It’s learning based on confirming to the dog that he did something right — without giving him verbal hints. To make things easier, we divided the behavior into small parts and rewarded every correct step.