The influence of prompt formulation on AI-Generated motion design quality : An experimental study of UX/UI micro-interactions with Figma AI

Sammanfattning

With the increasing integration of artificial intelligence into UX/UI design tools, designers can now create interactive behaviors and motion designs using text prompts. While these AI-based solutions offer efficiency and creative support, the way prompts are formulated may significantly influence the quality of the generated output — an area that remains underexplored in current research. This study uses an AI tool in Figma to investigate how different prompt frameworks impact AI-generated micro-interactions in UX/UI design. Motion design outputs were generated using three structured prompting frameworks representing different approaches to prompt structure: CARE, RTF, and Chain-of-Thought (CoT). The generated outputs were analyzed using qualitative deductive content analysis based on six of Nielsen’s usability heuristics, and a survey was conducted to evaluate the perceived quality of the generated motion designs among participants with design experience. The results indicate that prompt framework structure influences both the technical quality and perceived quality of AI-generated micro-interactions. The CoT framework produced outputs most closely aligned with UX motion design principles and received the highest perceived quality ratings across most dimensions, while all three frameworks generated animations that broadly satisfied the selected evaluation criteria.