Cognitive 2014 - Workshop on Cognitive Aspects of Computational Language Learning

The human ability to acquire and process language has long attracted interest and generated much debate due to the apparent ease with which such a complex and dynamic system is learnt and used on the face of ambiguity, noise and uncertainty. This subject raises many questions ranging from the nature vs. nurture debate of how much needs to be innate and how much needs to be learned for acquisition to be successful, to the mechanisms involved in this process (general vs specific) and their representations in the human brain. There are also developmental issues related to the different stages consistently found during acquisition (e.g. one word vs. two words) and possible organizations of this knowledge. These have been discussed in the context of first and second language acquisition and bilingualism, with crosslinguistic studies shedding light on the influence of the language and the environment.
The past decades have seen a massive expansion in the application of statistical and machine learning methods to natural language processing (NLP). This work has yielded impressive results in numerous speech and language processing tasks, including e.g. speech recognition, morphological analysis, parsing, lexical acquisition, semantic interpretation, and dialogue management. The good results have generally been viewed as engineering achievements. Recently researchers have begun to investigate the relevance of computational learning methods for research on human language acquisition and change.
The use of computational modeling is a relatively recent trend boosted by advances in machine learning techniques, and the availability of resources like corpora of child and child-directed sentences, and data from psycholinguistic tasks by normal and pathological groups. Many of the existing computational models attempt to study language tasks under cognitively plausible criteria (such as memory and processing limitations that humans face), and to explain the developmental stages observed in the acquisition and evolution of the language abilities. In doing so, computational modeling provides insight into the plausible mechanisms involved in human language processes, and inspires the development of better language models and techniques. These investigations are very important since if computational techniques can be used to improve our understanding of human language acquisition and change, these will not only benefit cognitive sciences in general but will reflect back to NLP and place us in a better position to develop useful language models.
Success in this type of research requires close collaboration between the NLP, linguistics, psychology and cognitive science communities. The workshop is targeted at anyone interested in the relevance of computational techniques for understanding first, second and bilingual language acquisition and language change in normal and clinical conditions. Long and short papers are invited on, but not limited to, the following topics:
Computational learning theory and analysis of language learning and organization
Computational models of first, second and bilingual language acquisition
Computational models of language changes in clinical conditions
Computational models and analysis of factors that influence language acquisition and use in different age groups and cultures
Computational models of various aspects of language and their interaction effect in acquisition, processing and change
Computational models of the evolution of language
Data resources and tools for investigating computational models of human language processes
Empirical and theoretical comparisons of the learning environment and its impact on language processes
Cognitively oriented Bayesian models of language processes
Computational methods for acquiring various linguistic information (related to e.g. speech, morphology, lexicon, syntax, semantics, and discourse) and their relevance to research on human language acquisition
Investigations and comparisons of supervised, unsupervised and weakly-supervised methods for learning (e.g. machine learning, statistical, symbolic, biologically-inspired, active learning, various hybrid models) from a cognitive perspective