Liina Pylkkanen - US grants

Although our understanding of the brain bases of semantic interpretation is beginning to be detailed at the level of individual lexical items, the neural mechanisms by which complex meanings are composed are still entirely unknown. With National Science Foundation support, Dr. Pylkkanen will conduct a three-year investigation aimed at elucidating the spatio-temporal dynamics of complex meaning composition. Brain activity will be measured with magnetoencephalography (MEG), which offers the best combination of spatial and temporal resolution of currently available cognitive neuroscience methods. One of the most fascinating features of human language is that many meanings are not expressed overtly. For example, when a sentence such as "the boy shut the door tight" is encountered, all English speakers immediately and automatically know that this sentence describes a causal relation between an event of shutting and a state of tightness even though no overt element in the sentence describes causality. Further, all speakers converge on the interpretation that the state of tightness does not hold of the explicitly mentioned door, but rather of some unmentioned closure between the door and its frame. This project focuses on the interpretation of covert meaning, as it offers a unique opportunity to study semantic composition while largely eliminating phonological and syntactic processing. The project covers multiple levels of representational complexity and involves three sets of studies: one investigating covert complexity in individual lexical items that appear to be simple, another on the effect of semantic opacity in the processing of overtly complex words such as "sweat-er", and a third on covert meaning composition at the sentence level. The project represents a novel type of combination of cognitive neuroscience methodology with detailed representational hypotheses emerging from theoretical linguistics, reflecting Dr. Pylkkanen's combined training in the two fields. An integral part of this research is bringing together a diverse group of students and faculty from linguistics, psychology and medical and/or neuroscience backgrounds. Finally, the clinical applications of the project are potentially substantial. Semantic impairment characterizes a vast array of neurological disorders, including aphasias, schizophrenia, developmental disorders such as autism, and neurodegenerative diseases. A prerequisite for understanding semantic malfunction is first understanding how the healthy brain accesses and computes meaning.

The advent of non-invasive brain imaging at high spatial and temporal resolution in awake, behaving human beings has had a profound impact on the study of language in the brain. It is now the ten-year anniversary of the first international conference on the Neurobiology of Language, a field that has grown substantially since then and continues to blossom. Studying the neurobiology of language requires highly specialized skills (e.g., brain imaging) and broad multidisciplinary knowledge (e.g., psychology, linguistics, neuroscience). In order to promote scientific inquiry of the highest quality, it is vital to promote dialogue and interaction among the relevant disciplines. This project explores how to facilitate this interaction.

The "neurobiology of language" is the biological implementation for representations and processes involved in the production and understanding of speech, sign, and language in context. Previously, brain structures and functions for language could only be inferred from studying humans with brain injury. By monitoring the anatomical encodings and physiological dynamics of language processing in awake, behaving human beings, researchers can now study human language at a biological precision previously possible only in animal models (of functions other than language). Concomitant with these new approaches has been a dramatic increase in the number of investigators dedicating their effort to understanding the biological basis of language. The growth in the number of investigators and the use of expanded techniques and theoretical perspectives have driven the need for enhanced transparency, integrity, and robust and rapid research dissemination. The PIs will explore these topics while developing systems and operations for open science, facilitating reproducibility and replicability of research, and dissemination of new findings.

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.

The proliferation of noninvasive methods to study brain function has revolutionized the neuroscience of language, a uniquely human ability that cannot be studied in animals. So far, most of this research has focused on the adult's brain, as opposed to the child's. Given our poor understanding of the neurobiology of language development, we lack an important background for understanding children's language disorders and for planning surgical treatments for children with neurological disorders such as epilepsy, since such surgeries need to spare language cortex. This project uses magnetoencephalography (MEG) to establish this critical background. The focus will be on two basic processes that form the foundation of human language: word retrieval and the composition of words into larger phrases. Elementary school children from New York City public schools will be invited to participate via science outreach activities in the schools. The data will be entered into a public database that will provide a resource for research and education. The experimental protocol is a combination of a tightly controlled design, built upon prior research on adults, and a naturalistic stimulus, containing fun science-themed narratives. Data from the naturalistic stimulus will offer vast analysis possibilities. To assure that the possibilities are maximally leveraged by researchers, these data will be released to the public on a rapid timescale, at the end each project year. This work will offer research opportunities for graduate and undergraduate students. The project is also a significant STEM opportunity for the participating children, as during the lab visit, the experimenter will explain the basics of MEG and how one can capture tiny magnetic fields inside human brains. Overall, the project aims to transform the neuroscience of language in children by providing the basic science that is needed for more detailed understanding to emerge in the future.

The specific scientific objective of this project is to characterize the spatiotemporal profile of lexical retrieval and phrasal composition in 1st, 3rd, and 5th graders, with their parents serving as the adult controls. The experimental protocol offers (i) the first systematic spatio-temporal characterization of children's neural processing of single words and phrases, as compared to the activity profiles of the children's parents; (ii) a tightly controlled within-subjects and within-items comparison of the neural circuits involved in language comprehension vs. production in children and adults and (iii) a measurement of a simple form of dialogue, engaging both the comprehension and production of words and phrases within a brief trial. An important outcome of the research is a database, from which users will be able to download activity time-courses for specific brain regions during the three experimental tasks. Study participants will also listen to a naturalistic stimulus, with a fun science themed content designed for children. The naturalistic stimulus will allow assessing whether results from the controlled laboratory tasks scale up to a more ecologically valid setting. All MEG recordings from the project will be made publicly available. The project is a large-scale endeavor to engage a diverse sample of New York City children in a neuroscience activity that will significantly advance our understanding of how language is processed in the brains of elementary-school children.

This project is co-funded by the Discovery Research preK-12 program (DRK-12) that seeks to significantly enhance the learning and teaching of science, technology, engineering and mathematics (STEM) by preK-12 students and teachers, through research and development of innovative resources, models and tools.

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.