A Collaborative Pilot Study with Algoma University for Phoneme Detection Using EEG Auditory Evoked Potentials

---

Project Lifespan: 2015

The main purpose of the study was to create a foundation for future work in the field of BCI, specifically in the language decoding areas. I used the typical AEP paradigm. However, I did not follow the typical P300 analysis in which analysis is made 300ms after the stimulus. Rather I looked at the time at which the stimulus was applied and when the brain reacted to it. The analysis was not for classification purposes but instead to try and understand the underlying cause or network pattern for when a person is thinking or trying to identify a heard phoneme. To achieve this I used the algorithm I helped develop at the Fields Institute. I observed clustering of electrical activity using the “clustering algorithm” in various brain regions during this short epoch. The results were inconclusive as some patients had very high source localized activity in the right uncus migrating to the right inferior temporal gyrus, while others had a combination of the precentral, postcentral, and superior parietal lobule.