Hierarchical control of sequential skills: Using EEG to decode the underlying representations

University Of Oregon Eugene, Eugene OR

Investigators

Abstract

Many of the most interesting human activities result from basic building blocks of behavior being structured into appropriate sequences. For example, speaking involves the combination of individual syllables into words and sentences; while, performing music involves the combination of notes into melody. The mind must represent these activities in terms of the hierarchical structures that combine the basic elements. Deficits in the ability to implement such structures, due to neurological disorders or aging, can have wide-ranging consequences for real-world functioning. Little is currently known about the mental representations that enable these sequential skills, or about how deficits in fluent sequential performance arise from disruptions to these representations. This project develops new methods for tracking these mental representations using electrophysiological signals recorded at the scalp (EEG). These methods will allow the pinpointing of the exact sources of individual and age-related differences and provide initial hints for the origin of deficits due to disease or brain insults. The project provides training and educational opportunities in sophisticated brain-imaging and data-analytic techniques for undergraduate and graduate students and contains a science-education component for underprivileged high-school students. Cognitive scientists have long assumed that the critical, hierarchically-organized structures that support sequenced behaviors rely on a special set of mental representations that organize complex sequences into smaller parts (often referred to as chunks), or that mark the position within a chunk. Yet, because of their abstract nature, it has been very difficult to empirically track these hidden representations. The central hypothesis tested in this project is that rhythmic, electrophysiological signals contain robust information about how strongly these mental representations are activated at a given point in time. The planned experiments will examine the validity and robustness of new, EEG-based methods of tracking mental representations. Individual experiments will determine how exactly positions or chunks are represented and will test specific hypotheses about how individual or adult, age-related differences in cognitive resources (e.g., working-memory capacity) affect sequential representations. This project will equip researchers with refined theories about how sequential information is processed, and more generally with a novel set of analytic tools to characterize complex, cognitive performance.