By Judi Fusco
Cyberlearning projects are quite varied as we've discussed previously, and you can learn more about Cyberlearning at the CIRCL website. In this post, we share an interview about a project that helps students invent their learning and it helps teachers support students during the process. Catherine Chase, is the lead researcher and an Assistant Professor of Human Development at Teachers College, Columbia University. This interview is also posted on the CIRCL website.
What is the big idea of your project?
How do we we help students transfer, in a flexible, adaptive way, what they are learning in school to novel contexts and situations? Our project focuses on transfer of concepts at the intersection of math and science, and one thing that we've found to be very successful at promoting this type of transfer is a method we call "invention" (Schwartz et al., 2011).
Invention is an exploratory task where students engage in inventing conceptual ideas through an exploration of data, often contrasting cases. Contrasting cases are examples that have many similarities but a few key differences that relate to deep principles and conceptual ideas. By contrasting the cases, students come to notice features that are important to understanding, but may not be obvious to novice learners.
As students explore the cases, they are asked to "invent" fundamental equations such as those for density or speed. The process of inventing or even attempting to invent equations on their own (even if they fail) prepares students for additional learning. After the invention process, we follow up with expository instruction (i.e., lecture) on a topic important to the concept, such as ratio.
We are currently building the Invention Coach -- a system that guides and scaffolds students through the messy and iterative process of Invention. The picture below shows the Invention Coach’s main interface. In this screenshot, a student is working to invent an index of “clown crowdedness,” which is a proxy for density (mass/volume).
How do you use cyberlearning in your work?
Prior classroom studies with paper-based invention activities show that the process of invention is really successful in promoting transfer. The cyberlearning part comes in with the technology we are developing to support the learning. Invention works well for promoting understanding, but students often need one-on-one time with a teacher or facilitator to engage in productive invention. Unfortunately, it’s not possible to give a teacher to each student in a classroom to support the process. Our project is working to create a technology that can reduce the demands on the teacher by providing individualized and timely feedback to students throughout the invention process.
Before building our technology, we wanted to know what a human would naturally do to promote transfer. We started by observing the guidance a human invention coach (a teacher) naturally gives in one-on-one invention tasks with students as the students invent formulas for density and speed. Our initial research showed human invention coaches did help students learn (Chase et al., 2015) and that much of the work the coach did during the task involved asking questions and not giving answers. In fact, the more explanations a coach gave, the lower the transfer test score for the student. One might think that a human coach gave explanations because a student was struggling, and perhaps it was a poorer student to begin with and thus the lower scores aren’t surprising. However, our analysis showed that frequent explanations were not related to how a student was doing on the task, and that those explanations hurt the student’s transfer. It may be that the explanations “cut short” the time the student spends exploring and generating ideas, so the student doesn’t do as well on the transfer task.
Our initial work helped us understand the human expertise we needed to include in the technological Invention Coach. Now we are working to develop that Invention Coach to support all students in the classroom so they can engage in a productive exploration and invention experience.
One of the design challenges we face is that we are essentially developing an intelligent tutoring system to scaffold the solving of ill-defined problems whereas most intelligent tutoring systems focus on well-defined problems (often in algebra) with clear steps to get to the answer. In our ill-defined problems it’s not clear what the goal is nor is it clear exactly what the path is to the goal. Thinking about how technology can scaffold students through that process without overly guiding is one of the critical challenges of our work.
Tell me more about the Invention Coach and what it looks like.
The Invention Coach is an exploratory learning environment that follows a student's trajectory through an invention task and provides adaptive feedback and scaffolding to help them engage in productive exploration to prepare them to learn from later expository instruction (Marks, Bernett, & Chase, 2016). We have designed the initial prototype (see figure above). The software gives students an invention activity and the students can ask for help, submit their solutions, and get feedback along the way. We’ve created a few different kinds of interactive modules that focus learners on diagnosing their own errors and thinking deeply about these concepts. Some modules focus learners on comparing specially designed contrasting cases. In the literature, there is a lot of research showing the benefits of compare and contrast activities for helping learners focus on the deeper features that novices often overlook.
The image below shows the “feature contrast module” where students are asked to compare the highlighted green “Crazy Clowns Company” bus with the highlighted blue “Bargain Basement Clowns” bus, to help learners realize that space (the bus size) is an important feature of density (e.g. “clown crowdedness”).
If we walked into a classroom, what would it look like to have students using the Invention Coach?
We’re gearing up to do our first classroom study with the Invention Coach in the Fall, but right now, students work individually with the computer. As discussed above, there is a mentor character on screen that guides them through the process and provides hints and feedback (see first figure).
Invention can be pretty frustrating, because it is a very novel task for kids and it’s also an iterative task, where students frequently fail to come up with the right solution. However, in our studies, we often see kids having “Aha!” moments when they come to discover critical pieces of a sensible solution. Or more often, this happens during later expository instruction when they realize the sophistication of the canonical solution. For example, after attempting to invent a ratio-based equation (Density = mass/volume), one student said during the post-lecture on ratio “Oh! Now I finally understand division!”
In the future, we are open to the possibility of students working in collaborative pairs and are currently toying around with ideas for how to do that. David Sears has done some work with Invention and found that it’s much more productive when students work in pairs. In our work, they are paired with a computer-based coach. Supporting students working together may lead to more discussions, argumentation, explanation with a live partner and could be future work.
I’m also interested in developing more teacher-focused technologies that would engender classroom-based discussion around the kinds of mathematical models that kids are inventing and building. I want to understand when those models are effective or ineffective.
Is there more about the project you would like people to know?
People can visit our website to learn more or see our publications or they read about our Cyberlearning award Developing a tutor to guide students as they invent deep principles with contrasting cases.
Further Reading and References
Marks, J., Bernett, D., & Chase, C.C. (2016). The Invention Coach: Integrating data and theory in the design of an exploratory learning environment. International Journal of Designs for Learning, 7(2), 74-92.
Chase, C. C., Marks, J., Bernett, D., Bradley, M., & Aleven, V. (2015, June). Towards the development of the invention coach: A naturalistic study of teacher guidance for an exploratory learning task. In International Conference on Artificial Intelligence in Education (pp. 558-561). Springer International Publishing.
Schwartz, D. L., Chase, C. C., Oppezzo, M. A., & Chin, D. B. (2011). Practicing versus inventing with contrasting cases: The effects of telling first on learning and transfer. Journal of Educational Psychology, 103(4), 759-775.
By Patricia (Pati) Ruiz
The NSF 2016 Video Showcase: Advancing STEM Learning For All featured 65 videos under the “Broadening Participation” keyword. This topic is an important one for those of us who work in the classroom. I learned through the Cyberlearning 2016 conference that the NSF established the Broadening Participation in Computing Alliance Program between 2006 and 2009 to address issues of engagement and education in computing and computationally-intensive disciplines across the K-20 education landscape.
One underrepresented group discussed in the NSF video showcase was students with diagnosed learning differences. Two of the videos on this topic were very interesting to me – these were Diverse Learning Technologies: Helping students with LD, ADHD, and ASD reach their full potential in STEM and Accessible PhET Simulations for Diverse Learners. In order to learn more about working with diverse learners, I spoke with Amar Abbott, a High Tech Center Access Specialist and faculty member at Taft College. Here is our conversation:
Question: You watched the Diverse Learning Technologies: Helping students with LD, ADHD, and ASD reach their full potential in STEM and Accessible PhET Simulations for Diverse Learners videos from the 2016 NSF Video Showcase. What did you find interesting about them?
Amar Abbott: I thought that they were very informative and I especially liked that all of the videos had closed captioning embedded. This makes them accessible to a wider audience. Also, I learned a lot about the technologies that are being developed for helping students with learning differences.
Question: How might these videos inform your practice?
Amar Abbott: These videos have helped me think more deeply about cognitive load and helping students and teachers monitor learning. One question that came up for me is: If a tool provides some support for the cognition of the student, what can the community [around the student(s)] also do to help support cognitive learning? In addition, I would also like to learn more about the situative and social supports that might help students.
Question: What will you do with what you learned from these videos?
Amar Abbott: I appreciated the introduction to Landmark College, their resources, and research group there; it’s a great model. Their focus on UDL is especially excellent because their work stems from direct experience. I am going to try to visit Landmark and hopefully develop long-term relationships with the researchers and students there.
As for the accessible PhET Simulations for Diverse Learners project video, I learned a great deal about what affordances learners need for activities like simulations. The video was to the point, and I like that it highlighted a blind student working with the PhET Simulations. Projects like these puts accessibility in the forefront and that helps all learners.
It was great to hear Amar’s thoughts, and as a teacher, I will be interested in following both of these projects. I am especially interested in what the University of Colorado team discovers about how simulations “are shaped by sociocultural norms of science, [and] can also be used to change the traditional norms of how students engage in the classroom.” This work will be helpful for everyone interested in broadening participation and engaging all learners in STEM topics.