I'm pointing to the rewards of manageable risk as a mechanism - probably an evolutionary mechanism - for exploring boundaries and thereby gaining things, even if you didn't know they existed.
It's all very well to talk about the rewards of new assets, freedom etc., but the reward from a risky venture isn't necessarily obvious; it may even be utterly unknown in the history of human kind. But if the risk itself being rewarding is a mechanism, it may encourage the discovery of such rewards.
The key misunderstanding problem here is the overloading of language. We have this talk of rationality, of evolutionary psychology, of emotions and drives. The key thing to understand, though, is that all may simply be different ways talking about the same things.
I'm saying that both things can be true: that it's rational consideration of long-term goals that cause us to risk things; and that it's the intrinsic utility of risk itself as encoded in the genome and proteome for the self-directed organisms we call humans. What I think is wrong is to take only a single terminology, and use it to say the other terminology is mistaken.