The third installment in our EF+Math blog series is a feature on the collaboration between the CueThinkEF+ & Pennesota R&D team.
Introduction to CueThinkEF+ & Pennesota
Together, CueThinkEF+ & Pennesota form one of three R&D teams in the EF+Math portfolio that is continuing to develop and study our core hypothesis: The integration of EF skill development in math learning approaches that address conceptual understanding and complex problem solving, in ways that afford equitable experiences in math learning, can dramatically increase students’ math outcomes. Each of the R&D teams in EF+Math’s portfolio uniquely address the intersection of executive function (EF), mathematics, and equity. The learnings from each team’s approach contribute to testing EF+Math’s hypothesis and inform continued research and development in this intersection.
Watch short videos about each of EF+Math’s R&D teams for more information.
The visual below demonstrates the complementary nature of each approach in the EF+Math Program’s portfolio.
CueThinkEF+ and Pennesota are engaged in a groundbreaking collaboration. They are focused on building technical capacities and assessments, and are redefining how we approach mathematical problem-solving and executive function (EF) development. Implicit in CueThinkEF+’s theory and approach is the idea that students’ EF, metacognition (the ability to plan, monitor, and reflect on your own thinking), and math perceptions (involving math identity, math beliefs, anxiety, and self-efficacy) are all crucial factors in their success in problem solving. The CueThinkEF+ & Pennesota project represents an equity-centered, inclusive approach to leveraging AI capabilities that honor student brilliance, support learner and teacher needs, and move the math education field towards more equitable educational experiences.
The Promise of CueThinkEF+ & Pennesota: Integrating Equity-Driven EF Supports and Ethical AI Across Problem-Solving Phases
CueThinkEF+ is an innovative application focused on improving math problem-solving and collaboration skills; the system has embedded scaffolded supports for executive function and metacognition skills so that students can progressively use these skills effectively in solving problems. At the heart of CueThinkEF+’s approach is a four-phase method designed to support effective student problem-solving: explore, plan, solve, and review.
CueThinkEF+ makes students’ mathematical thinking visible by allowing students to move back and forth across these stages to tackle complex problems, using prompts to start their thinking in the “Plan” phase, or selecting different tools to represent their thinking in the “Solve” phase. While the structure of each phase intends to build students’ habits of mind in breaking down non-routine tasks, CueThinkEF+ recognized that to truly empower learners, they needed deeper insights into how students engage in problem-solving activities – and where they may need additional support.
That’s where the Pennesota team comes in. The Pennesota team embeds just-in-time EF and metacognition supports in the CueThinkEF+ system through AI-driven learning detectors and probes to understand student behaviors and provide personalized interventions. They are able to mine the rich student data – text, visuals, audio, and interaction log data – emerging from the CueThinkEF+ platform to build these detectors.
These technical capacities include:
- detectors of self regulated learning (SRL) activity that provide a more nuanced understanding of student engagement and EFs while practicing mathematical problem-solving
- in-the-moment assessment of student annotations to support them in providing peers with high-quality feedback on their mathematical problem solving
- probes that gather in-the-moment data, about students’:
- engagement and mind wandering in math digital learning environments.
- self-efficacy beliefs and how confident they are in their ability to solve the math problems facing them
- pathways through problem solving as they reflect on their strategies and decide to keep or change their plans
With the data collected through these tools, CueThinkEF+’s platform can tailor EF and metacognitive supports to students, providing the right supports to the students who need them, at the right moment in time to be most effective for their learning. The combination of machine learning with deeply inclusive human-driven product development is an example for the sector of how to create AI-driven products that foster student agency and reasoning. Further, the platform allows teachers to access data about student learning in meaningful ways to inform their instruction.
Moreover, the teams have been vigilant about algorithmic bias, ensuring that the AI-driven components of CueThinkEF+ do not perpetuate inequities. When developing their detectors, which pick up on language and behaviors in the platform, the team intentionally trains their AI tools with culturally and racially diverse samples of student data and regularly checks for bias. By maintaining strong relationships with partners in school districts, the team gains invaluable insights into the contexts and systems where the platform will be deployed.
To Learn More, Watch The Product Demo
CueThinkEF+ supports learners in developing problem-solving schemas and provides support to exercise student agency and deepen their metacognitive skills. Early research suggests that students who repeatedly used the CueThinkEF+ platform improved their problem-solving abilities. Results from additional early efficacy studies indicate positive correlations between product usage and problem solving accuracy, as well as with students’ mathematical identities and beliefs.
Further, the team is finding new relationships between metacognition, EF, and affect in predicting problem-solving success (Rhodes, et al., 2023)1. We are looking forward to the results of ongoing research to further understand these correlations. We invite you to explore this recent blog from Dr. Rick Bryck, co-Principal Investigator of CueThinkEF+, regarding the role of EF in mathematics and specific examples of the strategies and tools CueThinkEF+ has incorporated to provide scaffolded EF supports that can help all students solve complex math problems .
Rhodes, S., Bryck, R., Gutierrez de Blume, A. (2023). Exploring factors influencing success in mathematical problem solving. In T. Lamberg & D. Moss (Eds.), Proceedings of the forty-fifth annual meeting of the North American Chapter of the International Group for the Psychology of Mathematics Education (Vol. 2) (pp.92-105). University of Nevada, Reno.
The Power of Inclusive R&D
What sets the CueThinkEF+ & Pennesota project apart is not just its technological advancements, but the equity-centered, inclusive approach which underpins its development.
Co-design has been at the core of CueThinkEF+’s process, bringing together Black and Latinx students, and students of all races experiencing poverty, along with educators, researchers, and developers to ideate and revise platform features. This collaborative effort ensures that the platform centers the needs of the people who will use the CueThinkEF+ platform – students and teachers. Empowering students to be able to shape their educational experiences is essential, and the process of co-design creates feedback loops between product developers, researchers, and their users that provide transparency and accountability around the development process for all involved.
Further, positioning students from historically marginalized communities at the center of curriculum design processes is necessary to ensure that the resulting products are reflective of cultures and values that are often ignored, making them more relevant and engaging for all students. Students and teachers feel like active partners, not just receivers, of the tool – and are collectively building knowledge regarding how students learn that can inform curriculum development at large.
The CueThinkEF+ & Pennesota team use the CEATL framework to guide the co-design process:
Dr. Aris Winger, an Associate Professor of Mathematics at Georgia Gwinnett College, has been an instrumental member of the CueThinkEF+ team in developing their process for engaging with students and teachers and facilitating co-design sessions. Dr. Winger remarked,
“It has been transformational for me as an educator and a person to take part in and witness a space where teachers and students of color can honestly give their perspectives about what they need from the application to improve their experience in mathematics. Moreover, watching the team receive their feedback, work diligently to apply it to the application, and bring it back to the users culminated, for me, an authentic process that gives me hope for a new way of improving mathematics applications to serve all of our students and teachers. “
The fruits of these co-design sessions are evident in the platform. For instance, students helped design an animated helper that supports metacognitive reflection by prompting learners to “Stop and Think” during problem-solving. The students workshopped language for prompts the animated helper could pose throughout the platform phases that were engaging and useful – significantly improving the usability and accessibility of the program.
The Road Ahead
The CueThinkEF+ & Pennesota project represents a thoughtful integration of AI into education—one that enhances rather than replaces human interaction and agency. By blending machine learning capabilities with deeply inclusive, human-driven product development, CueThinkEF+ offers students opportunities for collaboration and discourse while providing teachers with actionable data about student learning.
This approach acknowledges that mathematical problem-solving, while crucial, can be challenging to prioritize within teaching routines. CueThinkEF+ not only supports students in understanding problem-solving structures but also empowers them to exercise agency and deepen their grasp of mathematical concepts.
Looking to the future, CueThinkEF+ & Pennesota are individually and collectively expanding their platform and technical capabilities to iterate upon existing metacognitive supports based on recently collected data; they also aim to improve teacher features that can help all teachers build their capacity for equitable instructional practice through using the platform. Leveraging AI in intentional ways can offload the cognitive load for both students and teachers so that they can engage deeply in rich mathematical discourse and reasoning activities essential for student success.
For educators, school leaders, researchers, and product developers alike, the CueThinkEF+ story offers valuable lessons. It demonstrates that when we leverage AI capabilities in service of students’ mathematical brilliance and teacher needs, we can create tools that truly move the needle on educational experiences.
As we stand at the intersection of technology and pedagogy, projects like CueThinkEF+ light the way forward—showing us how collaborative innovation, rooted in empathy and equity, can transform learning for all students. We are thrilled for the CueThinkEF+ team as they recently announced their acquisition by Imagine Learning. It is exciting to know that millions of students will be able to benefit from CueThinkEF+’s AI-driven innovative math learning solutions that support equitable and engaging math learning for all learners. In a world where the discourse around AI often veers into concerns about replacement and depersonalization, the CueThinkEF+ project stands as a beacon, illuminating the potential for technology to amplify human connection, creativity, and growth in our classrooms.
