Funded by the National Science Foundation
Language Acquisition and Language Processing:
Finding New Connections
The special session the conference will be “Language Acquisition and Language Processing: Finding New Connections.” It will include presentations from six eminent researchers who bridge language acquisition and processing, a poster session dedicated to this theme, and awards for student presenters.
Two central questions addressed in psycholinguistics pertain to acquisition and processing. First, how do infants, exposed to a relatively small sample of language use in context, come to acquire a complete linguistic system that can express an almost limitless number of ideas and conceptions? Second, how do adults access this acquired linguistic knowledge in such an expert manner that they achieve interpretation in real-time as the speech unfolds, typically making distinctions on a millisecond time-scale? Traditionally, these two questions have been addressed separately, with the field taking a divide and conquer approach, often with great success. However, over the last two decades, the distinction between these questions, and their artificial divide in the discipline, have been blurred. Children as young as 24 months have been observed to interpret speech in real-time, deploying their incomplete knowledge of the language almost as quickly as their adult expert counterparts. Adults have been observed to be highly adaptive, learning new patterns of speech, new terms, and even new syntax, from brief exposures. Interestingly, developmental differences are beginning to be uncovered in both language processing (e.g., children’s failure to revise real-time interpretations) and language learning (e.g., whereas in some cases children regularize ‘noisy’ input, adults probability match).
The special session at the 2021 CUNY conference will bring together researchers investigating the mechanisms of language acquisition and process, seeking cross-pollination with mainstream (adult) psycholinguistic research. To be clear, it is not the case that the divide and conquer approach prevalent in the subfields of language acquisition and processing is ready to be abandoned. However, dialogue between these subfields is increasing to the point that important theoretical unifications are in sight. By bringing eminent acquisitionists and computationalists to CUNY, and inviting young researchers to submit work on this topic, it is our hope that we may begin to better understand how the processing of linguistic input (and the ambient referent world) is connected to acquisition, and whether linguistic universals should be best motivated by learning or processing constraints, or both. This bridging territory is fertile ground not only for new research but for new research questions.
We encourage abstract submissions on this special session topic.
Anne Christophe, École Normale Supérieure
Anne Christophe is Professor and Director of the Laboratory of Cognitive Science and Psycholinguistics (LSCP) at the École Normale Supérieure in Paris, France. Christophe is a leading scientist in the study of language acquisition. Her work focuses on how infants and young children acquire their native language, especially on how they learn the language’s phonological, syntactic and lexical characteristics. She is well-known in the field of infant language acquisition for her work on the synergies between early lexical and syntactic acquisition; she is an expert of infant experimentation, with a variety of real-time measures in children such as eye-tracking and EEG. Some of her most recent work examines how infants and children learn the meaning of novel function words (such as new articles and pronouns), examining how children respond to these terms in running speech, for purposes of parsing and interpretation. This work directly connects parsing, grammar learning, and reference, in ways of central interest to those CUNY attendees interested in similar issues in adults. She has received considerable attention for her work, garnering international invitations to speak, and numerous awards, including most notably the CNRS Scientific Excellence Award (2012-2015, 2017-2020) and the CNRS Bronze medal. She is an elected member of Academia Europaea.
Cynthia L. Fisher, University of Illinois
Cynthia Fisher is a Professor of Psychology at the University of Illinois, Urbana/Champaign. Fisher is a leading figure in the study of word and grammar learning in young children. Her work focuses especially on understanding how preschoolers use different sources of information to identify the meanings of verbs. In some of her most notable work, children hear novel verbs used in different syntactic contexts to describe the same event. She finds that these children, as young as 18-20 months, interpret these verbs as describing different aspects of the event depending on the particulars of the syntax. Her work has turned toward the origins of this syntactic bootstrapping process, by exploring how children dynamically update their discourse representation of a dialogue, by making predictions about what is likely to be referred to in the next utterance. This work makes direct connections to CUNY attendees interested in real-time interpretation of discourse in adults. Other work in her lab explores phonotactic learning, examining how infants represent the phonological properties of speech. Fisher has been awarded several grants for her research from NSF and NIH. She is Editor of Language Learning and Development and is an elected fellow of the Association for Psychological Science.
Jeffrey Lidz, University of Maryland
Jeffrey Lidz is a Professor in the Department of Linguistics at the University of Maryland. Lidz earned his PhD from the University of Delaware in 1996 and was a postdoctoral fellow at the University of Pennsylvania and the Laboratoire de Science Cognitive et Psycholinguistique in Paris. Lidz is a highly distinguished linguist, who is best known for his theoretical and experimental work on language acquisition and comparative syntax-semantics. His work focuses especially on the relation between syntax and semantics and how children discover the mapping between the two. He has studied this topic in the domains of quantification, negation, verb raising, and verb argument structure. Two aspects of his work will be of particular interest to CUNY attendees. First, he has examined how incremental parsing in in 2- to 4-year-olds affects what can and cannot be learned about novel words in these sentences. Second, he has asked how quantifiers connect to children’s representation of number in perceived referent worlds. Both topics (real-time interpretation, quantification and reference) are frequent topics of this conference. Lidz has been awarded several research grants for his research from NSF and the McDonnell Foundation, and has earned awards for teaching and graduate mentorship. He is Editor in Chief of Language Acquisition: A Journal of Developmental Linguistics and has been Associate Editor of the Journal of Semantics. He is a member of the Linguistics Society of America.
Christopher Manning, Stanford University
Christopher Manning is the Thomas M. Siebel Professor of Machine Learning in the Departments of Computer Science and Linguistics at Stanford University. He is a leading figure in the field of Natural Language Processing (NLP). He is best known for his work on using Deep Learning methods within NLP, including Tree Recursive Neural Networks, the GloVe model of word vectors, sentiment analysis, neural network dependency parsing, neural machine translation, question answering, and language understanding. Manning distinguishes himself by connecting his work to the broader field of language research, in both linguistics and psycholinguistics. He has co-authored with those in psycholinguistics, and wishes to examine how computational principles might inform theories of human language processing and acquisition. For these reasons, his work is of central interest to the CUNY attendees who wish to understand the computations that support human language comprehension and production. Manning has coauthored influential textbooks on statistical approaches to NLP and information retrieval. He is an ACM Fellow, an AAAI Fellow, and an ACL Fellow, as well as past president of the ACL (2015).
Elissa Newport, Georgetown University
Elissa Newport is a Professor in the Department of Neurology and Director of the Center for Brain Plasticity and Recovery at Georgetown University. She is head of the Language Learning and Development lab and the Pediatric Stroke Research Project. Newport is a highly distinguished scientist in the study of language acquisition. The breadth and significance of Newport’s work makes contact with several regular themes of the CUNY conference: her work on statistical learning directly connects to use of statistical cues in parsing; her work on critical periods directly connects to issues about second language (L2) processing; her work on neural plasticity and infant stroke recovery directly pertains to the neural basis of language processing. She has been elected a fellow of the Association for Psychological Science, the Society of Experimental Psychologists, the Cognitive Science Society, the American Association for the Advancement of Science, the American Academy of Arts and Sciences, and the National Academy of Sciences. Her research has been supported by grants from the National Institutes of Health, the National Science Foundation, the McDonnell Foundation, and the Packard Foundation. In 2015 she received the Benjamin Franklin Medal in Computer and Cognitive Science.
Linda Smith, Indiana University
Linda Smith is the Distinguished Professor and Chancellor’s Professor of Psychological and Brain Sciences at Indiana University. Over a highly distinguished career, Smith has conducted ground-breaking work on infant development in language, perception and action. She is particularly interested in understanding how changes in these different systems support early language learning. Her work is notable for its examination of behavior and development at multiple times scales, from the millisecond time scale of perception and interpretation to the multi-month time scale of learning and development. Two aspects of Smith’s work are of central interest to CUNY attendees. First, her work on the dynamics of infant attention in natural language tasks (via head-mounted cameras and eyetrackers) connects directly to the “visual world” eyetracking approach now common in the field of adult spoken language processing. Second, her interest in naturalistic settings and input connects directly to those interested in similar issues in spoken dialogue, a frequent topic of CUNY presentations. She is an elected fellow of the American Psychological Society, the American Academy of Arts and Sciences, the Cognitive Science Society and the National Academy of Sciences. She has earned numerous awards for her research, including the William James Fellow Award (APA), and the David E. Rumelhart Prize in Cognitive Science.