that traditional methods are highly effective and have stood the
test of time. On the other hand, behaviorists and other trainers
are saying that the older methods are not as humane or as effective
as positively based ones. This leaves many very confused. How
can they decide what is right for them and for their dogs? Those
on both sides of this issue make some powerful and persuasive
arguments, so let’s examine some of the claims in more detail.
1. Traditional techniques work.
This statement is true, up to a point. For specific
types of dogs, and for specific types of behaviors and activities,
forceful techniques can be effective. However, just because something
works in certain situations, that doesn’t mean it can or
should be applied to the general population. For example, a highly
driven and intense Border Collie may be able to ‘take’
a very strong correction from his handler in order to get him
under control and keep him from charging the sheep in herding
lessons. The correction may have the desired effect, yet not lessen
the dog’s drive and desire to do the work.
However, this does not mean that you could apply
the same type of correction to a different dog in a different
setting (or a different dog in the same setting, or the same dog
in a different setting), and get the desired effect. For example,
a fairly sensitive Poodle who doesn’t drop on recall during
Open obedience training would probably melt into the floor and
avoid working if given the same correction. You could interpret
this to mean that different dogs need different types and levels
of corrections. In order to make this distinction and to use correction-based
training effectively, you would have to be a trainer who was very
good at reading dogs and applying just the right level of aversive
at the right time. Unfortunately, most trainers are not that good
at what they do. And if you make a mistake using this method,
you can see some very severe and long-lasting consequences. You
might ruin a dog for a particular activity, or for any type of
activity.
Another issue to consider here is what philosophers
call the ‘means-end analysis’. This kind of logic
states that the ‘ends’ justify the ‘means’.
So, if you get the desired result, it doesn’t matter how
you got there. Most rational people can see that this is a very
dangerous road to start traveling down. Justifying the means by
pointing out the successful end result doesn’t take into
account how ethical the mean might be.
2. Positive reinforcement is not effective
for all dogs.
Again, this is a true statement. However, no
method is 100% effective, it simply doesn’t work that way.
We have to take into account individual differences in trainers
and dogs. While many people have been very eager to find a ‘better’
way to train, they have not always been as educated and skilled
in these techniques as they might be, which leads to errors. In
their enthusiasm to use techniques that are reward-based, some
people may take the ‘no corrections necessary’ idea
to mean ‘no control necessary’. So, you might end
up seeing dogs who are described as clicker trained who are behaving
like spoiled brats, while their owners extol the virtues of positive
reinforcement.
Done correctly, clicker training and positive
methods can be highly effective. A lack of control is not an inherent
facet of clicker training. In fact, psychological control can
be obtained in a much faster and more efficient way using positive
methods. Once you have psychological control, direct physical
control is not necessary. A dog who wants to work for you because
it is in his best interest, and who has learned that you control
access to all the good things he wants, is a very attentive and
compliant animal indeed.
3. A balanced approach, using a mixture
of positives and negatives, is the best way to train.
This is an interesting argument, but, in reality,
it is not effective. While it might sound appealing, the Chinese
menu approach to training (one from Column A, two from Column
B), often leaves the dog confused and unsure. Using a mixture
of corrections and positive reinforcement sends mixed signals
to the dog, and typically results in a dog that does not reach
his training potential. The reason for this is that the basis
of the two types of training is completely different. Using negatives
and aversives is based on teaching the dog that he should not
do anything unless given a command or permission to do so. This
leads to dogs that learn that, in many situations, doing nothing
is the safest way to go. On the other hand, dogs that have been
clicker trained, especially those who have been taught using shaping,
have learned that trying new behaviors (in training sessions)
is the way to earn reinforcement. We really want our dogs to have
an openness to trying new things as it gives us a variety of behaviors
to choose from and to mold into the desired final action. These
two underlying foundations (do nothing or do many things) are
clearly incompatible.
What many trainers really mean when they talk
about using a balanced approach is that they use traditional,
force-based techniques, but they also add positive reinforcement
when the dog is correct. This is really not a true mixture of
traditional and positively based training. The real heart of positively
based training is the trust the dog learns in the process. In
training, the dog is safe to try and do many new things. Those
that are desired will be reinforced, and will then be repeated.
Undesired behaviors are not punished, they are ignored, and they
disappear (in general).
4. Positive reinforcement will not work
for all behaviors in all situations.
Again, this is very true. But the same could
be said for traditional training. And it is not a failing of the
method, but a failing of the application. Training that is based
on operant and classical conditioning has to work, if done properly.
Scientists talk about the LAWS of learning. They are laws, not
suggestions. If a trainer understands learning theories, and applies
them correctly, they will work. I have 100% confidence in the
effectiveness of the positively based techniques that I use because,
as a scientist, I know that they will be successful. Thousands
of research studies provide evidence that this is true.
I don’t know how many times I’ve
heard an owner/trainer say “I need to use a correction to
make sure he knows he can’t get away with that”. I
always wonder to myself where they ever got the idea that corrections
are more powerful than positives. There is no evidence to back
up that idea. What makes behavior strong is a strong reinforcement
history.
This is not to say that there are never, ever
times when an aversive may be necessary. If an animal is harming
himself or others, I will do whatever is required to make it stop.
However, I am not under the illusion that I am training in those
situations; that is simply crisis management.
5. In the real world, we learn from
the bad things that can or do happen.
This is correct. We learn to avoid dangerous
and unpleasant things. A negative experience will change our actions.
That is the way the world works. But, just because there are bad
things out there, I don’t think that gives me the right
to use negatives and aversives against other living creatures.
Why add to the unpleasantness that already exists in the world?
I want my training relationships to be based
on mutual respect and on trust. Just because I am bigger, stronger,
and smarter (sometimes!), that doesn’t give me the right
to hurt another creature just because I can. Even if the aversive
is mild and/or short-term, it goes against the way I want to interact
with my animals and the way I want to live in the world. Cooperation
is much more pleasant than confrontation, and the results are
well worth the effort.
IN CONCLUSION…
I would like to end this consideration
of training methods by encouraging trainers to think carefully
before they decide what is appropriate for them and for their
dogs. Many people seem willing to try any new technique that claims
success, without thinking through the possible implications and
consequences. One thing I have learned is that it is very important
to carefully consider your training methods and techniques. In
addition to being effective, will they strengthen the relationship
between you and your dog? Are they consistent with your ethics
and ideals? Are they clear and fair? It is our responsibility,
as trainers and owners, to do what is best for our pets. I encourage
you to think very carefully about how you train.
Back
to Articles Index
© 2005 K9inFocus.com
Built and Maintained by Pet
Web Designs
Home
| About | Books
& Videos | Reviews | Workshops
| Calendar | Seminar
Contact Form | Photos | Articles
