Structured over Chat-Based Feedback: Comparing User Preferences on Different Presentation Feedback Reports

Authors

  • Noemi Giulia Bontempo Goethe University, Frankfurt am Main, Germany
  • Nina Mouhammad DIPF | Leibniz Institute for Research and Information in Education, Frankfurt am Main, Germany
  • Daniele Di Mitri German University of Digital Science, Potsdam, Germany; DIPF | Leibniz Institute for Research and Information in Education, Frankfurt am Main, Germany

DOI:

https://doi.org/10.34190/ecel.25.1.5389

Keywords:

Presentation Training, Feedback Visualization, Reflection, Educational Technology, Learning Analytics

Abstract

Digital presentation training systems provide automated feedback during presentation practice, but less attention has been paid to how feedback should be presented after a training session to support review and reflection. This paper investigates how learners perceive different feedback representations for understanding presentation performance and reflecting on it. A Rehearsal Report was designed for Presentable, a digital presentation training system that provides automated feedback on presentation delivery. Three concepts were developed: a timeline-based design that connects feedback to specific moments in the presentation, a chat-based design that presents feedback conversationally, and a storytelling-based design that organizes feedback into thematic categories. The concepts were evaluated in an online within-subject study with 84 participants. Participants reviewed all three designs and evaluated them regarding clarity, usefulness, reflection support, preference, and perceived support for understanding presentation performance. The results show that the chat-based design received lower ratings than both structured alternatives. Participants preferred designs with clear visual organization and contextual information. The timeline-based design was rated most helpful overall, while the storytelling-based design was rated highest for ease of understanding. These findings suggest that structured visual representations may be perceived as more supportive for post-training reflection than conversational feedback formats.

Author Biographies

Noemi Giulia Bontempo, Goethe University, Frankfurt am Main, Germany

Noemi Bontempo holds a bachelor’s degree in computer science from Goethe University Frankfurt. Her research interests include educational technology and human-computer interaction. In her bachelor’s thesis, she explored how different representations of post-training feedback can support learners’ reflection and understanding of their performance in digital presentation training.

Nina Mouhammad, DIPF | Leibniz Institute for Research and Information in Education, Frankfurt am Main, Germany

Nina Mouhammad is a doctoral researcher at DIPF | Leibniz Institute for Research and Information in Education in Frankfurt. She holds a master’s degree in computer science with a minor in Human Sciences. In her PhD she explores how software can support students in learning to create more effective presentation structures.

Daniele Di Mitri, German University of Digital Science, Potsdam, Germany; DIPF | Leibniz Institute for Research and Information in Education, Frankfurt am Main, Germany

Daniele Di Mitri is Professor of Multimodal Learning Technologies at the German University of Digital Science and an associated researcher at DIPF Leibniz Institute. He leads the Augmented Feedback group, developing AI-driven multimodal learning technologies that interpret learner behaviour to provide personalised feedback, adaptive support, and intelligent tutoring systems and innovations.

Downloads

Published

2026-10-07