Emotionally enriched AI-generated feedback: Supporting student well-being without compromising learning
Omar Ali Saleh Alsaiari, Nilufar Baghaei, Hatim Lahza, Jason M. Lodge, Marie Bodén, Hassan Khosravi
Abstract
The use of AI-generated feedback in higher education has received growing attention, with most existing research emphasising its accuracy, usefulness in improving student work, and scalability. However, little attention has been paid to the role of emotional cues such as encouragement, praise, and empathetic language in shaping how students perceive and respond to feedback. This study addresses this gap by investigating whether enriching AI-generated feedback with motivational language influences students’ emotional responses and engagement. Drawing on the Control-Value Theory of Achievement Emotions, we conducted a randomized controlled experiment involving 395 participants, in which the experimental group received AI feedback enhanced with motivational elements, while the control group received neutral feedback. Our results show that the enriched feedback was perceived as more helpful and significantly reduced negative emotions—particularly anger—towards receiving feedback. However, it did not significantly affect students’ engagement with the feedback or the quality of their revised work. These findings highlight the potential of emotionally enriched AI feedback to foster more supportive and emotionally attuned learning environments without compromising learning outcomes, and underscore the importance of designing affective feedback systems that balance emotional well-being with sustained improvements in performance.