Reward Based Training: Intermittent Reinforcement at Work

Now that we understand the technical aspects of “Intermittent Reinforcement,” how do we effectively put it to use with our pups?

Initially, when teaching your puppy a behavior or command, we will be rewarding every single time, keeping the dog on a continuous reinforcement schedule. This helps the puppy learn the exact behavior to receive the reward. Because dogs have a difficult time to generalize a command, you will want to do the behavior in many environments and while working on the 3Ds (distance, distraction, duration), but maintaining the continuous reinforcement schedule. Once your pup has seemingly mastered a command in most variations of locations, etc., then you will want to move to a variable ratio reinforcement schedule. This means you will intermittently give rewards for correct responses, but not in a predictable manner.

So, these are the general steps for teaching a new behavior:

1. Use a LURE for a few days to one week. Make sure to practice frequently and in as many places as possible during that time to help your pup generalize.
2. Move to a REWARD every single time you ask pup for behavior. This will last an extended amount of time, approximately a couple of months for that specific behavior. Make sure you are building on the 3Ds during this time and working the behavior in many different locations and situations to help your pup build a solid generalization and foundation for the command.
3. Move to INTERMITTENT REINFORCEMENT once your dog has a solid understanding of the command. You have been rewarding every time you ask for the behavior for several months and now your pup is able to give you the behavior, no matter the distraction, distance, or duration! Great! Now start to move to rewarding intermittently and in a variable schedule.
4. Gradually decrease your rate of reinforcement. When first moving to intermittent reinforcement for a behavior, you will want to reward more often than not. As the pup builds an even more solid understanding of the behavior, you can lower the amount of times you give a food reward.
5. Raise your expectations for the intermittent rewards. So now your pup has an amazing and solid response for a behavior! Lower your rate of reinforcement even more by only rewarding with food when your dog gives a perfect response or does the behavior in a difficult situation (ie – another dog walks closely by your dog while he is correctly performing the command you gave him).

Tips!

• Always include praise and touch! No matter which stage you are in of teaching a behavior to your puppy, you should ALWAYS include praise and occasional touch (the type your pup likes best) to build meaning to praise/touch and so that you always have those as “rewards” when not giving food rewards.
• Do not go back to a LURE! Once your pup has learned a behavior and is on REWARDS, you should not be luring a pup to get them to perform a behavior. Instead, ask your pup for the command 2 times, then change something if pup hasn’t performed by the 3^rd time. What can you change? Run through the 3Ds and determine which of those you can adjust to make it easier for your pup to follow the command. Also, consider the “Box of Acceptable Behaviors” and if your expectations for the behavior are too high.
•  If you are teaching a new behavior, start at the beginning with a LURE, no matter your pup’s age or where they are for other behaviors! If your pup is 10 months old and learning “stand” for the first time, but has other behaviors like “sit” and “down” at the intermittent reinforcement stage, you will still start with a lure for “stand” and move through all 5 steps. Your puppy has not had a chance to generalize the behavior of “stand” to all the situations to which he has generalized “sit” and “down.”
• You can always increase your rate of reinforcement, when needed. Although you should not return to luring, you can return to more rewards in situations that require it, such as walking past a big distraction. Even if you are at the stage of very rare rewarding, increasing your rate of reinforcement temporarily can help remind your pup how great it is to work with you!

Combining your knowledge and skills with the “yes” marker word, luring vs. rewarding, and intermittent rewarding, you can give your pup a great foundation solid skills!