It’s hard to crack a journal in developmental psychology or education without bumping into an article on mindset. Research has shown that students who believe they can improve their intellectual abilities work harder and perform better academically than those who believe their abilities are a fixed trait. But plenty of questions remain about the contexts that best help students develop “growth mindsets,” a term coined by Stanford University psychologist Carol Dweck, PhD, who has been a leading figure in the area of mindset and children’s beliefs about success for more than 30 years.
Answering those questions means thinking bigger, says David Yeager, PhD, an associate professor of developmental psychology at the University of Texas at Austin (UT), a former graduate student of Dweck’s. “The education field is littered with studies from the local district next door to a given university,” he says. And those samples often fail to represent the makeup of the nation’s schoolchildren. Enter the National Study of Learning Mindsets. Co-led by Yeager and Dweck, the study tested a short online growth mindset intervention in more than 16,000 high school students across the United States—the first ever national randomized controlled trial of such an intervention (Natureopens in new window, Vol. 573, No. 7774, 2019).
“The 50-minute intervention is free and scalable. Kids think they’re just completing a survey, but in fact they’re getting all sorts of information about their brain and their ability to learn,” says Yeager, who received a 2020 APA Award for Distinguished Scientific Early Career Contributions to Psychology. Among lower-achieving students, the intervention reduced the prevalence of fixed mindset beliefs. “We also found the treatment improved grades in kids with lower academic achievement and increased kids’ likelihood of taking advanced math the following year.”
Designing such a large-scale study takes a lot of time, energy, and money. Yeager thought there must be a better way to usher his field forward. So, in the early months of 2020, he folded his independent Adolescent Development Research Group into the newly launched Texas Behavioral Science and Policy Institute (TxBSPI), which he co-created with three UT colleagues: social psychologist Christopher Bryan, PhD, and sociologists Robert Crosnoe, PhD, and Chandra Muller, PhD. “We decided to create a new kind of research institute to focus on contextual factors that moderate the treatment effects of psychological interventions,” Yeager says. “We’re reshaping the idea of the psychology lab and pooling our resources and shared infrastructure to get more done.”
Breaking down barriers to success
The TxBSPI focuses on issues of inequality and barriers to social mobility, says Bryan, a longtime collaborator of Yeager’s who relocated from the University of Chicago in 2020 to help launch the new institute. “We mapped out five policy areas that we want to build strength in: education, criminal justice, political participation, workplace inequality, and health,” Bryan says. Some of the collaborators maintain independent lines of research outside the partnership, but Yeager has largely subsumed his lab into the institute, he says.
Much of the team’s previous work has been in adolescent development and education—an area that has long fascinated Yeager. He spent two years as a middle school teacher and earned a master’s degree in education before pursuing a psychology career. “I loved working with teenagers and felt it was a calling. But I got more interested in the science of helping young people thrive and flourish than I was in teaching the content,” he says. “A lot of the advice I got about how to connect with young people and keep them engaged wasn’t very compelling, and I wanted to do new research to address those problems.”
He completed his PhD in developmental and psychological science at Stanford in 2011. During the last two years of graduate school, he also worked as an associate researcher at the Carnegie Foundation, applying psychological insights to a large initiative to reform community college remedial math education. An assessment found the Carnegie Math Pathways program significantly improved community college student success rates, with 54% of students earning college-level math credits in one year, compared with 6% of students in traditional developmental math programs (Huang, M., Six years of results from the Carnegie Math Pathwaysopens in new window, 2018). “That [project] taught me the power of ambitious shared projects to bring people together in teams—and the value of psychological research in addressing broad questions about human development and social problems,” Yeager says.
Yeager then joined the faculty at UT, where his work centers on helping adolescents overcome the psychological barriers that stand in the way of success, such as the stress and anxiety that can go along with school. In one example, Yeager and colleagues found that adolescents who completed a 25-minute reading and writing exercise emphasizing the belief that people have the potential to grow and change had improved cognitive and behavioral responses to acute stress compared with classmates who completed a control exercise. They also showed more adaptive neuroendocrine and cardiovascular responses to stressors (Psychological Scienceopens in new window, Vol. 27, No. 8, 2016).
Following that finding, Yeager and Bryan, along with collaborator Jeremy Jamieson, PhD, at the University of Rochester, developed a 25-minute digital intervention to improve people’s responses to stress. The treatment blends a growth mindset intervention with another intervention to help people reframe their stress as useful for enhancing performance rather than as a sign they’re going to fail. In a series of studies being prepared for publication, they tested the intervention in high school and college students and found participants had improved appraisals of hypothetical stressors and a reduced physiological response to stress.
In a happy accident, one of those studies spanned the early days of the COVID-19 pandemic. In this study, college students completed baseline anxiety measures in January 2020. Then, researchers followed up with participants as planned in April, during the high-stress early days of pandemic lockdown. In people who started with poor stress-management mindsets, the intervention group showed a significant reduction in overall anxiety symptoms, especially among women. “That suggests even a one-time message can persist months later,” Yeager says.
While the treatment was tested in high school and college students, he says the same approach could apply to adults going back for technical training after losing a job or changing careers—often an anxiety-inducing prospect. “We’re developing treatments to help people deal with the stress and anxiety of needing to shore up [skills] in the new economy, especially after COVID,” he says.
Changing teachers’ mindsets
Many of Yeager’s interventions aim to change an individual’s mindset. But a growth mindset can only get you so far if you’re surrounded by naysayers. “Growth mindset interventions for kids aren’t likely to be effective if their teacher has a fixed mindset,” he says. “But instead of thinking of teacher mindsets as a limitation, we decided to say, ‘What if the challenge of changing teacher mindsets is the most interesting scientific puzzle to be working on now?’ Rather than only trying to change the mindsets of underrepresented groups, we think the field should be working on changing the mindsets of people in power who are gatekeepers to opportunity.”
To that end, Yeager and Mary Murphy, PhD, a psychologist at Indiana University, launched the Texas Mindset Initiative. The project mirrors the National Study of Learning Mindsets—but this time it’s teachers, not students, who will receive the interventions. Those interventions will be delivered through an innovative new research panel dreamed up by Yeager and his colleagues at the TxBSPI. “When I started the national study, I flew to 60 schools to personally recruit them. That’s obviously not sustainable, and it doesn’t make sense to create a whole new research infrastructure for every study,” he says. Instead, he developed a kind of supercollider for intervention science.
“In physics, researchers solved problems by creating supercolliders and letting everyone share time on that shared resource so they can run their experiments and get answers quickly,” he explains. His version involves a panel of more than 1,000 Texas high school teachers who work with OnRamps, a dual-enrollment and distance-education program that allows high school students to engage in college experiences and earn college credit. Those teachers are available for experiments twice a year as part of their employer-mandated professional development—a perfect opportunity for engagement. “Researchers can apply for space on the panel, and we can run five different experiments, twice a year,” Yeager says. “And because our partner is a service provider, we also have access to student outcomes data.” The team launched the first five randomized controlled trials on the platform in fall 2020.
One trial is spearheaded by Kyle S. H. Dobson, PhD, a postdoctoral researcher who joined the lab in 2020. Dobson is studying how people’s differences can be seen as strengths and how teachers can support students’ authentic selves to help them stay engaged in learning. He developed an online intervention and tested it with 1,000 high school teachers in the fall. “The goal was to show teachers how to support students’ authenticity so that students feel heard and stay engaged,” Dobson says. He doesn’t define the differences: Some students may feel different because of their race or socioeconomic status, while others have a learning disability or an atypical family structure, he says. “This can apply to any student who is disengaged and doesn’t fit the mold of the ‘ideal’ student.”
Before launching the intervention, Dobson pilot-tested it with teachers to make sure it was something that really supported them in their goals. “One of the lab’s philosophies is the idea that you need to figure out what people care about and meet them where they’re at. Teachers want to make sure their students are heard and engaged,” he says. “There’s also a feeling of responsibility that it’s our job as scientists to make sure our research isn’t just theoretically interesting but that it’s going to positively impact people.”
Creativity and collaboration
Yeager isn’t afraid to rethink the status quo to make a difference. “People always say, ‘I could do that if I had time.’ But David set grand goals and said, ‘If we don’t have the time, then we need more people and should put together a team,’ ” Bryan says. “He’s an engine of energy and productivity.”
Though Yeager emphasizes that the TxBSPI has been a collaborative effort from the start, it was born in part from that energy. And despite launching just a few months before the COVID-19 pandemic, the team has been able to take advantage of their new research platform to study interventions with teachers and students during a challenging period in which many schools are in hybrid or distance learning. “We’re the only game in town for teacher-focused randomized controlled trials during the pandemic,” he says.
In addition to his three co-investigators, the institute includes four PhD-level permanent research associates and a computer scientist, as well as graduate and undergraduate students and postdoctoral researchers working with each of the principal investigators, from fields including education, business, social psychology, and sociology. “It’s a huge and interactive team, with other faculty members pitching in to offer advice and insight,” says Yeager’s fifth-year PhD student Melanie Gonzalez, who studies racial and socioeconomic inequality in teachers’ use of motivational practices.
Yeager also maintains active collaborations with researchers outside the institute, including a separate line of research to improve psychology methodology. To improve transparency and address critics concerns about the replicability of the research on growth mindset, Yeager and colleagues arranged for an independent evaluation of the National Study of Learning Mindsets, which found results consistent with their own (Zhu, P., et al., Using a growth mindset intervention to help ninth-gradersopens in new window, MDRC, 2019). He’s also shared materials with outside research groups in the United States and abroad and has established a rigorous protocol for preregistration for anyone using TxBSPI data sets. “Psychology’s approach to preregistration has a long way to go. Rather than wait around for the field to develop better standards, we’ve developed our own,” he says.
Whether he’s improving opportunities for young people or honing the field of psychological science, Yeager is counting on a combination of creativity and collaboration to make a difference. “Psychologists are so often stuck in the model of having their own independent labs. But we could get so much more done if we collaborated and pooled our resources,” he says. “When we combine forces, we can do so much.”
“Lab Work” examines the work of psychologists in research labs.
