Connecting Visual Perception, Attention, and Probabilistic Models of Task Performance with EEG Measured Brain Activity

• Author / Creator

  Sheldon, Sarah S

• The neural mechanisms underlying visual perception and attention continue to elude researchers despite decades of research. Developing novel methodology and improved analytical techniques may provide key insights into these processes that traditional approaches have been unable to reveal. In this dissertation, we pursue this idea in a series of studies whose overall aim is to better understand visual perception, attention, and their underlying neural mechanisms. First, we demonstrate how the novel adaptation of visual working memory probabilistic models can turn simple performance measures into metrics that quantify the quality of participants’ internal perceptual representations. When paired with EEG analysis, we find evidence that perceptual representations will vary only within a fixed range of values, but where in that range its precision falls changes from trial-to-trial as a function of post-stimulus neural activity. Next, we extend this approach to test the effects of covert attention modulation while simultaneously questioning previous assumptions about the role of periodic oscillatory activity, particularly in the alpha (8-14 Hz) frequency, during the cued version of the orientation perception task. From this novel combination, we find evidence that the conflicting reports on the role of alpha oscillations in visual perception and attention are, at least partially, due to measures being confounded by the overlapping and task-related 1/f aperiodic activity. Finally, we used multivariate pattern analysis (MVPA) to address a long-standing question regarding how alpha-related brain activity prior to the presentation of a visual stimulus corresponds to attention, perception, and subsequent task performance. Our results suggest that it is the complex spatiotemporal dynamics of alpha amplitude that best represents what makes trials with covert spatial attention distinguishable from trials without. However, we found little evidence that those same spatiotemporal patterns of activity are predictive of subsequent behavioral responses. Overall, this series of studies demonstrate the value of using novel techniques that can take full advantage of the inherent multidimensionality of EEG data as well as highlights the opportunities these methods present for future research.

• Subjects / Keywords

  • Attention
  • Visual Perception
  • EEG
  • Standard Mixture Model

• Graduation date

  Fall 2022

• Type of Item

  Thesis

• Degree

  Doctor of Philosophy

• DOI

  https://doi.org/10.7939/r3-ykwf-wb80

• License

  This thesis is made available by the University of Alberta Library with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.