<p>Bidirectional Feedback may benefit students by facilitating learning success and goal achievement and can help teachers address students’ needs and/or assess and improve learning material. Current feedback practices have limitations, such as the difficulty for students to formulate understandable feedback for teachers, the effort required for students and teachers to engage in feedback dialogue, and the challenge for teachers to manage multiple feedback responses across a wide range of learning material. Thus, there is a need for computational support in a digital learning environment that facilitates bidirectional feedback creation, processing, and dialogue. To this end, we present a process model of bidirectional feedback, a conceptual design of a computational support to assist this process, and its implementation for the self-assessment task type in Moodle. It was implemented and evaluated in a distance learning course in a B.Sc. Computer Science program. With the use of computer-assisted support for the bidirectional feedback process, an increase in feedback from more students was observed and was related to more self-assessment tasks. Feedback included not only negative but also positive feedback. Teachers confirmed the usefulness of computational support as well as the usefulness and understandability of student-generated feedback messages, both for helping students and for assessing and improving the quality of assignments.</p>