The States and Traits of EEG Variability

The States and Traits of EEG VariabilityErin Gibson Doctor of Philosophy Institute of Medical Science University of Toronto 2021 Neurons in the brain are seldom perfectly quiet. They continually receive input and generate output, resulting in noisy, highly variable patterns of ongoing activity. Yet the functional significance and behavioral consequences of this variability remains largely unknown. We hypothesize that brain signal variability is tightly coupled to the processing of task-relevant information and serves as an important indicator of cognitive function. To test this, we examine EEG activity in young, healthy adults as they perform a cognitive skill learning and resting state task. Several measures of EEG variability and signal strength are calculated at multiple timescales, or in overlapping time windows that span the trial interval. We perform a systematic examination of the factors that most strongly influence the variability and strength of EEG activity. Study 1 examines the relative sensitivity of each measure to trait-level variation across subjects and state-level variation across task blocks. We find that EEG variability is most sensitive to trait-level differences across individuals that remain relatively stable across blocks. Study 2 examines the sensitivity of each measure to different sources of state-level variation across task blocks. We find that key task-driven changes in EEG activity are better reflected in the strength, rather than the variability, of EEG activity. Study 3 examines trait-level variation in each measure and its relationship with behavior. We find that differences in learning speed and acquired skill across individuals are better reflected in relative changes in the variability, rather than the strength, of EEG activity. These results demonstrate that EEG variability is particularly sensitive to stable trait-level variation across individuals and relative changes in EEG variability around a particular level, rather than the level itself, are a strong indicator of behavioral performance in young, healthy adults.Ph.D