ITALK is a collaborative EU-funded research project at the intersection of robotics, computer science, developmental psychology and cognitive linguistics.
The aim of the project is to explore interactions between different resources in cognitive development by investigating the co-development of action knowledge, conceptual structure, social interaction and a restricted form of language in an infanoid robot.
Work at USD is concerned with a variety of language-related issues such as
- developing suitable scenarios for language acquisition experiments with our robotic learner, iCub
- providing empirically motivated classifications of the linguistic input available to child language learners, i.e. constructional usage patterns in child-directed speech
- investigating user preconceptions and relevant interactional peculiarities of human-robot communication
Linguistic research in ITALK accommodates insights from usage-based Construction Grammar and is based on experimental and quantitative corpus-linguistic methodologies.