Lessons Learned from Using Physiological Sensing in Learning Environments

Nataliya Kosmyna
February 13, 3-4 pm
MIT Campus
MIT Community

Knowledge work has become increasingly complex and demanding.

Given the magnitude of information surrounding us, our never-ending connection to the Internet and the constant shifting between increasingly complex tasks, it is no surprise many of us suffer from low attention span, engagement issues, cognitive overload and increased fatigue. All these may negatively affect our performance on cognitively demanding tasks. Despite several recent research efforts which investigate the measurement of attention, cognitive load, fatigue and engagement, few of these projects have made it to the real world.

In this talk, Dr Nataliya Kosmyna will present a novel platform to help tackle these problems in real-world scenario of learning environment. The platform consists of an improved version of an existing smart glasses research prototype called AttentivU, which measures physiological data of a person: their brain activity using Electroencephalography (EEG) and their eye movements using Electrooculography (EOG).

In her talk Kosmyna will share the results of the interventions she has performed using the glasses; challenges of such projects; ethical concerns related to using physiological data; and possible future use cases of technology involving physiological monitoring like brain signals in the learning environments and scenarios.

Nataliya Kosmyna is a Post-Doctoral Associate in the MIT Media Lab, helping MITili evaluate the real-time biofeedback of learners to monitor and improve their ability to sustain attention. Most of her projects are focused around EEG-based BCIs in the context of consumer grade applications.  In 2016 Nataliya also created her first start-up, Braini, for consulting in the domains of Artificial Intelligence for Cognitive Enhancement as well as Neuroscience. She has also participated in two TEDx talks. 


Read MIT student Melissa Cao's reflections on Dr Kosmyna's xTalk.