top of page

BUILDING LEARNING TOOLS

By learning tools, we mean two things.​

​

  1. Real-world devices that autonomously adapt to new settings, users, and tasks, by using the data they capture as they work. This makes an important testbed of our interactive learning algorithms.

  2. Tools that facilitate human learning and inquiry, whether that be in education, exploration outside of formal education, or at the forefront of scientific research. Such tools can, in particular, greatly enhance distance learning and collaboration.

ag_how_it_works.png

One form factor is particularly attractive. AI coupled with augmented or virtual reality (XR) allows us to process the user’s surroundings and overlay feedback onto the user’s field of view, all while capturing data about the experience. Thus far, this sort of paradigm has been used in the Autism Glass Project, in which we prototyped and tested an XR aid, meant to be worn by children on the autism spectrum, that recognizes social cues from those around the wearer and provides immediate visual and auditory feedback from which they might learn.

​

  1. Voss, Catalin, Jessey Schwartz, Jena Daniels, Aaron Kline, Nick Haber, Peter Washington, Qandeel Tariq et al. Effect of wearable digital intervention for improving socialization in children with autism spectrum disorder: a randomized clinical trial. JAMA pediatrics 173, no. 5 (2019): 446-454.

  2. Daniels, Jena, Jessey N. Schwartz, Catalin Voss, Nick Haber, Azar Fazel, Aaron Kline, Peter Washington, Carl Feinstein, Terry Winograd, and Dennis P. Wall. Exploratory study examining the at-home feasibility of a wearable tool for social-affective learning in children with autism. NPJ digital medicine 1, no. 1 (2018): 1-10.

  3. Washington, Peter, Catalin Voss, Aaron Kline, Nick Haber, Jena Daniels, Azar Fazel, Titas De, Carl Feinstein, Terry Winograd, and Dennis Wall. SuperpowerGlass: a wearable aid for the at-home therapy of children with autism. Proceedings of the ACM on interactive, mobile, wearable and ubiquitous technologies 1, no. 3 (2017): 1-22.
     

bottom of page