Research Projects
NIH Grant HD-01994 - Nature and Acquisition of the Speech Code and Reading
(Jay Rueckl, PI/A40)
Research Goals: The overall goal of the A-40 program is to understand how the language apparatus, biologically specialized for speaking and listening, becomes adapted to reading and writing. The program of research just approved for funding consists of four projects. For the first time in the more than 40 year history of this project, all projects include research on both speech and reading. (In past cycles, for the most part, a project was either about speech or about reading.) The first project is meant to extend our study of “articulatory phonology” or more generally our theory of phonological “practice” (that is, public use of language forms). In the domain of reading, we extend our earlier cross language research meant to expose the nature of the phonological information (we presume articulatory phonological information) accessed by skilled readers. Projects 2 and 3 develop our researchers’ theoretical account of the brain systems underlying language use in speech and reading. In its research on reading, Project 2 integrates neuroimaging, behavioral, and computational modeling results to extend Haskins’ model of skilled reading. Project 3 is a longitudinal study of children learning to read Finnish, English and Chinese. Developing proficiency in each group will be assessed with common psychometric, adaptive learning, and neurobiological (ERP and fMRI) measures so that relationships among attained skills, cognitive processing, and cortical organization can be tracked. The focus of Project 4 is on the relationship between reading ability and lexical phonology and how that relationship is changed for (1) disabled readers and (2) individuals whose dialect of English differs from standard English.
NIH Grant DC-02717 - Links Between Production and Perception in Speech
(Douglas H. Whalen, PI/A93)
Research Goals: The goals are to obtain data to substantiate the claim that speech production and perception are related and to validate the articulatory gesture as the link between production and perception. The method of inquiry involves attempting to show that those acoustic parameters that vary together as a result of some articulator movement also cohere in perception.
NIH Grant DC-04669 - Sensorimotor Control of Human Jaw Movement
(David Ostry, PI/A111)
Research Goals: The project is to explore the control of human orofacial movements, focusing on the jaw. The aim is to understand the ways in which neural signals interact with orofacial biomechanics to determine movement outcomes in speech. A common procedure is used throughout. The investigators use a servo-controlled robotic manipulator to deliver precise mechanical perturbations to the jaw. The research documents the jaw’s biomechanical properties and the relationship between jaw stiffness in static and kinematic variability during vowel production. It characterizes jaw stiffness during movement and assesses the comolexity of the information that the nervous system encodes concerning the biomechanical periphery and external loads in orofacial motion planning.
NIH Grant DC-008780 - Variability and Error in Speech Production
(Louis Goldstein, PI/A158)
Research Goals: This project addresses the relationship between 'normal' token-to-token variability in the production of phonetic units and tokens that can be characterized as containing ‘errors.’ Traditionally, speech errors are thought to occur when speakers produce something that was not intended through a categorical mis-selection of units during planning (e.g., 'fonal phonology' for 'tonal phonology'). Systematic studies of such sublexical errors have been largely carried out using phonetic transcription without any investigation of speech articulator activity. Conversely, studies of speech kinematics (and its variability) typically exclude tokens that are perceived as containing 'errors.' However, recent studies of articulatory kinematics during error-producing tasks have discovered a class of productions that sound like errors, but that do not involve categorical mis-selection of units. Rather, although an intended unit’s vocal tract constriction is correctly produced, it is accompanied—to various degrees across tokens—by an unintended constriction (an 'intrusion'). Thus, unintended constrictions represent a neglected source of token-to-token variation, and provide support for the hypothesis that errors and variation are more intimately linked than previously thought. Since the factors causing token-to-token variation are themselves poorly understood, Louis and colleagues propose to investigate patterns of variability and their relation to errors with three specific aims, for which they will:
• manipulate experimental factors that should contribute to making an utterance ‘hard to say,’ and measure their effect in three tasks: planning time (RT), articulatory variation in a speeded production task, and errors (as operationally defined) in a repetitive task, uncovering the patterns of correlations of these measures;
• test a dynamical account of errors as shifts in the behavior of pairs of coupled planning oscillators from more complex to simpler frequency ratios, with associated changes in fluctuations, i.e., variability; and
• examine the acoustic and perceptual consequences of intrusions to test the possibility that what have been called errors are simply instances of variation extensive enough to shift perceptual categories.
Relevance: Diagnoses of different speech disorders are often based on the clustering of error types and on the perceived variability of patients’ speech. The analyses will provide a sound theoretical and empirical basis for such acoustic and perceptual evaluations.
NSF ARRA Grant BCS-0922437 - Stochastic Time Models of Syllables
(Adamantios Gafos, PI/A174)
Research Goals: The proposed research develops analytical and modeling methods for evaluating the fit between theoretically posited syllabic parses and experimental data. Given a hypothesized syllable parse of a segmental string, our models will generate simulated temporal structure using a probabilistic version of Optimality-theoretic temporal alignment constraints. The simulated data will then be compared to the experimental data for their goodness of fit. In this way, a rigorous evaluation of hypothesized syllable parses is made possible. Furthermore, the methods to be developed enable analysis of the relation between posited abstract phonological organization, in the form of a syllable parse, and its predictions in terms of phonetic indices of that organization. The experimental data to be used are articulatory and acoustic data from English and Moroccan Arabic, two languages claimed to parse similar strings into very different syllable structures.
NSF Grant BCS-0966411 - From Endangered Language Documentation to Phonetic Documentation
(D. Whalen, PI/A179)
Research Goals: The proposed research determines whether satisfactory phonetic documentation can be based on textual materials from a documentation corpus that was not designed for phonetic analysis. It will then be determined as to how much material (tokens of a sound, number of speakers) is required to have an accurate estimate for the phonetic measures of an entire language. The study will provide a preliminary phonetic description of three endangered languages and will address a small number of specific phonetic questions in those three languages. The results will produce recommendations for best practices in language documentation to make future corpora as useful for phonetic analysis as possible.
NSF Grant IIS-1010336 - Subcontract agreement with the Rehabilitation Institute of Chicago, on a project entitled A Joint Database of Experiments and Models of Reaching Movement
(D. Ostry, Co-Investigator/A180)
Research Goals: David Ostry will provide critical insight into the data that should be shared for a wide range of theoretical and experimental approaches. He will assist in optimizing the database formats so that they are maximally useful.
NIH Grant HD-044073 - Subcontract agreement with Vanderbilt University, on a grant entitled Cognitive and Neural Processes in Reading Comprehension (K. Pugh, Co-Investigator/A181)
Research Goals: The project examines brain and cognitive measures of children with specific comprehension disorders in reading. Ken Pugh will collaborate on fMRI task development and brain / behavior analyses.
NIH Grant HD-065794 - Neurobiological Predictors of Spoken and Written Language Learning
(K. Pugh, PI/A182)
Research Goals: The project employs multiple neuroimaging methods to test predictive models of individual differences in learning consolidation for: 1) novel spoken and written word learning tasks that vary in demands on component processes relevant to reading, and 2) language and nonlinguistic tasks that systematically examine contributions from those cortical and subcortical neural systems associated with procedural and declarative memory as a window on basic learning and consolidation deficits in reading disability (RD). The overarching goal is to develop causal models of the ways in which any one or some combination of, these functional and structural factors might act to impede language and reading-related skill acquisition in RD.
NIH Grant DC-009263 - Development of Bimodal Bilingualism
(K. Pugh, PI/A184)
Research Goals: The proposed project is a subcontract with the University of Connecticut to provide consulting services to Dr. Lillo-Martin of the University’s department of Linguistics. The study will include testing participants at research fairs in New England, and bimonthly home visits for longitudinal participants, and then transcribing the findings following the conventions established using EUDICO Linguistic Annotator (ELAN) format.
Haskins Center - Haskins Language and early assessment research network (LEARN)
(J. Preston, J. Irwin, PI/A185)
Research Goals: The proposed study will focus on developing a scientific understanding of how children acquire language and overcome difficulties in this process. The goal of this knowledge is translational: to inform and improve educational practice and, to the degree possible, to help prevent language difficulties through early intervention and prediction.
NIH Grant DC-011342 - Neurobiological Signatures of Audiovisual Speech Perception in Children in ASD
(Irwin, PI/A186)
Research Goals: The proposed study will address a critical problem in the study of autism spectrum disorders (ASD); heterogeneity in language functioning. The research will study the neural response associated with perception of auditory and audiovisual speech in children with ASD by pairing EEG/ERP and eye-tracking technology to collect perceptual data to identify biomarkers associated with language outcome.
NIH Grant HD-067364 - Neurocognitive Determinants of Second Language Literacy Development in Adolescents
(K. Pugh, PI/A187)
Research Goals: The project comprises a comprehensive investigation of the neurocognitive parameters that affect how adolescents acquire and learn to read a new language. The project will employ a longitudinal design in which we will recruit cohorts of adolescents ranging from a basic to medium literacy level in a second language (L2) and track skill development with both behavioral and fMRI measures over 24 months. Cohorts will be recruited in both Israel and the U.S.; thus, each language will serve as both L2 and L1. Specific aims are 1)To investigate how learning to read in L2 is jointly determined by the linguistic structure of L1 and by individual differences in neurocognitive capacities of the reader; 2)To investigate whether acquiring reading fluency in a second language necessarily depends on acquiring “native-like” neurocognitive markers; 3)To investigate the linguistic and general neurocognitive consequences of learning a new set of statistical regularities in L2. In addition, a cross-sectional fourth aim contrasts Hebrew vs. Spanish as L1 in order to assess both the generality of findings from Hebrew and investigate the impact of qualitative differences in the underlying linguistic structures of an L1 on neurocognitive indices of reading English. This research will directly inform theories of second language learning, and holds promise to inform research on optimal approaches to second language curriculum development.
NIH Grant DC-012502 - Speech Motor Learning and Sensory Plasticity in Children and Adults
(D. Ostry, PI/A191)
Research Goals: The proposed studies will focus on the neural substrates of speech motor development in children. The plan is to use a multi-modal approach that combines advanced psychophysical and neuroimaging techniques. This will test the idea that patterns of brain activity during task based learning that are related to these measures can be used to predict durable changes in functional connectivity that are observed after the completion of the learning task.
NSF Grant IIS-1161962 - Multilingual Gestural Models for Robust Language-Independent Speech Recognition
(H. Nam, PI/A193)
Research Goals: In previous research, great strides were made in developing a gesture-based automatic phone recognition system for American English that shows rubustness to noise and coarticulation. In this proposal, this previous work will be extended to build a large vocabulary automatic speech recognition system for American English and language-independent gestural models to perform multilingual ASR.
NIH Grant HD-071988 - Individual Differences in Learning Potential for Language and Literacy
(D. Braze, PI/A195)
Research Goals: The unacceptably high incidence of poor literacy skills among American young people is a public health crisis that is both insufficiently understood and understudied. Low reading skill in adults is consistently associated with many negative outcomes, including lesser economic success, increased risk of poor mental and physical health, and poor outcomes for offspring (Kutner, et al., 2007; National Institute for Literacy, 2008). There is a growing appreciation that we urgently need to identify the learning mechanisms for those skills that underlie reading comprehension, and to better understand how they fail in low literacy individuals. Our focus is on young adults, but not the predominantly middle class university students who are most often the subjects of reading research in this age range, but a far less restrictive community-based sample obtained from adult education programs, community colleges, organizations that serve young people and from the community at large. Thus, our sample encompasses a wide range of skill levels including many who have or are at risk of dropping out of the educational system after experiencing repeated learning failure. To date there has been remarkably little direct study of learning processes pertaining to literacy skills and the cognitive abilities that make reading comprehension possible. Clarifying the connections between learning capacities and literacy skills is essential for gauging potential for remediation. This project will build on research from our own group and elsewhere showing that poor readers exhibit reliable differences in learning of linguistic and orthographic structure. An important innovation of the project is its use of experimental learning tasks to determine connections among capacities for learning in several theoretically important domains of linguistic and orthographic structure. These tasks include probes of (a) learning relationships between sounds of words and word meanings; (b) learning relationships between spellings and word meanings; and critically; (c) learning relationships between sounds and letters; (d) learning for improved efficiency in processing complex syntactic and semantic structures typical of written language. Individual learning profiles derived from the proposed experiments will be related to scores from a comprehensive test battery indexing capacities that are known to differentiate better from less skilled readers within our study population. This will allow us to connect to the existing knowledge base on individual differences in reading comprehension. A second key innovation of the project is to target a large understudied population that is socially and economically important. Whereas most past studies of individual differences in literacy focus on children at early stages of acquisition, our project focuses on older individuals who are products of our primary and secondary schools but who are in many cases unprepared for further education or the demands of the workplace. The prevalence of poor readers in our samples allows us to study the outcomes of atypical developmental trajectories, providing important constraints on current theory. Our research will elucidate mechanisms for the promotion of effective reading skill within this segment of the population. Thus, through improved understanding learner subtypes, this research speaks to the issue of creating more effective learning environments for low-literacy secondary and post-secondary students.
NIH Grant HD-075740 - Training-Induced Plasticity in Human Motor and Sensory Systems
(D. Ostry, PI/A196)
Research Goals: The planned studies focus on the sensorimotor system and explore the idea that training induced changes to the brain spread from the motor to somatosensory areas of the brain and vice versa. The plan is to address the effects of motor learning on sensory systems and of somatosensory perceptual training on motor systems by using an approach that combines psychophysical, neurophysiological and neuroimaging techniques. The ability to quantify changes to brain plasticity that accompany both somatosensory training and motor learning may permit a better understanding of the broader effects of neurological rehabilitation on sensorimotor disorders. Imaging the sensory and motor networks of the brain that are associated with both somatosensory and motor learning may also lead to better diagnoses and tracking of brain neuroplasticity during therapy. Our approach may aid in the development of neuroscience-based strategies for training and rehabilitation.
NIH Grant HD-073288 - Retrieval Interference in Skilled and Unskilled Reading Comprehension
(J. Van Dyke, PI/A197)
Research Goals: Poor reading ability has profound cognitive, emotional, and behavioral consequences for the developing child, and—if unremediated—eventually has economic consequences for the adult. Indeed, the 2003 report on Adult Literacy and Life Skills (Statistics Canada & OECD, 2005) estimates that 51% of US adults aged 16-25 can read only simple texts and make only low-level inferences, a level termed “below-basic,” which is insufficient for attaining advanced educational and occupational goals. These statistics point to the need for understanding factors contributing to poor reading ability beyond the single-word level in the adult population. Unfortunately, the bulk of linguistically-based research into sentence and text-level comprehension has emphasized a limited set of general cognitive capacities (especially working memory capacity) as the source of comprehension difficulty, and focused primarily on the college-level population. This proposal brings together findings from three so far unintegrated research communities (memory, adult sentence and discourse processing, and reading disability) and an alternative research sample to create a novel approach towards understanding poor comprehension. The project is built around an architectural framework that emphasizes memory retrieval as the mechanism connecting word-reading skills and higher-level integrative skills. Building on memory research pointing to a severely limited active memory capacity, even for skilled readers, we assume that comprehension is primarily determined by the successful retrieval of information from passive memory. Thus, in our Specific Aim 1 we investigate the conditions leading to successful retrieval, including the way that different types of linguistic cues (phonological, morphological, syntactic, semantic, pragmatic) are combined, and the conditions leading to failed retrieval—namely the presence of memory interference. Our Specific Aim 2 combines long-standing conclusions from memory research, suggesting that sensitivity to interference will be determined by the quality of to-be-retrieved representations, with research in reading disability, arguing that individuals will vary in the average quality of their word representations, which are determined by their reading experience and facility for processing different types of linguistic information. Finally, our Specific Aim 3 investigates the neurological bases of sensitivity to retrieval interference, with primary focus on the neural networks that link skilled word reading to higher level processing, and the contribution of regions responsible for mediating interference at each level, especially left inferior frontal gyrus. In order to have broad impact, we target a community-based sample of mainly non-college-bound individuals (age 16-24), who we believe constitutes a more representative sample of reading ability in the population at large. In addition, recruit a supplemental sample of beginning readers (age 7-9) to investigate the development of retrieval and sensitivity to interference. We expect that this project will result in a new conception of sources of individual variability in reading comprehension, and a deeper understanding of how these develop and persist in the adult population.
NIH Grant DC-012350 - Kinematic Patterns of Phonetic Convergence
(T. Mooshammer, PI/A198)
Research Goals: In this proposal we pioneer a new approach that for the first time supports simultaneous direct observation of speech articulation by two face-to-face talkers, and apply it to studying the mutual adaptation that occurs between them as they interact in conversation. This will lead to improve understanding of the articulatory aspects of foreign accent and suggest possibilities for improving methods of speech remediation and second language instruction