Achievement

IGERT-trainee Constantine Lignos (2nd-year, Computer Science) has already begun a substantive integrative collaboration across departments. With IGERT faculty members Mitch Marcus (Computer Science) and Charles Yang (Linguistics), he has developed a cognitively plausible computational model of how children learn morphology (tense, plurality) in English. Work in morphology learning has thus far been primarily divided into two lines of research: cognitively-motivated models of morphology learning, which attempt to model human development and competency, and engineering-oriented models, which attempt to maximize application performance. In their recent talk and published paper, they address the gap between these approaches by presenting results from applying the learning model to child-directed data and comparing its learning process to research in child language acquisition. This highly successful collaboration was directly facilitated by our active IGERT program in Language Sciences.