1. Finish normative morphologic data analysis and carry out additional statistical measures such as t-tests and ANOVA's to fully exploit the multidimensional relationships in the three year longitudinal data. Two papers, both co-authored with Dr. Diane Kurtzberg, are in progress: one on lateral asymetries of the AEP to Speech Sounds and the second on normal longitudinal developmental patterns in the AEP to Speech Sounds from birth to two years of age. 2) Carry out 16-electrode topographic analysis of the same data to separate morphologic features overlapping in time, localize different neural generators underlying different effects, and trace how these generators change in the course of development over the first three years of life, and publish topography paper (also co-authored with Dr. Diane Kurtzberg). 3) Apply AEP norms to evaluate longitudinal AEP data in population of high-risk infants (birth weight less than or equal to 1500 gms, and/or neonatal asphyxia-Apgar Score less than or equal to 4 at 5 min., and/or neonatal ventilatory support greater than or equal to 72 hrs.), also collected over the last three years, correlate with profiles of medical, neurobehavioral, and cognitive language data available on the same children, and publish results. Further incorporate use of resulting early AEP indices into clinical evaluation and management. 4) Pilot a new AEP study on the phonological, semantic, and syntactic levels of language development appropriate for following these same high-risk infant followup and control populations from 4 to 6 years of age. The stop consonant vowel CV syllable stimuli and discriminations were appropriate to the known competence of infants from birth to early preschool age, but higher-level stimuli and paradigms are needed to probe higher language functions between 4 and 6 years of age. The rationale is that by the end of this age range, normal children are expected to have mastered most of the phonologic, semantic, and syntactic rules of their native language so that their failure to do so should be detectable.
