For decades, the science of learning has grown through the formulation and testing of theories pertaining to student growth. Scientists, researchers, and educators alike have attempted to answer questions as to how learning occurs, the right contexts, settings that foster learning, and the right actions, interventions, and practices that promote better learning strategies.
At the basis of many of the most prominent learning theories to emerge is the idea that failure is a necessary part of the learning process and provides opportunities to grow. It is important, therefore, for students to recognize experienced difficulties and struggle as opportunities to learn, but similarly important for educators to guide students in how to improve themselves.
Among the most widely-recognized theories of learning is focused on this concept of growth. Dr. Carol Dweck’s idea of a growth mindset seeks to instill the belief that all individuals are able to grow and learn, and it is through this manner of thinking that such an individual is able to recognize struggle as an opportunity for improvement.
Similarly, this theory of learning as well as others including Dr. Angela Duckworth’s concept of grit, and Dr. Robert Bjork’s idea of desirable difficulty, recognizes that motivation and persistence play an important role in the learning process. In this way, these theories recognize that keeping students engaged in their learning, especially while experiencing difficulty, is “half the battle,” so to speak, in helping students overcome their struggle to grow as learners.
But what about the other half? Is persistence, itself, enough to learn?
Motivation and engagement help to promote persistence in students, but without useful guidance, it is often difficult for students to know what to do in order to improve. It is from this realization that we recognize the role of feedback in the learning process and design much of ASSISTments around the idea of improving the feedback that is given to students.
While there are several existing tools and features within ASSISTments and more currently in development focused on improving and promoting various forms of feedback, the DRIVER-SEAT project exemplifies our effort to supporting teachers in the process of communicating feedback to their students; the project aims to leverage the power of artificial intelligence while keeping teachers “in the driver’s seat” when it comes to interacting with their students.
There are many ways in which ASSISTments can help automate feedback to students while they are working through assigned content. Features including immediate correctness feedback, common wrong answer messages, and teacher-authored hints and explanations all help to guide students in understanding and overcoming gaps in knowledge or misconceptions.
This is particularly easy for a computer to automate on close-ended problems that expect a single or finite number of acceptable correct answers (e.g. when solving for x in 5x+2 = 12). However, it is more difficult to automate feedback in more open-ended cases (e.g. when asking students to explain their reasoning).
Moreover, automating every aspect of feedback can reduce and detract from meaningful teacher-student interactions that are similarly important in the learning process; teachers know their students well, and it is with this in mind that the DRIVER-SEAT project is focused on augmenting the teacher’s ability to attend to student work and writing meaningful feedback with greater efficiency.
The DRIVER-SEAT project is leading to several tools that are currently in development. Most notably of these is QUICK Comments, which has been previously featured in the ASSISTments blog. This tool, which is currently undergoing pilot testing before being released to all ASSISTments users, aims to help teachers write messages to their students in regard to their answers to open-ended problems. Writing directed, meaningful feedback in response to student work can be time consuming, and this tool is attempting to use machine learning and natural language processing - representing state-of-the-art methods in the field of artificial intelligence - to save teachers time and direct their attention to the students most in need of help.
However, there are other “open-ended” aspects of student learning that the DRIVER-SEAT project is aiming to address. While QUICK-Comments is focused on providing feedback to student work on individual open-ended questions, teachers often want to also write messages to students in regard to an assignment as a whole.
Helping teachers to recognize and commend students for persisting and persevering through difficult problems or helping students who are struggling through content can be just as important as providing other forms of feedback. Similarly, helping teachers identify students who are rushing through problems or misusing hints and explanations can help them communicate and promote better learning and help-seeking strategies.
Currently in development, shortly behind QUICK-Comments, is a set of tools that will help teachers provide this kind of assignment-level feedback within ASSISTments, integrating with existing and future reports.
The support of teacher-provided feedback to their students through the DRIVER-SEAT project is designed to promote motivation and engagement as well as offer guidance to help students improve. This project exemplifies our effort to promote a growth mindset-approach to instruction and similarly illustrates how we are currently building learning theory into practical teacher- and student-facing tools.
These and other tools under development for ASSISTments are made possible through close collaborations with teachers. If you are interested in helping to develop or test these and future projects in ASSISTments, please contact us and sign up to join our Teachers for Research and Feedback community.