The speech signal provides listeners with information about both who is speaking and what is being said. Research on typical adults suggests that efficient comprehension requires integrating these two sources of information. Our project uses behavioral and neuroimaging methods to examine how babies learn to integrate these two sources of information, how bilinguals integrate this information across their two languages,and whether children with language impairment show deficits in integrating talker and linguistic information.

Approximately 7% of school-age children meet diagnostic criteria for Specific Language Impairment (SLI): low language ability despite nonlinguistic abilities in the normal range. While there has been little work investigating the neural organization of language in SLI, some previous studies suggest atypical lateralization in SLI: language processing relies heavily on the left hemisphere in children with typical language development, but may be more bilateral in children with SLI. This project tests whether these differences replicate in a larger sample with better controls, and adds an examination of prosody (the melody and rhythm of speech), an aspect of language that typically relies most heavily on the right hemisphere. This project was initiated by a UConn undergraduate student, and brings together an interdisciplinary team from SLHS and Psychological Sciences. The results will form the basis for a grant application to the National Institutes of Health.

Our project aims to shed new light on the organization of human lexical knowledge using graph theory.   We will consider both static and dynamic representations of human phonological and semantic knowledge.  By using cutting-edge graph-theoretic techniques for characterizing the dynamics of functional networks, we will develop new methods for analyzing the behavior and structure of computational models of human word recognition.

Research on genetically engineered mouse models is growing rapidly, with particular emphasis on phenotypes that can tie causal genetic mutations/variations to clinical conditions such as autism, depression, schizophrenia, and language disorders. To accomplish this, well-established and "mouse-friendly” tasks (Morris and other mazes, Open Field, novel object, rotarod, and various social paradigms) are often used. The application of engineered mouse models to the study of complex human cognition, however, calls for the development and validation of new tasks that can link to additional aspects of higher-order cognition such as categorization, object-constancy, structural logic, and other forms of rule-learning. A set of 4 new touch-screen operant testing stations recently acquired by the MBNF will be used to develop novel testing programs (via custom software) that can tap such cognitive measures in mice, and thus be used to study genetic modulation of cognition. Once validated, the tasks will seed future behavioral neurogenetic projects for the PIs, as well as others in the field.

This project investigates the psychological and neurological determinants of language variation in school-age and adolescent children with Autism Spectrum Disorder (ASD). Over the past decade, Naigles and Fein have collected intensive early language data from children with ASD and typically-developing controls. In this follow-up project, now in collaboration with Skoe, the children will be visited again in their homes to obtain neural measurements in the form of auditory brainstem responses.

Events typically entail change, but an under-studied topic in cognitive psychology concerns how we encode and track the changes that individual objects undergo as events unfold. On hearing “The chef will chop the onion”, we must keep track of multiple versions of the onion; before the chopping, and after. We are using simultaneous fMRI and EEG recording to explore the role of the brain’s memory systems and other structures during the comprehension of such events.