I'm a professor/researcher in artificial intelligence and cognitive science with a special interest in language learning - computational cognitive psycholinguistics. That is I am particularly interested in building computationally and cognitively plausible models of how we learn and communicate about the world, and how to get computers/robots to learn and communicate about the world in a similar way, as well as how better to teach people languages and about language.

From the perspective of learning, my focus is on unsupervised learning - children learn to speak without a teacher, without being taught what nouns and verbs are. From the perspective of language, I like Tagmemics and Cognitive Linguistics - consistent with my focus on unsupervised learning, language and learning boil down to what we think is similar.

Formally this gives rise to similarity or distance measures used in learning, and we also need to account chance levels or prevalences. It also gives rise to metaphor, metonymy, grammar and related linguistic constructs, which define Cognitive Linguistics, as well as the basic ideas of Phonemic and Tagmemic analysis, Contrast in Analogous Environments and Complementary Distribution. It similarly explains the richness of polysemy and why there is really no such thing as synonymy, as when anything is different we look for other things to be different to learn from, using the part that is similar as context or paradigm, a substrate for the learning. We are continuously developing and modifying the range of meaning of language with our personal, social and cultural development, as we press old words or morphemes into new roles.

I also edit the new Springer journal, Computational Cognitive Science that aims to bridge the gap between Computational Intelligence and Cognitive Science.