MIT Open Learning
Published in

MIT Open Learning

MITili research grant showcase

Researchers across MIT share projects and results from investigations into how we learn

A robot navigates a cell
Gameplay of Cellverse by MIT Education Arcade

By Katherine Ouellette

The MIT Integrated Learning Initiative (MITili) funds, supports, and shares research into learning effectiveness at all ages, levels, and scales. As part of this effort, MITili provides research grants to MIT faculty and research teams investigating myriad aspects of brain and cognitive science, digital and augmented technologies, and learning approaches. This fall MITili hosted a showcase of new and ongoing research resulting from the 2018 and 2019 grant cycles, hosted by associate director of MITili, Jeff Dieffenbach, along with MITili faculty directors Prof. John Gabrieli and Prof. Parag Pathak.

The presentations concluded with an announcement from Prof. Gabrieli that MITili would be able to extend a new round of funding in February 2022. Check the MITili website in January 2022 for more details.

Converting zombies into learners: Improving MOOC learner retention
Eva Ponce, Inma Borrella, Sergio Caballero, and Chris Caplice (MIT Center for Transportation and Logistics)

Of 16,000 learners who register for a given Massive Open Online Course (MOOC), 1,500 will continue on the verified track, 1,200 will complete the course, and only 1,000 will pass the course. The team hoped to predict which learners were likely to drop out, and then identify which interventions could effectively reduce the potential dropout rate. Their predictive model was able to identify 3 out of 4 dropouts using random forest and logistics regressions algorithms, with “grades” and “time spent in course” as the most relevant predictors from the raw data. They then tested four methods of interventions based on increasing “support” (e.g. motivational emails and exam preparation materials) and modifying “content” (e.g. gradual increase of difficulty in tests and improvement of challenging content). Content interventions reduced dropout rate, while support interventions had no effect on the dropout rate.

Integrating NYC schools: the role of admission criteria and family preferences
Clémence Idoux and Prof. Joshua Angrist (Blueprint Labs, MIT Department of Economics)

As of 2018, New York City schools were among the most segregated in the US in terms of race and income. This segregation persists because of “school side” factors (e.g. admission criteria based on academic performance and home address) and “family side factors” (i.e. preference for nearby schools; homily). Angrist and Idoux studied two different school districts that adopted admission reforms in 2019 that reduced the role of academic criteria screening: the Upper West side reserved 25% of seats for low income families and Northwest Brooklyn eliminated academic credentials altogether, and both policies reduced racial and economic school segregation. Disadvantaged students applied to more competitive schools in response to policy changes. To learn more about Idoux and Angrist’s research, watch the video Open Learning Talks: Educational policy’s impact on diversity, accessibility, and achievement.

Impact of infusing computation and visualization into introductory physics subjects
Kyle Keane (MIT Quest for Intelligence); Anna Musser, (MIT Sinha Lab, Department of Brain and Cognitive Science); Lauren Berk (MIT Operations Research Center, Sloan); Michelle Tomasik (Online Education, MIT Department of Physics); and Andrea Griffin (Online Education, MIT Department of Materials Science and Engineering).

This experimental design was broken up into six phases: a pre-assessment, interventions each week for four weeks, and a post-assessment. The lessons explained core concepts using programming, visuals (i.e. graphs), non-visuals (e.g. charts), and further questions (not graded). The experiment did not produce statistically significant results, but the team learned a lot about how to conduct experiments for their related ongoing projects.

AttentivU: evaluating the effectiveness of real-time biofeedback to monitor and improve ability to sustain attention
Nataliya Kosmyna and Prof. Pattie Maes (MIT Media Lab, Fluid Interfaces Group)

After 10 full, peer-reviewed publications with users since 2018 — plus several awards and honorable mentions — the prototype of these glasses has proven effective for gathering metrics on attention (auditory, visual, internal, external), engagement, fatigue, drowsiness, and cognitive load of the user. AttentivU gives the user data in real time, and the biofeedback helps the user sustain attention. This has huge implications for users with 12- to 18-year-olds with ADHD, and more.

Adaptive learning of motor skills
Learning of maker skills using digital games
Dishita Girish Turakhia and Prof. Stefanie Mueller (MIT CSAIL)

CSAIL’s research has three components: 1) adaptive learning; 2) game-based learning; and 3) reflective learning. The toolkit they developed, Adapt2Learn, had an auto-adaptive mode powered by a user interface and visualization tool to configure and assess the learning algorithm — which resulted in higher learning gains from the users. Similarly, the FabO fostered adaptive skills for digital/maker skills using video games. This interactivity excited learners about fabrication. The next step in their process will be codesigning a workshop with maker educators they have already interviewed, and run the study with students.

Enhancing learning via “novelty insertion”: employing the neuroscience of learning to create more effective pedagogical approaches
Annie Cardinaux, Matt Groth, Anna Musser, Lara Cavinato, Sidney Diamond, Riccardo Barbieri, and Prof. Pawan Sinha (MIT Department of Brain and Cognitive Sciences)

Prof. Sinha discussed his ongoing research into the plasticity of the brain in children who have corrective vision surgeries, pulling together domains of neural learning and classroom learning. Brain plasticity is commonly thought of as an age-limited resource, with exponential rate of synapse formation in the sensory pathway for vision in the first 11 months. Sinha’s research points to brain plasticity as a resource on demand — e.g., when in novel environments, the brain will enhance its ability to change, and therefore, learn. To test this hypothesis, subjects in the study were presented with both novel videos and familiar videos. Contrary to the hypothesized expectations, the novel video metrics were indistinguishable from familiar videos, but subjects did perform marginally better with novel insertions (presented within eight minutes of video-watching).

Dialogic co-reading interaction for improving child-parent conversational turn-taking
Hae Won Park, Huili Chen, Sharifa Alghowinem, and Prof. Cynthia Breazeal (MIT Media Lab, Personal Robots Group)

From a young age, children don’t learn language through reading and writing, they learn by socially interacting with others. Language learning is cognitive, social, affective, and relational. That, plus John Gabrieli’s group’s work on conversational turn takes, created the foundation for the Personal Robots Group to use social artificial intelligence (AI) agents as an intelligent tutoring system. During story time between parents and children, they facilitated child’s learning, empowered the parent(s), and mediated parent-child conversations, then tracked the effect of story time in everyday contexts. This data set helped them develop predictive algorithms for effective language learning in AI agents. The paper is currently under review.

Cellverse: Cellular biology through virtual reality
Meredith Thompson (MIT Scheller Teacher Education Program and The Education Arcade)

“Cellverse” is a collaborative tablet-based virtual reality game that teaches cell biology. Their research is currently aimed at high school students but there are possibilities for wider audiences. The game is set inside a cell that has a type of cystic fibrosis, and the player needs to figure out what type of cystic fibrosis so they can find the cure. In their studies, it helped students take an abstract concept like cell processes into a hands-on concept. Student feedback said, “You actually have to pay attention to reach your goals and finish the game.”

--

--

--

News, ideas, and thought leadership on the future of learning. Discover more at openlearning.mit.edu.

Recommended from Medium

10 Things To Consider Before Joining Architecture School

Youth Justice: Principles of Effective Practice

BIG SISTER ADVICE: What you ACTUALLY need to know

Five London boroughs join forces to help troubled families stay together

TL42: 89

What is Cognitive Learning? — Benefits, Challenges, Use Cases and More

We are there for your kids

While we may not be physically in the classroom tomorrow, teachers like us are still there for your kids.

The Art of Teaching Inquisitively

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
MIT Open Learning

MIT Open Learning

Transforming teaching and learning at MIT and around the globe through the innovative use of digital technologies.

More from Medium

Q&A with Saif Rayyan, Assistant Director, Higher Education Engagement Units J-WEL

Illustration of three students pulling educational materials out of a laptop screen

What is Gravity?

How to Find and Understand Research Papers

evolution as a mental model