PhD Thesis Proposal: Jenny Qiu

Wednesday, March 14, 2018, 2:00–4:00pm

Jackson Conference Room, Cummings Hall

“Learning State Estimation with a Brain Computer Interface”


Brain-Computer Interfaces (BCI), which are platform systems that utilize one or more neuroimaging methods to translate central nervous system impulses into a response, have found applications in rehabilitation and assistive technologies as well as brain-performance augmentations. Although several BCIs, such as P300 spellers and neural prosthetics, have been successfully implemented, these BCIs rely on the binary existence of neural signals such as ERP. However, as humans have complex neural signals that are influenced by both internal and external factors and by time, developing BCIs should not only determine whether the signal exists but also consider how the signal evolves.

This proposed research will focus on the following BCIs aspects in complex systems: 1. The extraction of baseline neural signatures resulting from passive cognitive tasks such as symbol recognition and those signatures resulting from active engagement with a complex learning task from a designed user interface feedback system, 2. Identifying learning states of a subject using subject behavior and neural data, and 3. Incorporating both 1 and 2 into a classifier for a real-time BCI design that utilizes results from neural engagement to classify complex neural behaviors and perform tasks. This research will benefit the growing BCI industry that requires translating neural signals through complex cognitive tasks by incorporating the human element back into the control loop. This research is especially beneficial to patients who require assistance with learning and to e-learning companies who wish to evaluate effectiveness through a neural perspective.

Thesis Committee

For more information, contact Daryl Laware at