Skip to Content
MIT News feature

How Should We Teach the Teachers?

MIT rethinks education for the 21st century.

Standing in front of several hundred education leaders over coffee and croissants in the ballroom of Cambridge’s Royal Sonesta Hotel, Arthur Levine is gleefully taking apart the modern education system. “Imagine if your GPS worked like testing does today—it would give you a reading every hour,” quips Levine, president of the Woodrow Wilson National Fellowship Foundation, a Princeton, New Jersey–based nonprofit focused on education and leadership development. What’s more, “when you started you’d be 20 miles from where you are going; and now you are 50 miles from where you are going, and heading in the wrong direction.”

Known for his withering critique of teacher training at most education schools, Levine, the former president of Columbia’s Teachers College, is delivering the keynote address at a gathering of top higher-education professionals from around the country. First, he says, we have to admit that the American education system is broken. Built during the industrial age, it evokes the best technology of the era, the assembly line—forcing all students to learn the same thing at the same time.

That system is totally out of step with what students need in a global information society. “The textbook, the old technology we are familiar with, is dying,” he says. In its place, Levine imagines a classroom that takes advantage of virtual reality to teach students about ancient Greece, that can teach students remotely wherever they are in the world, that paces itself to the needs of students and tests them every step of the way to ensure that they’re on track. Given the requirements and capabilities of our modern world, he says, “we are going to see education become more individualized, more customized, and more adaptive.”

Just a few blocks away from where’s he giving the speech, he and his foundation are working with MIT to construct a new model for a graduate school of education to implement his vision. Over the next five years, MIT will serve as the incubator for the Woodrow Wilson Academy of Teaching and Learning, an independent teacher-education school that plans to offer a master’s in education as well as a license to teach a STEM (science, technology, engineering, and math) subject in grades 5 to 12. (The degrees will be granted by the academy, not MIT.) When it opens its doors to teacher candidates in the fall of next year, the academy will use cutting-edge technology and the latest brain research on learning to dramatically transform the way teachers learn—and the way they teach middle school and high school students.

MIT’s partnership with the academy is part of an even more sweeping initiative to improve teaching all the way down to kindergarten. “K–12 education is where we get our undergrads from, but more than that, there is also an ‘MIT way’—a ‘hands-on, minds-on’ way of teaching—that we thought we should get out into the world,” says Sanjay Sarma, MIT’s vice president for open learning. The impetus to share the MIT way sprang from discussions that began with the Institute-wide Task Force on the Future of MIT Education convened by President L. Rafael Reif in 2013. “One of the things that came up was a heartfelt and deep interest in encouraging more STEM education and reaching out to students before they came to MIT,” says Sarma, who cochaired the task force.

In February, President Reif announced two other new initiatives to further the goal of spreading MIT’s version of STEM education: the pK–12 Action Group, which will integrate and expand MIT’s many outreach efforts to pre-­kindergarten through high school learners, and the MIT Integrated Learning Initiative ­(MITili), which will help translate the latest research in such disciplines as psychology and neuroscience directly into educational approaches and encourage more education research at MIT.

Getting kids excited about STEM

While creating an ed school has been proposed in the past, says Sarma, “we’ve never had the faculty numbers or the concerted interest in taking on a large number of graduate students in teaching per se.” On the other hand, the Institute has a long history of researching education and packaging complex science for younger minds. In the 1950s, MIT worked to get the newest science and engineering concepts into children’s textbooks. In the 1960s, a team from MIT led by Seymour Papert developed the Logo programming language to teach math to kids. More recently, Media Lab professor Mitchel Resnick, SM ’88, PhD ’92, created Scratch, a graphics-based program to teach kids to code. Yet another example: Hal Abelson, PhD ’73, co-designed App Inventor, which lets children develop apps for phones and tablets (see “Upward Mobility”). In all, Sarma estimates, there are more than 120 efforts targeting K–12 education under way at MIT.

“But there has never been a major concerted effort to bring it all together,” says Eric Klopfer, a professor whose own research focuses on developing computer games to teach kids STEM topics. In addition to leading the Education Arcade, which develops games, simulations, and technology tools for teachers, Klopfer directs the Scheller Teacher Education Program, a licensing program for undergraduates. “This is the first time where we are really making an effort to come up with ideas and structures for programs that could really help leverage what we are doing across these areas to make an impact,” he says.

Getting kids excited about science is crucial to inspire the next generation of scientific discoveries, says Angela Belcher, director of the pK–12 initiative. Belcher, who is the W. M. Keck Professor of Energy as well as a professor of both biological engineering and materials science and engineering, regularly visits K–12 classrooms. She’s denatured DNA from strawberries and used dodge balls to explain peptide chains and crystal structures. She even teaches kids as young as preschoolers about solids, liquids, and gases: she has them pretend to be gas molecules by running around, gets them to flow like liquid by moving slowly and lightly touching hands, and has them form a lattice with tight bonds to become solids. When she calls out temperature changes, the kids shift their state. “Should first-graders understand hydrogen bonding? Of course they should,” she says. “It’s one of the keys to life. When they jump into a swimming pool, they should think about how amazing all of those electrons and protons are, how hydrogen bonds hold it all together.”

“A lot of kids decide in third or fourth grade whether they want to be a scientist or an engineer,” she continues. Too many decide they are not interested in science before they’ve been exposed to its potential. “What if we get kids excited by encouraging their intellectual curiosity and hands-on learning earlier, by giving them the ability to understand what it means to be a creator?” she says. She thinks it’s possible to get even very young kids excited about big challenges like developing clean energy, or get them to understand the beauty and importance of DNA by having them develop their own tools to study genetic material.

In addition to working directly with students, MIT has been actively involved in training teachers both on and off campus. Klopfer developed a series of edX courses about educational technology and games that has reached about 75,000 people; the Edgerton Center, run by Kim ­Vandiver, professor of mechanical and ocean engineering and dean for undergraduate research, works in Boston schools to distribute science curriculum materials and run workshops for teachers on new ways of teaching about atoms, molecules, and DNA; and the Center for Materials Science and Engineering brings hundreds of Massachusetts high school and middle school teachers to campus every year for an intensive weeklong boot camp in using innovative techniques to teach science to young people.

The pK–12 Action Group is working to coördinate and expand these programs, as well as launching new efforts to bring hands-on science teaching directly to classrooms. “I’d like to see something like that, with minds-on, hands-on learning, that would be available with this kind of curriculum for every child in the country—every child in the world,” says Belcher. “We won’t be able to invite them all to MIT, but we would be able to give them an MIT experience.”

Transforming the ed school

Giving teachers in training an MIT-like experience is part of the goal of the ­Woodrow Wilson Academy, which grew out of the Woodrow Wilson Foundation’s desire to find a university partner for its own efforts to improve education. “MIT was at the top of the wish list,” says the academy’s director, Deborah Hirsch.

In 2001, foundation president Arthur Levine began a four-year research project looking at teacher education programs throughout the country, originally intending to refute criticism of education schools. “Unfortunately, the data showed the criticism was largely correct,” Levine says now. His scathing report found that most schools of education were engaged in a “pursuit of irrelevance” and had “not kept pace with changing demographics, technology, global competition, and pressures to raise student achievement.” Most university ed schools, he determined, were overly academic, focusing on education theory and policy, not teaching. Undergraduate teacher training programs, meanwhile, were outdated and underfunded, favoring an assembly-line approach that left teachers ill prepared for the real world.

By contrast, the academy will use a so-called competency-based program in which teacher candidates will move at their own pace, depending on what they need to learn.

MIT will serve as the incubator for the academy, which in turn will serve as a lab for MIT education researchers. Since last summer the academy has been housed on the east side of campus, in shared space with a startup vibe. Before the first class of two dozen teacher candidates enters in the fall of 2017, it will move to a permanent home, most likely near MIT. By then, says Hirsch, the academy will grow into a graduate school with a chief academic officer and faculty comprising teachers of math, biology, and chemistry. (Other subject areas will be added later.)

The curriculum, which the first class of students will help shape, will combine three elements: face-to-face instruction by teacher trainers, online courses that candidates can work through independently, and experience teaching in Boston-­area schools. Teacher candidates will be constantly assessed to ensure they assimilate key concepts, and the online curriculum will slow down or change course as needed. Levine likens it to the way a GPS recalculates a route.

In a room down the hall from the academy, MIT’s Teaching Systems Lab (TSL), which aims to help teachers effectively harness technology in the classroom, is coördinating with MIT faculty to develop a curriculum. Starting with biology—which has a shortage of qualified teachers at the middle and high school levels—TSL executive director Justin Reich has been working with MIT faculty to identify what teachers will need to learn: both the latest biology knowledge and the most effective techniques to teach it. To gain proficiency in these areas, he explains, teacher candidates will tackle specially designed “challenges” both online and in the classroom.

One challenge, for example, might be to create a welcoming culture in a high school biology lab; another might be to assess problems with a test on which the whole class did poorly, and figure out how to retool it.

Klopfer’s group, meanwhile, is working to create games and simulations that train the teachers and can help them embrace the use of technology in their future classrooms. One simulation, for example, might include watching a video about a volatile moment in class in which a boy says something offensive about girls in science. A teacher candidate might answer guided questions online during the video and then later discuss his or her choices with an instructor.

“That helps the teacher candidate understand the millions of choices they would make in a class period and reflectively rehearse how they would handle those choices in the future,” says Reich, adding that in some ways that kind of training is more effective than in-person training, since it exposes teachers to a wider range of situations than they might come across as student teachers in physical classrooms. “We want to really make sure every student encounters these kinds of decisions,” says Reich. “We can at least, in a simulated way, make sure that happens.”

Unlike most graduate schools, the academy will also continue to support teachers for three years after they graduate, providing mentoring and forums, both online and in person, in which to discuss classroom experiences with instructors. “That’s something schools struggle with,” says Hirsch.

Learning research

MIT won’t just be providing the knowledge and technological know-how for the academy’s curriculum; it will also enhance the teacher training with the latest in brain and cognitive research about how children learn. “One area in which MIT has incredible strengths, maybe unparalleled strengths, is the area of learning and cognitive science,” says Hirsch. “I would say that is missing from pretty much every school of education.”

That’s where MITili (pronounced “mightily”) comes in.

John Gabrieli, PhD ’87, who investigates the neurological underpinnings of learning and memory in the human brain, is spearheading the MITili initiative to promote new research on teaching and learning that builds on work across the Institute in fields including psychology, economics, neuroscience, engineering, and public policy. MITili will encourage interaction among researchers in different disciplines and fund new research to evaluate the effectiveness of teaching and learning environments.

“One of the problems in education is there are a lot of unproven and in some cases disproven things out there,” says Gabrieli, the Grover M. Hermann Professor of Health Sciences and Technology and Cognitive Neuroscience. “Almost every few weeks, a parent will say to me, if I paint my child’s room red, they’ll do better. We are taking a skeptical, data-driven perspective on whether something is helpful or not.”

In two studies, Gabrieli is using functional magnetic resonance imaging and other measurements to gauge the effectiveness of teaching techniques in two Boston charter schools. In the first study, he is testing the effect of meditation training on memory and cognitive performance; in the second, he is assessing how behavioral interventions by parents and attention-building exercises for students might affect school performance. “We’re asking, do these interventions change brain structures—and do those brain measures provide any insight into how to improve performance for those children?” he says.

MITili researcher Laura Schulz, a professor of cognitive sciences, studies how children construct concepts at a young age, and how curiosity and exploration can continue to be encouraged as they grow older. In recent studies, Shulz has found that the scientific method may be deeply ingrained in the way even young children experience the world, helping them draw much more sophisticated conclusions about natural phenomena than previously thought.

In addition to helping disseminate such findings through traditional methods like scholarly journals and conferences, MITili will work to make them more accessible online. It will also incorporate the teaching techniques that research finds effective into the curriculum of the Woodrow Wilson Academy, and share them through outreach efforts led by the pK–12 Action Group.

To that end, the Teaching Systems Lab is offering MIT researchers “teaching and learning innovation grants” of $25,000 to $100,000. “We are hoping to spur some research that is related and extend the work of faculty to focus on areas that are important to the academy,” says Hirsch.

The first round of recipients includes biological engineering instructor Natalie Kuldell, president of the BioBuilder Educational Foundation, who will design a high school biology curriculum to teach principles of evolution using synthetic-biology experiments; Schulz, who will research the impact of an informal science education initiative on children’s understanding of basic scientific concepts; and Resnick, who will develop new strategies that draw on the personal interests of students, particularly girls and students of color, to help them learn computer coding.

The ultimate goal will be to transmit the best teaching and learning practices gleaned from MIT research to newly trained teachers and, through them, directly to students. “We have no plans to create an ed school at this point,” says Sarma. “We also don’t have a medical school,” he points out, but through research centers such as the Koch Institute for Integrative Cancer Research and partnerships with Harvard and other medical schools, MIT has an outsize impact on biological and medical research. In the same way, says Sarma, efforts like the partnership with the Woodrow Wilson Academy could have a more profound impact on education than a dedicated education school would.

While the first class of teachers in 2017 will number 24 or 25, the plan is to ramp that up to 100 before long—and perhaps to as many as 200 eventually. More significant, however, the academy will disseminate its work far and wide to other schools of education and anyone else who would like to make use of it. “Everything we develop will be nonproprietary,” says Hirsch. “We are not trying to create the perfect teacher accreditation program in a petri dish. We are trying to create and test strong, innovative approaches and then help other teacher programs test and employ them. We are looking to improve teacher preparation at large.”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.