Backward chaining describes...

• A. Begin with the end behavior; reinforcement then leads back to the start.
• B. Start with the first step and build forward.
• C. Use random reinforcement.
• D. Only reinforce after a complete sequence.

Answer: (A)

Backward chaining builds a multi-step behavior by securing the final action first and then adding previous steps in front of it. You start by teaching and reinforcing the last part of the sequence so that it’s performed reliably. Once that final action is solid, you train the step before it so that the dog must complete the preceding action and then execute the end action to earn reinforcement. By continuing this pattern, you end up with the entire sequence—sit, fetch, and drop, for example—being performed in order, with reinforcement tied to the completed end of the chain.

This approach is especially effective because it ensures the dog always has a clear, rewarding end to work toward, which anchors the entire behavior pattern. It also makes it easier to shape complex sequences, since you don’t rely on the dog reliably starting with the first step and carrying through every time before receiving reinforcement. For instance, if you want a dog to perform a chain like sit, fetch, and drop, you’d start by reinforcing the final action (dropping the item into your hand), then add the fetch before it, and finally the sit before the fetch, so the whole sequence ends with the reward.

The other methods described—starting with the first step, using random reinforcement, or waiting to reinforce until a complete sequence is finished—do not align with this step-by-step backward buildup, which is why this option is the best match.