Though the idea that mimics learn their sounds from other species might seem painfully obvious (why else call them mimics?) the theory had not been tested, at least not extensively. But university researchers studying bowerbirds have figured out that they do, indeed, learned by listening directly to other species, and not from other bowerbirds. Now the questions become ‘Why don’t mimics learn from each other?’ and ‘Is there an evolutionary advantage to not teaching one’s own young a species-specific song?’
It seems to me that there might be an evoluntary advantage to learning as many unique sounds as possible. If you can spend the time to learn, you must be in good condition, which might translate into your relative fitness as a potential mate. A form of sexual selection, or at least my pet theory.