Colin Phillips, Ph.D. - US grants

The National Science Foundation will support a workshop on The Relation between Parsing and Production, to be held at the University of Maryland, College Park, as a part of the 17th Annual CUNY Conference on Human Sentence Processing, March 25-27, 2004. The workshop will bring together leading researchers on language production and language understanding from psychological, computational and neuropsychological perspectives, with the goal of generating a new synthesis of the parallels and differences between production and comprehension. Until recently, research on speaking and understanding investigated largely different phenomena, using largely different methodologies; this led to the view that sentence comprehension and sentence production are largely independent cognitive systems. Recent advances in both fields now suggest a different picture: parallel experimental and computational investigations of speaking and understanding indicate that there is substantial overlap in the mechanisms used for both tasks. The specific questions addressed at the workshop will include the following: What do sentence comprehension and production have in common, beyond the fact that they use the same language? Do comprehension and production operate on the same time-scale? How can experimental findings of close parallels between speaking and understanding be reconciled with the finding of dissociations between speaking and understanding in language disorders? What methodologies will allow closer comparison of comprehension and production?

The workshop will have broad impacts for both research and education. A proper understanding of the relation between language production and language comprehension is important for a number of different fields, including language learning and especially the study of language disorders, where language comprehension and language production have often been found to be differentially impacted. An improved understanding of the relation between comprehension and production also has implications for the development of computational models of interactive natural language agents. The workshop will foster interaction among researchers in different fields, and by supporting the participation of students, it will enhance the training of a new generation of researchers. Results of the workshop will be made available through publication in a special journal issue.

A central problem for a theory of language acquisition is to determine how children learn both what is possible and what is not possible. This problem is especially acute for phenomena residing at the boundary between syntax and semantics, where (a) the mapping from surface form to meaning is often complex and (b) languages vary in how form and meaning align. This project aims to use the acquisition of language-specific constraints on scope interpretation as a probe into the character of the syntax-semantics mapping and the learning of this mapping. Several constructions in Japanese do not show scope ambiguities that the corresponding English sentences exhibit. Under the direction of Dr. Colin Phillips and Dr. Jeffrey Lidz, Mr. Takuya Goro will investigate Japanese preschool children's interpretations of those constructions, using the Truth Value Judgment Task. The results from the project will expand the empirical coverage of studies on the acquisition of scope. Much previous research in this domain has investigated the pragmatic and processing constraints that might underlie children's bias for surface scope interpretations of scopally ambiguous sentences. In contrast, this project focuses on children's mastery of constraints that exclude certain scope interpretations. The project will also contribute to an improved understanding of the mechanism that is involved in the acquisition of language-specific constraints on scope interpretations Furthermore, the project will determine whether children have initial biases towards particular kinds of interpretations, and will therefore contribute to an understanding of the initial state of language learning. One broader impact of the study is that the research project will help to establish new partnerships for language acquisition research on Japanese, based on developing institutional connections with preschools, and training Japanese researchers in state-of-the-art research methods in the study of language development. Partnerships with preschools are common in the US, but relatively rare in Japan. The new relationships developed in this way will benefit not only this research project, but also future international collaborations on comparative language acquisition studies on Japanese and English.

Human language is both universal within the species and highly variable across populations. This Integrative Graduate Education and Research Traineeship (IGERT) project will train young scientists and engineers to understand language diversity by combining the tools of behavioral, computational and biological research, drawing upon an extensive collaborative network that spans nine departments in five colleges at the University of Maryland. The project aims to promote sustainable change in the science of language by building infrastructure for interdisciplinary research on diverse languages through local and international collaborations and outreach efforts, by strengthening links between basic science and clinical and engineering applications, and by building awareness of the science of language through high school and undergraduate partnerships. The training plan provides coursework, research training, and an environment geared towards preparing students for interdisciplinary research and equipping them to build collaborative networks in their future careers. Preparation for interdisciplinary research will be provided by broad coursework, integrative pro-seminars and a post-candidacy lab rotation that will pair trainees with students from other disciplines. A central component of the project is the Winter Storm, an intensive two-week workshop that will provide foundational skills training, research planning, and professional development. The project will enhance the use of computational and neuroscientific techniques in studies of atypical language and second language learning, and will partner with an NSF-supported Science of Learning Center based at Gallaudet University that focuses on visual language. IGERT is an NSF-wide program intended to meet the challenges of educating U.S. Ph.D. scientists and engineers with the interdisciplinary background, deep knowledge in a chosen discipline, and the technical, professional, and personal skills needed for the career demands of the future. The program is intended to catalyze a cultural change in graduate education by establishing innovative new models for graduate education and training in a fertile environment for collaborative research that transcends traditional disciplinary boundaries.

This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5).

Studying activity of the human brain non-invasively is a major scientific challenge, yet it is essential for enhancing our understanding of the neural bases of action, emotion, and thought. A major technological advancement in studying the neural basis of behavior has been the development of functional magnetic resonance imaging (fMRI), a hemodynamic technique based on the tight coupling between neuronal activity and oxygenated blood flow. fMRI is a powerful tool for non-invasively measuring local changes in the brain with high spatial resolution (~1 mm) in the blood oxygen level dependent (BOLD) signal. Additionally, structural imaging using MRI can characterize volumetric differences in brain tissue and specify major pathways of neural processing and transmission. These approaches can be combined with other neuroimaging data examining the temporal dynamics of brain activity, to establish a more complete understanding of the human brain and the neural processes underlying human cognition, action, and emotion.

A state-of-the-art 3-Tesla Magnetic Resonance Imaging (MRI) scanner will provide access to this powerful technology to the University of Maryland College Park community for studying human brain activity. The MRI scanner will serve as the centerpiece of the Brain Imaging Center at Maryland (BICAM) and will transform the research and educational environment at the University of Maryland. The scanner will provide the foundation for research in cognitive and affective neuroscience, with specific foci on human development, attention and memory, decision making and risk, motor-control, and language and communication. The center will also create opportunities for innovations in signal processing and magnetic resonance physics. The center and its shared instrumentation will foster an intensive learning environment through the integration of research and education within the University of Maryland and through its partnerships in the local community. The MRI scanner will enhance graduate and undergraduate education through directed research projects, courses with a hands-on focus in functional neuroimaging, and accessibility to students from underrepresented groups. The center will also sponsor a summer institute in developmental cognitive neuroscience which will bring experts in the study of brain development and neuroimaging to the University of Maryland.

BICAM is part of the Neuroscience and Cognitive Science (NACS) program at the University of Maryland. This program consists of faculty from traditional behavioral and neuroscience departments such as Psychology, Human Development, Linguistics, Hearing and Speech, and Kinesiology, as well as faculty from Computer Science, Physics, Applied Mathematics, and Electrical and Computer Engineering with expertise in imaging, signal processing, and the physical basis of magnetic resonance technology. Acquisition of the new scanner will lead to broad interdisciplinary collaboration in areas of the basic physical and behavioral sciences with the goal of understanding the neural bases of behavior.

This project aims to understand how speakers encode and navigate structured mental representations, through a series of studies of how speakers' knowledge of grammatical constraints guides online interpretation processes. Previous research has found that the real-time use of grammatical constraints shows a remarkably uneven profile of successes and failures. Comprehenders show impressive facility in deploying some rather complex grammatical constraints, but they show striking inaccuracy in other cases that one might expect to be rather easy. This "selective fallibility" profile is unexpected under current theoretical models, and it presents an opportunity for new insights into how speakers store and manipulate sentence structures in memory. The project explores the hypothesis that the uneven profile reflects the different contributions of prospective and retrospective memory search processes. The project contrasts the predictions of two approaches to linguistic memory retrieval: one based upon structure-guided search and another based upon parallel cue-based retrieval in content-addressable memory. The project consists of six sets of studies on constraints such as subject-verb agreement, forwards and backwards coreference relations, bound variable anaphora, negative polarity item (NPI) licensing, and null subject licensing. The experiments use four experimental techniques (self-paced reading, event-related brain potentials (ERPs), speeded grammaticality judgments, and eye-tracking during reading). The project will gather data on the processing of English, Spanish, Hindi, and Brazilian Portuguese. It will also support a new collaboration with researchers in Brazil.

The project addresses a basic challenge to the integration of linguistics, psycholinguistics, and the cognitive neuroscience of language. It has been common to assume a clear division of labor between the mechanisms that speakers use to survive the pressures of real-time language processing, and another set of mechanisms that capture speakers' linguistic acceptability judgments. However, if it is possible to understand linguistic computation as real-time computation, then it becomes feasible to integrate the concerns of linguistics, psycholinguistics and cognitive neuroscience. The project supports broad-based training of students, including members of underrepresented minorities and undergraduate and high school students. The project will also create new international partnerships with teams of researchers in the UK, Germany, and Brazil, and will help to develop infrastructure for experimental research on Portuguese and the indigenous languages of Brazil. More broadly, by elucidating the detailed mechanisms used to encode and navigate structured representations in memory in healthy adults, the project will contribute to an understanding of normal and atypical language learning and age-related difficulties in language processing.

Language learners must identify linguistic properties that differ across languages in the language input that surrounds them, and much recent research has explored the potential importance of distributional regularities in the language input for successful learning. However, other recent findings on child sentence understanding have shown that children's immature language comprehension system is prone to mis-parsing of the input. This raises the possibility that informative distributional information might be missed by the learner: if a child misanalyzes sentences in the input, then the true input distribution from the perspective of adults and researchers may be different from the 'intake', i.e., the effective input distribution that feeds into the language learning mechanism. This project investigates this issue through studies of incremental sentence parsing and reanalysis in question constructions in English and Japanese.

Under the direction of Dr. Colin Phillips and Dr. Jeffrey Lidz, Mr. Akira Omaki will conduct studies using eye-tracking, question-after-story and truth value judgment measures in English and Japanese in order to assess a) whether children, like adults, make early commitments to the interpretation of questions ('active dependency processing'), and b) whether children are able to successfully reanalyze in cases where their initial interpretation turns out to be incorrect. The experimental findings will be supplemented with a corpus analysis and a computational modeling study will be combined with the experimental findings, in order to generate an estimate of how the distribution of wh-question constructions appears from a child's perspective. A novel feature of the project is that it combines experimental and corpus-based approaches to gain an understanding of how children's language comprehension system might lead them to apprehend distributional regularities that do not correspond to what is actually present in their input. Thus, one broader impact of the project is that it could have important implications for any research that emphasizes the role of distributional regularities in the language acquisition process. A second broader impact of the project is that it will help to establish new partnerships for language acquisition research on Japanese, based on developing institutional connections with preschools, and will facilitate future international collaborations on comparative language acquisition studies.

Most infants around the world, and many in the USA, are raised learning two languages. Recent research suggests that growing up bilingual leads to cognitive benefits. Bilinguals show advantages during tasks that rely on short-term memory (involved in the temporary holding of information) and attention (necessary for selecting and inhibiting different forms of information). However, bilinguals' language and vocabulary development differs from that of monolinguals, and as a result they are sometimes misidentified as having language disorders. Most studies have examined ways in which bilinguals are better or worse than monolinguals. However, it is possible that bilinguals simply approach tasks differently, which leads to advantages in some tasks and disadvantages in others.

The current research examines whether differences in language exposure (monolingual vs. bilingual) and age influence individuals' ability to learn and comprehend words in in difficult listening conditions (e.g., when speech is heard in the presence of background noise). Listeners of different ages often find themselves in noisy settings where they are spoken to in the context of competing sound. Hence, examining language processing under these circumstances has great relevance to everyday life. The project includes 4 experiments with infants and adults that rely on behavioral measures (e.g., eye-tracking, verbal responses). It examines monolinguals' and bilinguals' ability to focus their attention on a particular sound signal (i.e., the target speech), while ignoring a competing signal (i.e., the background noise) during two tasks that are extremely relevant for becoming a proficient language user (learning and comprehension).

Given the rapidly growing rate of bilingualism worldwide, understanding whether or not human faculties such as cognition and language develop differently depending on the number of languages in the environment is of great importance. Discoveries from this project will provide evidence for generating more accurate tools for diagnosing language delays, developing teaching approaches, and developing recommendations for policies related to language, caregiving, and education. This project will also support the scientific training of a promising scholar.

Language unfolds rapidly in time, and is often contaminated by various kinds of noise. Despite these challenges, native speakers? comprehension is generally robust and efficient. A key cognitive mechanism that underlies the robustness and efficiency of language understanding is the ability to predict the future based on knowledge acquired from past experience, i.e., memory. Much recent research has shown that comprehenders are very good at predicting upcoming words. However, little is known about exactly how comprehenders generate predictions in real time, as they hear or read sentences. In order to understand predictive mechanisms, we need to understand the nature of the memory systems that are engaged, the mechanisms that guide memory access, and how those mechanisms relate to the linguistic information that provides the memory cues. Understanding these mechanisms is important for figuring out how human languages are processed in the brain, and it may serve as a foundation for understanding how human language processing mechanisms can be impaired or repaired (in clinical settings), trained (in educational settings), and simulated (in technological settings).

This research aims to examine the mechanisms underlying word prediction by studying native English speakers (in the US) and native Japanese speakers (in Japan) using electroencephalography (EEG). The project will support the doctoral dissertation research of Shota Momma. The studies build on previous research that shows that some linguistic predictions are computed rapidly while others are computed more slowly. The guiding hypothesis for the current project is that the speed of linguistic predictions depends on the compatibility between linguistic cues and the format of memory encoding. Linguistic cues that mismatch the format of memory are slower to compute. The project tests this hypothesis via experiments that vary the nature of the linguistic cues. The Japanese studies will be carried out with partners at Waseda University in Tokyo, where they will contribute to a growing scientific partnership between the American and Japanese laboratories.

Language learning, in humans and machines, has far-reaching relevance to global technology, commerce, education, health, and national security. This National Science Foundation Research Traineeship (NRT) award prepares doctoral students at the University of Maryland, College Park with tools to advance language technology and language learning. The program provides trainees with an interdisciplinary understanding of learning models from cross-training in linguistics, computer science, and psychological and neural sciences, and with the tools to work with multi-scale language data. The training program contributes to the public understanding of science through a policy internship program that engages trainees with federal agencies and Washington-area professional organizations. Moreover, by contributing to the development of a free public digital linguistic tool, Langscape, it will provide a valuable resource for researchers, the public, the government, and nongovernmental agencies to discover geographical and linguistic information about languages of the world.

Flexible and efficient language learning, in humans and machines, is the research focus of this NRT program. The research hypothesis is that improvements in learning in machines and in humans will come from the ability to use training data more efficiently at multiple scales. Through interdisciplinary team approaches, trainees will explore efficient use of language data, with a focus on the informativity of data to human and machine learning. Through a suite of training activities that includes intensive summer research workshops, engagement with undergraduates and K-12 schools, and policy internships, trainees will become flexible communicators in writing and speaking and also learn to apply their research to diverse contexts.

Mountain rivers play a distinctive role as the primary agents of distributing sediment and nutrients from mountains to lowlands, as critical freshwater source areas for the Western United States, and as vital aquatic habitats. However, mountain rivers are highly susceptible to the compound hazards presented by global climate change and shifting precipitation patterns. Increasing rainfall intensity can result in more frequent flooding and landsliding while increasing drought severity enhances the threat of wildfire, vegetation loss, and extreme erosion risks. This project aims to develop a physical model to assess the sensitivity of mountain river channels to these shifts in climate and their associated hazards. This research will help to identify where river channels are susceptible to significant change and guide management and engineering practices for mitigation and sustainable restoration. Project results will directly support underserved and minority high school and undergraduates with hands-on engaging learning and research STEM experiences at project Universities.<br/><br/>This research aims to develop an understanding of how mountain river channel geometry dynamically responds to flooding and other watershed perturbations. The research is organized around the central question of separating river response due to a perturbation from the inherent natural variability present within watersheds. Separating signals of change from variability involves three components: a physics-based model for the expected river conditions under natural forcing, quantification of the inherent natural variability within the river channel system across the riverbed and reach scale, and a physical description with quantified adjustment times for river response to perturbations across these scales. This research will accomplish all three components by: (1) leveraging high resolution lidar topography and sediment transport data to establish a baseline level of variability and the dependence on the scale of variability on climatic and geologic factors; (2) developing a physical link between hydraulic perturbations, river planform instability, and the threshold processes underpinning sediment transport through high resolution laboratory experiments; and (3) pilot a signal-to-noise framework at sites where documented perturbations have both destabilized the system or appear to have been buffered by river processes. Expected results will aid researchers and engineers in determining which river systems are vulnerable to erosion due changing climate and landscapes, and provide a foundation for treating rivers dynamically within the next-generation of river flood hazard forecasting models.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.