As we read and hear sentences, our intact brains quickly settle on the intended interpretation-this despite the fact that (a) individual words generally have multiple dictionary meanings (baseball vs lemonade pitcher, adopt a child vs a stance), and (b) the words alone sometimes omit details crucial for sentence interpretation (begin the book, reading? writing?). This is a long-standing problem with no consensus on an adequate, biologically realistic, theory of sentence interpretation, based on the available data. There are many theoretical alternatives, each accounting for certain phenomena of meaning indeterminacy (too many or unstated word meanings or senses). Thus, we believe what is needed for theoretical progress are new- and new kinds of-brain data that characterize in greater depth and breadth what specific content of the interpretation is computed and with what types of information (e.g., gender stereotypes, event knowledge), and when. This is now possible via systematic investigations of real-time sentence interpretation across a wider-than-previously-examined range of meaning indeterminacy, using temporally-precise measures of neural activity in intact, dynamically interpreting brains (both halves). Our experiments use brainwaves- event-related brain potentials (ERPs)-to track relevant brain processes with millisecond resolution, in both proven and novel experimental paradigms, to investigate a broad spectrum of meaning indeterminacy (homonyms, polysemes, metonyms), including words such as shoe and shoot (coined hermenesemes) which seem unambiguous in context but may be interpreted differently, e.g., his vs her shoe and shoot with a rifle vs camera. Remarkably few studies have used brain waves to these ends, much less to figure out how human brains handle the wide range of language input rates that boggle current voice recognition software. We thus test proposals that neural oscillations adjust to word input rates and thereby impact the speed and efficiency of interpretative processes. All told, some of our proposed experiments test current alternative theories. Others, by necessity, are systematic explorations designed to address critical questions, pushing beyond the boundaries of current debate, an approach which past work shows can lead to the discovery of new phenomena and drive theory in unforeseen directions. Ultimately, we aim to answer these open questions in enough detail to be able to understand, assess, and remediate language comprehension breakdowns in developing and compromised brains across the lifespan. Understanding how the brain's two halves cope with the meaning indeterminacy of words arriving at variable rates is also likely to provide insight int the design of computer algorithms for assistive technologies for communicative and language translation purposes. In sum this project investigates sentence comprehension in the brains of healthy young adults in order to lay critical groundwork necessary for advancing both theory and therapy.
We all use language routinely for multiple purposes - to communicate, educate, inquire, negotiate, and help, to name but a few - yet we still have much to discover about how our two half-brains use what we know to construct interpretations from words that often have many possible meanings or senses, sometimes unstated, so quickly and effortlessly. Understanding how human brain's (neural systems) flexibly and successfully adjust to variations in rate, meaning, and voices in dynamically changing contexts will have untold benefits for typical and compromised language users in all walks of life. With 'normal' electrical brain activity patterns in hand, we can better assess possible 'abnormal' patterns (e.g., in children, SLI, schizophrenia, brain damage) and learn what assessments/interventions may best test/benefit developing and/or compromised brains.
Showing the most recent 10 out of 118 publications
