Wednesday, 16 September 2009

An introduction to positive reinforcement training

Recent years have seen a move away from the use of punitive methods in dog training. Where once trainers were set on making dogs bend to their will now more and more are moving to a system where good behaviour is encouraged rather than 'bad' or incorrect behaviour disciplined.

The idea of positive reinforcement training lies at the heart of popular modern training methods such as clicker training. Typically the reward is food based but this need not always be the case. If you have a dog who is very keen on a particular toy, or who particularly loves to be made a fuss of then these can also be used as rewards. Many specialist 'working' dogs such as Police sniffer dogs are trained using this method and rewarded with the ultimate prize of a squeaky ball having correctly sniffed out drugs or explosives.

Key to using positive reinforcement is c
orrect timing. It is essential that the reward must occur immediately—within seconds—of your pet performing the desired behaviour, or your pet may not associate it with the proper action.

When your pet is learning a new behavior it is important to reward them every time they correctly do what you want, which means continuous reinforcement. Initially it may be necessary to use a technique called "shaping" in order to achieve the desired result. Shaping means reinforcing when your dog gets close to the desired result and then gradually requiring that your dog do more in order to earn the treat.

For example, if you're teaching your dog to "shake hands," you may start by rewarding your dog for lifting their paw off the ground, then for lifting it higher, then for touching your hand, then for letting you hold their paw, and finally, for actually "shaking hands" with you.

Once your dog has successfully learnt a behaviour you can move to using intermittent reinforcement . At first, reward your dog with the treat three out of every four times they do the behavior. Then, over time, reward them about half the time, then about a third of the time, and so on, until you're only rewarding her occasionally with the treat. Continue to praise your dog every time—although once your dog has learned the behavior, your praise can be less effusive, such as a quiet, but positive, "Good dog." Use a variable schedule of reinforcement so that they don't catch on that they only have to respond every other time.

No comments:

Post a Comment