AI-prompted peer feedback: Can an AI feedback coach enhance peer feedback quality?

Publikation: KonferencebidragPosterForskningpeer review

Abstract

Peer feedback and cooperative learning are concepts that are often attributed with having an impact on student performance, and therefore a focus in the development of educational technology. The integration of Generative AI into higher education practices raises questions about its potential role in enhancing student learning.
This submission discusses the use of LLM-based chatbots in educational settings, emphasizing the technology’s unique ability to simulate interactions and conversation. The proposed study examines how chatbots can support learning through dialogical processes, thereby facilitating self-paced reflection and engagement with subject matter, that are traditionally assisted by peers. The proposed study aims to determine if students regularly using AI chatbots for reflection and feedback achieve comparable learning outcomes to those engaging in peer discussions, and potentially better outcomes than students practicing solo self-reflection. The following research question is therefore proposed: How can AI chatbots support student learning through dialogical processes?
A quasi-experimental intervention study will be conducted with four groups: one using a chatbot, another encouraged to do reflections and feedback with peers, a third group focused on self-reflection and self-initiated feedback, and a control group without intervention.
The study will use pre and post-test surveys to measure cognitive load, emotion, and immersion, as ways of evaluating student learning and retention.
This research design builds upon a pilot study indicating that AI coaching significantly affects the style and characteristics of written feedback.
The study is proposed as the main part of a future Ph.D. project, which is currently in the preparatory phase, that will also include a qualitative study based on observations and interviews, as well as a comparative Natural Language Processing analysis of feedback comments, to explore the broader impacts of AI chatbots on teaching and learning processes, this roundtable submission invites discussions and suggestions concerning both method and theory.
OriginalsprogEngelsk
Publikationsdato21 mar. 2024
Antal sider3
StatusAccepteret/In press - 21 mar. 2024
BegivenhedJURE 2024 - Faculty of Psychology of the University of Sevilla, C. Camilo José Cela, s/n, 41018 Sevilla., Sevilla, Spanien
Varighed: 24 jun. 202428 jun. 2024
Konferencens nummer: 28th
https://www.earli.org/events/jure2024

Konference

KonferenceJURE 2024
Nummer28th
LokationFaculty of Psychology of the University of Sevilla, C. Camilo José Cela, s/n, 41018 Sevilla.
Land/OmrådeSpanien
BySevilla
Periode24/06/202428/06/2024
Internetadresse

Emneord

  • læring
  • generative AI
  • universitetsdidaktik
  • læringsteknologi

Fingeraftryk

Dyk ned i forskningsemnerne om 'AI-prompted peer feedback: Can an AI feedback coach enhance peer feedback quality?'. Sammen danner de et unikt fingeraftryk.

Citationsformater