top of page

Using Capturing to Train New Behaviors

Capturing is a training technique that rewards naturally performing behaviors and puts them on cue. It requires the behavior to already exist in the animal’s repertoire. You can’t capture behaviors that the learner doesn’t demonstrate.

Example: Can’t teach “speak” to a dog that never barks.


Develop the learning environment.

  • Identify contexts and factors that …

    • Increase the likelihood of the behavior manifesting.

      • Include these in the learning environment.

    • Decrease the likelihood of the behavior manifesting.

      • Remove these from the learning environment.



            Teaching a cat to meow – train around dinner time

            Teaching a dog to roll on their back – go on fresh cut grass

            Teaching a chicken to scratch the ground – give them some fresh dirt

In our example, we’ll be teaching a dog to lie down using capturing.


Factors to include:

  • Quiet room

  • Main living space

  • Favored napping area

  • Patience!


Factors to remove:

  • Other pets or people

  • Lots of noise and commotion

  • Lingering high energy (exercise first!)

Part I – Identifying the Behavior

  • Set up the learning environment

  • Wait for the behavior to occur

  • Mark the behavior

  • Reward with a high value treat

  • Reset the behavior

  • Repeat


You’ll set up the learning environment to ensure a high likelihood of the target behavior manifesting naturally. Your job is to stay quiet and watchful. You want to be sure to mark every time the animal performs the target behavior. This ensures quick understanding of the behavior being rewarded and reduces frustration. Expect the first training session to last 20-30 minutes. There’s a lot of waiting involved.


As soon as the animal finishes the target behavior, mark (clicker, verbal “yes,” etc.), then give a high value reward.


The first couple times, the animal will likely be confused and unsure what prompted the reinforcement. Through repetition, they will eventually identify your target behavior, then more actively perform it to obtain further reinforcement.


Depending on your target behavior, using resets is necessary to prime the learner for trying again. In our example, we’re trying to teach a “down.” If the dog is already lying down, however, they can’t lie down again. They need to reset their bodies to either a standing or sitting position before they can offer it again. This isn’t necessary for all behaviors, for example, speaking. They speak just once, mark, reward, and they’re already reset.

Part II – Naming the Behavior

  • Set up the learning environment

  • Wait for the behavior

  • Cue just before the behavior

  • Mark the behavior

  • Reward

  • Reset


Up to this point, you are not prompting the behavior. You are simply waiting for it to spontaneously occur. You know you’re ready to add the cue when you can tell the learner is just about to offer it. For a “down,” most dogs will shuffle their feet a little, then lie down. For speaking, many cats get a bit of tension in their cheeks and make strong eye contact before meowing. You need to identify your learner’s “tell” before adding the cue. This is important because you’ll want to add the cue right before they start the behavior. This is because the cue is a predictor for the behavior and then finally a prompt for the behavior. If you use the cue and the behavior doesn’t happen too frequently, the cue means nothing to the learner. Timing is crucial.


Just like Part I, you’ll set up the environment and wait for the behavior. Consider doing a few warmups without adding a cue first. This allows you to get your observation skills in place and reminds the pet of what is expected in this context. When you’re ready and can easily predict the behavior, start adding the cue. Remember, aim to give it just before they start the behavior, then mark and reward like before. Repeat at least 10 times to make the pairing of the cue and behavior clear to the learner.

Part III – Testing the Behavior

Up to this point, you’ll be adding the cue when the animal decides to use the behavior. Now we need to test it when it doesn’t look like they’ll offer it. The learning environment should still be the same, but your timing will be different. Wait for a lull in offerings, use the cue, then wait for a response. If the animal does not perform the behavior immediately after cueing, going back to Part II and continue pairing the cue and behavior. We don’t want to build a habit of “human says weird word and I do nothing.” One “no response” and we go back a step. We’re build a habit of “human says ‘down’ and I lie down.”


If the animal does understand the cue, you’ll use it and they will immediately perform the behavior. Make sure you give them a big “jackpot” (lots of treats in a row) to solidify their understanding.

Part IV – Generalizing the Behavior

Finally, we’ll want to start removing the environmental factors we put into place when we first started. Though we’ve taught the learner a distinct cue, many of them will still rely on the learning environment to be part of the cue. We want to take away those factors by practicing in a new room, with a different person, standing versus sitting, etc. Be sure to change the context slowly or you may need to “reteach” the behavior in this next context. Some animals will rely on context cues more heavily than others. Such learners will need extra help realizing that “down” means “down” no matter where you are.

14 views0 comments

Recent Posts

See All


bottom of page