UROP Openings

Have a UROP opening you would like to submit?

Please fill out the form.

Submit your UROP opening

AttentivU: wearable systems for measuring physiological signals




MAS: Media Arts and Sciences

Faculty Supervisor:

Pattie Maes

Faculty email:


Apply by:

February 1, 2021


Nataliya Kosmyna, nkosmyna@media.mit.edu

Project Description

While today's pervasive digital devices put the world’s information at our fingertips, they do not help us with some of the cognitive skills that are arguably more important to leading a successful and fulfilling life, such as attention, memory, motivation, creativity, mindful behavior, and emotion regulation. Building upon insights from psychology and neuroscience, the Fluid Interfaces group creates systems and interfaces for cognitive enhancement. Our designs enhance cognitive ability by teaching users to exploit and develop the untapped powers of their minds and by seamlessly supplementing users' natural cognitive abilities. The solutions we design are compact and wearable to support real-life studies and interventions, rather than for use in laboratory settings. Our work is highly interdisciplinary and combines insights and methods from human computer interaction, body sensor technologies, machine learning, brain computer interfaces, psychology, and neuroscience to create new opportunities for studying and intervening in human psychology in-the-wild. One of the projects we aim to find two students for this IAP/Spring term to help us with is AttentivU. AttentivU addresses a problem attention, fatigue, cognitive load and engagement: it is increasingly hard for adults and children alike to be attentive given the increasing amounts of information and distractions surrounding us, thus we developed a device, in a socially acceptable form factor of a pair of glasses, that a person can put on in moments when he/she wants/needs to be attentive. The AttentivU glasses use brain activity (electroencephalography - EEG) as well as eye movements (electrooculography - EOG) sensors to measure engagement, attention, cognitive load, fatigue of a person in real-time. We have tested the first generation prototype of the device in workplace and classroom settings with over 100 subjects. We have performed experiments with people studying or working by themselves, viewing online lectures as well as listening to classroom lectures. We have now finished the first tests of the glasses (second prototype) with more than 30 subjects who were performing driving task in the simulator and falling asleep while driving. To help us investigate the usability of the project in different scenarios (driving, learning, communication for users with health challenges like CP or ALS) and with different populations (adults and children), we are looking for one or two students during this IAP/spring term to mainly: (option 1) improve our current ML models or (option 2) help improve the app for the glasses Participation of a minimum of 10 hours per week during the school year is preferable, but there is flexibility. This UROP may be taken for pay (direct funding), credit, or volunteer. Relevant project URL: https://www.media.mit.edu/projects/attentivu/overview/


Depending if the interest is in option 1 or 2: If interested in Option 1: one or more of the following: Experience in Artificial Intelligence using programming languages such as Python and platforms such as MXNet, TensorFlow, Caffe, Torch, etc. If interested in Option 2: One or more of the following: Languages: Swift, Java, Javascript, Objective C, Python, C#; Tools/Frameworks: React Native, React.js, Node.js, Java RESTful API, Azure Speech SDK, Docker, AWS (S3, RDS, Elastic Beanstalk, ECS), Mediapipe CV Pipeline, Git
 Design: Adobe Illustrator, Figma, Human-centered design, hypothesis-driven testing, UI/UX design