User interviews, like many other insight-rich UX research techniques, can be enhanced by AI-driven tools. While AI has been employed for tasks such as copywriting and paraphrasing, its potential for deeper activities like reflection and probing remains underexplored. To bridge this gap, we developed intavue.ai, a tool designed to support UX researchers by generating interview questions tailored to research goals. We evaluated the tool through a validation study (n=6), where results from SUPR-Q and cognitive load measures were positive. Building on this, we conducted a between-subject study (n=12) to investigate whether intavue.ai is more effective during its first or second use. Findings revealed a reduction in effort during the first use, while other metrics remain areas for further investigation. Finally, we provide design recommendations, including the concept of “probing with quintets” and extending applications to other reflective UX research methods, such as focus groups and brainstorming sessions. Our work underscores the transformative potential of AI-driven tools in enhancing qualitative research practices, fostering more meaningful and efficient user engagement.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

The Probing Machine: Can Using GenAI Tools Help With Better Reflections During User Interviews

  • Corey Tran,
  • Richard Lance Parayno,
  • Bharat Krishna Venkitachalam,
  • Janna Aika Deja,
  • Jordan Aiko Deja

摘要

User interviews, like many other insight-rich UX research techniques, can be enhanced by AI-driven tools. While AI has been employed for tasks such as copywriting and paraphrasing, its potential for deeper activities like reflection and probing remains underexplored. To bridge this gap, we developed intavue.ai, a tool designed to support UX researchers by generating interview questions tailored to research goals. We evaluated the tool through a validation study (n=6), where results from SUPR-Q and cognitive load measures were positive. Building on this, we conducted a between-subject study (n=12) to investigate whether intavue.ai is more effective during its first or second use. Findings revealed a reduction in effort during the first use, while other metrics remain areas for further investigation. Finally, we provide design recommendations, including the concept of “probing with quintets” and extending applications to other reflective UX research methods, such as focus groups and brainstorming sessions. Our work underscores the transformative potential of AI-driven tools in enhancing qualitative research practices, fostering more meaningful and efficient user engagement.