Join Us

Join the Animalinfo's online directories. Silver Listings are free. Find out more >

Get a gold plan Get a silver plan
Join >

Positive Reinforcement Training for Dogs

Search the summaries below for articles on many aspects of reward based positive reinforcement training for dogs. Click on 'Read more' for the full article or search for books in our online book resource.

Positive reinforcement training

How Clicker Trainers Correct Behavior

Category:
Positive Reinforcement Training for Dogs
Author(s):
Cindy Ludwig, Canine Connection LLC

Part I: How Clicker Training Minimizes the Need for Correction

Introduction

One of the most commonly asked questions by people new to clicker training concerns how to correct an animal for mistakes since clicker trainers don't use correction, at least not in the traditional sense of the word. How can an animal learn what it has done wrong without being corrected?

How Clicker Training Works

To answer this question it is necessary to first review the basics of clicker training. In clicker training, a clicker or hand-held noise-making device is used to make a click-click sound at the precise moment an animal does something the trainer likes and wishes to see repeated. The click is followed by a reward or reinforcement, usually a food treat.

This motivates the animal to repeat the behavior again and again. We say that we are building a "reinforcement history" with the animal which makes the behavior stronger or more likely to be repeated in the future. This new behavior is put "on cue," the treats are phased out and the behavior itself becomes reinforcing to the animal. Once the animal understands what is expected the clicker is no longer needed. It is just a training tool.

The 80% Rule

Clicker trainers keep their animal students successful every step of the way by adjusting their criteria for performance up or down, depending on how the animal is doing.  They tend to look at mistakes a little differently than trainers who use more "traditional" methods which include physical or verbal correction or intimidation.

If an animal fails to provide the expected response or provides the wrong response to a cue, a clicker trainer first asks him or herself where the training broke down. Perhaps the animal isn't motivated enough, is too distracted, or did not understand what was expected.

Clicker trainers have an 80% rule. When the animal is performing at an 80% or higher success rate, they move to the next level in the training process. Success rate is easily calculated by dividing the number of correct responses by 5 or 10 opportunities - 4/5 or 8/10 correct responses equals an 80% success rate.

No Click Means Try Again

Just like the game of hot and cold played by children, click means you're headed in the right direction. No click means you're not.  A word to indicate to the animal that he has made an error would not provide any more information than the absence of a click, so clicker trainers do not use what are commonly referred to as "no reward markers."

We humans are verbal creatures, but animals are not. The beauty of the clicker is that it provides a great way to bridge the communication gap. As a neutral, unique and consistent sound, the click-click sound of the clicker clearly communicates to the animal that he has done something right or wrong in terms of what the trainer expected.

A clicker trainer has a picture of the final behavior he or she wants to see, and breaks this behavior down into steps that make it easy for the animal to learn. By shaping a behavior in small steps which keep the animal successful, clicker trainers are able to keep their animal students engaged and motivated. If the success rate falls below 80% a clicker trainer will adjust the training plan and situation to bring the animal's success rate back up.

Cues Versus Commands

In clicker training we use cues rather than commands. This is an important distinction and not just semantics. A cue is an opportunity for to earn rewards of some sort and a command is an order which means, "Do it or else!"

Unlike other types of training that use commands, cues are taught after the animal understands how to perform the behavior. The reason for this is twofold. One reason is that cues spoken during training before an animal understands what is expected only serve as a distraction which makes learning more difficult and secondly, clicker trainers want to ensure that the cue is clearly associated with the final goal behavior and not some intermediate form of the behavior.

By the time a behavior such as sit, down or come is put on cue the behavior is strong and essentially error-free. Training continues as the trainer works with the animal in different situations to make it even stronger and more reliable, but this is done in an incremental fashion to maintain that 80% success rate.

Animals Learn from Success

Research has shown that physical punishment, also known as "correction" is not necessary for dogs to learn, and in fact can have untoward effects such as inhibition of learning and aggression.

Additionally, we know from a recent study at Massachusetts Institute of Technology that success promotes learning whereas failure does not. In this study, done with monkeys, researchers observed that when monkeys were rewarded for correct responses they performed better than when they made an incorrect response and were not rewarded. In other words the monkeys were more likely to provide a correct response if a previous correct response was rewarded, rather than if they provided an incorrect response and received no reward. They learned from their successes, but not from their mistakes!

So, not only is physical punishment unnecessary and potentially hazardous to learning, but punishment in the form of withholding a food reward may not be very helpful to learning either.

Summary

In summary, animals are motivated to repeat behaviors they find rewarding. Clicker training helps an animal understand exactly which behavior to repeat to earn a food reward because the click marks the behavior and is followed by a food treat.

In clicker training, the cue is added after the animal learns the behavior through repeated reinforcement. This differs from "traditional" or military-style compulsion training where an animal learns a behavior along with the name of the behavior (command) through repetition and is corrected for non-compliance. Adding the cue after the animal has learned what will be reinforced helps prevent an animal from failing to respond correctly to the cue for the behavior.

Clicker trainers shape behaviors incrementally in order to keep an animal motivated and successful at each step of the training process. Research corroborates what clicker trainers already know - that animals do not have to be physically corrected for learning to occur and that success begets success. In the next section we will discuss specific training problems and how to correct them using clicker training.

Click here to go to

Part II: Correction, the Clicker Training Way

  • How Clicker Trainers Correct Behavior
  • How Clicker Trainers Correct Behaviors: Teaching Abby to Sit
    How Clicker Trainers Correct Behaviors: Teaching Abby to Sit