Can lack of experience delay the end of the sensitive phase for song learning

Some bird species will modify their songs in adulthood, whereas in others, once developed, song appears relatively fixed. However, even in some of the latter, social experience may lead birds to learn songs later than was previously thought possible. Do age-limited learners really exist or is failure to learn later in those species where it has been described a result of inadequate stimulation? If learning is normally restricted to a certain age period, is the decline in sensitivity due to age as such or to enough adequate information having been received? We discuss some of the issues involved and describe some experiments on zebra finches which we have carried out. Our results suggests that drastic changes in adulthood are rare. Birds deprived of tutors in the sensitive phase are more likely to produce elements learnt from the father before that stage, but show little evidence of learning from tutors encountered when adult (at 120 d+). Birds that are female-raised in groups and deprived of any male tutor till 120 days learn little from their tutors thereafter but have quite normal songs; these converge strongly within groups, as do the songs of all group-reared birds. Birds reared by both parents but without siblings, and then deprived of song during the sensitive phase, show learning both from the father and from the tutors encountered after 120 days. The most abnormal songs are those of birds reared by the mother alone and then kept singly till they were placed with singing adults at 120 days. The songs of these birds were more variable than those of the other groups, and they showed more evidence of learning between 120 and 240 days. These results do suggest some flexibility in the timing of song learning in zebra finches depending on experience but argue against profound change occurring once birds are adult even where their early experience has been greatly reduced.