The objective of this randomized crossover study is to test the hypothesis that scores on the Bayley Scales of Infant Development-III are higher than scores on the Bayley Scales of Infant Development-II in the same group of infants who were born very preterm. First, however, the investigators must demonstrate that significant "learning" does not occur when the tests are administered 4-8 weeks apart. The investigators hypothesize that scores on the Bayley-III will be higher than scores on the BSID-II, irrespective of the order in which the tests are administered.
