Why does sleep produce an immediate deteriorating effect on singing skill when the bird's performance is still rather low? This is presently an unsolved question. Experimental deafening can induce a deterioration in song performance in adult song birds similar to that seen after sleep in juveniles (Nordeen and Nordeen, 1992), and it has been speculated that neuronal reactivations of newly acquired song representations during sleep produce such deterioration, because they likewise take place in the absence of auditory feedback (Nick and Konishi, 2005a and Nick and Konishi, 2005b). Thus, in the initial learning phase sleep-associated reactivations induce an unsupervised learning process that might also enhance inaccurate aspects of song performance, due to the lack of auditory feedback. After a more generalized representation of the song has been formed, during sleep such pre-existing representations might replace the missing acute auditory feedback. Those pre-existing representations might then serve as a reference template for the reactivation of newly encoded sensorimotor information to produce a further shaping and improvement in song quality. Brain maturation might play an additional role. Crandall et al. (2007) found that the degree of song deterioration across sleep was inversely correlated with the amount of neuronal spiking in the HVC. Juveniles with the greatest deterioration showed the lowest HVC firing rate during sleep. Importantly, there is also a general increase in HVC firing rate during development. Juveniles during song learning show generally a distinctly lower firing rate of the HVC during sleep than adults. Thus, it was proposed that the increasing spiking activity of HVC neurons stabilizes the maturing song (Crandall et al., 2007). Whatever the underlying mechanisms, the similar dynamics in the effects of sleep on skill memory in song birds and humans during development further adds to the notion of a general role for sleep in long-term memory formation that is independent of the species