AI-assisted Wireframing: Comparing AI and Human Approaches to UX Wireframing: Evaluating Usability Through Heuristics and Cognitive Principles
2025 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE credits
Student thesis
Abstract [en]
With a focus on UI element placement and structural logic, this study looks into the differences between wireframes created by AI and those created manually. It is important to determine if AI technologies like Uizard improve or hinder the design process as they become more integrated into design workflows. This study used a qualitative research approach to explore how users interact with AI-assisted design tools versus manual wireframing. Through heuristic evaluations, screen/audio recordings, and post-task interviews with both experienced and novice designers, the research examined user experiences, cognitive load, and design decision-making based on principles like Hick’s Law and choice overload.
The results showed that AI-generated wireframes often created clean and well-organized layouts that helped reduce choice overload and simplified user interaction. However, they also sometimes missed important details and required well-structured prompts to work effectively. Designer feedback showed that while AI tools like Uizard can speed up early design stages and offer inspiration, they still lack the deeper user understanding and creative flexibility found in manual designs. Overall, AI-assisted wireframing can serve as a useful starting point but still needs human input to achieve optimal usability.
General terms such as heuristics and user-centered design that are used throughout this study are primarily judged through the lens of choice overload- and cognitive load (informed by Hick’s Law), speed of completing tasks, and subjective satisfaction.
Place, publisher, year, edition, pages
2025. , p. 60
Keywords [en]
AI-assisted wireframing, UX Design, Human-Computer Interaction (HCI), Heuristic Evaluation, Hick’s Law, Choice Overload, Cognitive Load
National Category
Computer and Information Sciences
Identifiers
URN: urn:nbn:se:hj:diva-68933OAI: oai:DiVA.org:hj-68933DiVA, id: diva2:1975289
Presentation
2025-05-26, Jönköping, Sweden, 09:00 (English)
Supervisors
Examiners
2025-07-012025-06-232025-10-13Bibliographically approved