arXiv preprint
We tested whether photorealistic embodied conversational agents can improve the quality of online survey responses. With 80 participants and 2,200+ responses, the study shows both the promise and the challenges of making surveys feel more like real conversations.
Read article
arXiv preprint
We put a long-standing card sorting shortcut to the test: asking participants to sort random subsets of cards. With 160 participants, we found subsets can reproduce overall mental model structures but distort finer thematic details. The study also reveals how personality traits and cognitive reflection shape sorting behavior.
Read article
Smashing Magazine
Discussion on the significance of context in the creation of relevant follow-up questions for usability testing, how an AI tasked with interactive follow-up should be validated, and the potential — along with the risks — of AI interaction in usability testing.
Read article
Computer Standards & Interfaces
In a controlled study, we collect data for mouse dynamics utilizing a variety of mouse parameter configurations. We discuss methodological implications — how mouse dynamics studies should account for the diversity of mouse-related conditions.
Read article
International Journal of Human–Computer Interaction
We evaluate the merit of low-level interactions for the task of repeat purchase prediction and study feature sets utilizing low-level interactions, together with non-interaction data.
Read article
arXiv preprint
An in-depth study in which we experimented with data from 28 card sorting studies involving 1,399 participants from real projects across various industries. We tasked LLMs with simulating these studies using various prompting methods and then compared the results.
Read article
Information and Software Technology
To understand the methodological impact of tree testing tools on how users navigate through information, we conducted a set of experiments with 180 participants interacting with the same information architecture but represented by different tree testing variants. We found key differences.
Read article
Journal of Systems and Software
User research practitioners are leaving hotspot hints enabled (either unknowingly or deliberately). We investigate the impact of this seemingly helpful feature by analyzing how hotspot highlighting affects usability testing data.
Read article
Engineering Applications of Artificial Intelligence
Using a simple webcam, machine learning can estimate where on the screen you’re looking. But does it have the potential to democratize eye tracking? We developed a state-of-the-art gaze-tracking solution and discuss its potential for UX research.
Read article
International Journal of Human-Computer Interaction
We examine the implementation of AI-generated follow-up questions to assess their impact on usability testing insights. An experiment in which we compared feedback from no follow-up questions, static questions prepared by researchers, real-time GPT-4 questions, and a blend of static and AI-generated questions.
Read article
Smashing Magazine
We look under the hood of how first impressions are affected by time, the test participant’s cognitive abilities, and the visual complexity of the tested image, and how UX researchers and product owners can ensure that users get off on the right foot from their first steps.
Read article
arXiv preprint
In a systematic review of 155 publications, we provide a comprehensive overview of the current state of the art for automated usability issue detection. We analyzed trends, paradigms, and the technical contexts in which they are applied.
Read article
Computers in Human Behavior Reports
Exploration of how mouse dynamics and verbal responses can reveal deception in online personality questionnaires under realistic conditions. We show that while mouse-based lie detection struggles, analyzing speech cues offers a robust method.
Read article
Smashing Magazine
How to design a toggle button that shows the selected option effectively is a long-term open question among UX and UI designers. Despite our vigorous googling efforts, we have not found any guidelines based on real user research. So, we decided to step up and provide the UX community with a proper data-based answer
Read article
