Animal self-efficacy?

As I sit here trying to think of what to write in this post, my dog jumped up on my bed with a ball in his mouth, his way of telling me that it is time to play fetch. And as this class has been more of a behaviorism class rather than a self-efficacy class, I decided I'd run with this thought.

As with all dogs, we originally trained him to fetch by throwing the ball and then rewarding him with a treat whenever he brought it back to us. But since he enjoys fetch so much, we haven't had to reinforce him for bringing the ball back; he seems, and I'm probably giving him too much credit here, to intrinsically value the game itself, making the reward unnecessary. But other tasks/games we've tried to teach him, like running through obstacles at the dog park, which seem equally fun and which we rewarded just as often, he won't spontaneously try to engage in, like he does with fetch. I wonder if there is some sort of animal-equivalent of self-efficacy at play here, in that he is really good at fetch but not nearly as good at the obstacle courses. I would say, at a basic level, I'm wondering about the interaction of self-efficacy and operant learning and how this interaction plays out in educational settings.