Smile to Like: Effects of Facial Expression-Based Evaluation on Media Engagement and Emotional Self-Awareness

Social media platforms widely use the like function to evaluate content and stimulate user engagement. However, while this function was originally intended to express authentic emotional responses, it is now often used for impression management or social signaling. This study proposes an approach that directly links the bodily expression of smiling to evaluation. This approach enhances user engagement with video content by removing the need for manual operations and is also expected to improve the user’s emotional state based on the facial feedback hypothesis, which suggests that deliberately forming a smile induces positive psychological effects. We developed a video-viewing smartphone application that detects smile intensity via the front-facing camera, converts detected smiles into likes , and delivers real-time visual feedback. In a within-subjects study with 18 participants, we compared this smile-based evaluation with a traditional button-based method. Quantitative analysis showed that the smile-based evaluation significantly improved user engagement in terms of attention to content and feeling of involvement. Qualitative analysis revealed Embodied Feedback Loops comprising smile generation, system response, and amplified positive affect, strengthening emotional self-awareness. At the same time, the study highlighted challenges related to variability in users’ expressiveness and biases against content that elicits less visibly expressive content. These findings indicate that smile-based evaluation can curb strategic behavior while enriching affective engagement, offering design insights for perceptually grounded, expression-based interaction techniques in immersive media applications.