Human-Computer Interaction

Permanent URI for this collectionhttps://hdl.handle.net/10125/112509

Browse

Now showing 1 - 9 of 9

Governing Generative AI Use through FAIR User Experience: Designing for Interactional Integrity
(2026-01-06) Abhari, Kaveh; Lahiri, Monjima; Paila, Sumanth
Prevailing HCI paradigms in GenAI design prioritize harm prevention and abstract principles, such as transparency, but often neglect the embodied and interactive ways in which AI shapes user behavior and identity. This paper advocates for a fundamental shift, positioning “interactional integrity” as the central design goal and moving beyond a reductionist focus on harm minimization. We use the FAIR framework—grounded in the existential principles of Freedom, Authenticity, Intentionality, and Responsibility—to cultivate interactional integrity through interface-level affordances. By embedding these principles into user experience, the FAIR framework offers a systematic approach for integrating responsible AI into design practice, ensuring that responsible use becomes a lived interaction rather than a regulatory afterthought.
Towards Simulating User Behavior for Automating Usability Tests by Employing Large Language Models
(2026-01-06) Griesbach, Marie; Lütke Stockdiek, Janina; Winkelmann, Hendrik; Grimme , Christian
Large Language Models (LLMs) enable the automation of tasks that typically require substantial manual effort. This work investigates their applicability in the context of usability testing. First, we evaluate whether LLM-agents can navigate in and interact with different applications to accomplish given tasks. Second, we compare LLM-generated streams-of-thought with human think-aloud comments collected during usability tests. Results show that, based on GPT-4o, LLM-agents can successfully interact with websites and perform tasks such as information search. However, they often fail to recognize task completion and tend to engage in actions beyond the intended goals. The comparison further reveals clear differences between LLM-based and human observations: while human users overlook certain issues, LLM-agents identify them. These findings demonstrate the potential of LLMs as a preparatory step in usability testing and outline directions for advancing their adaptation and improvement.
Vibe Design: Human-in-the-loop AI Agents for UI Design with Large Language Models
(2026-01-06) Hwang, Ji Sun; Kang, Juyoung
New tectonic shifts are looming in the UI design process due to advances in LLMs. These shifts have led researchers to adopt LLM-based approaches to develop design systems. However, such systems often focus on narrow functions and rely on code-based data that limits real-world applicability. We propose Vibe Design, an agentic framework that enables designers to co-create and evaluate prototypes with the LLM-based agents. The framework consists of a design agent that supports UI creation and a user testing agent that generates usability feedback through simulated user interviews, guided by the Persona-Scenario-Goal methodology. We also incorporated a human-in-the-loop mechanism to enhance the reliability and quality of the LLM-generated responses. We successfully generated functional prototypes and extracted user requirements from diverse personas. This study provides insights into the use of LLMs for both UI generation and user feedback collection, offering practical implications for future design workflows.
Caught Between Compliance and Resistance: Understanding Users’ Conflicted Responses to Confirmshaming Techniques
(2026-01-06) Sha, Kaixin; Xu, Jingjun (David); Wang, Qi
Despite the widespread use of confirmshaming in digital interfaces, prior research has largely focused on its impact on user compliance, with limited attention to its negative business consequences and underlying psychological mechanisms. To address this gap, we draw on dual-process theory and psychological reactance theory to theorize how confirmshaming produces competing psychological effects through two distinct pathways: experienced coercion and financial cost perception, which exert differential influences on customer word of mouth (WOM) intention. Our online experiment shows that confirmshaming increases customer acceptance of promotional offers by amplifying financial cost perception, while simultaneously reducing positive WOM intention through increased experienced coercion. These dual mechanisms reveal a critical trade-off: compliance gains come at the expense of customer advocacy. Theoretically, our study contributes to IS literature on digital nudging by demonstrating that reactance responses compete with rational decision-making processes. Our findings suggest that platforms can reconsider the use of manipulative designs, as short-term benefits may be offset by reduced customer WOM intention.
Comfort with Uncertainty Predicts Acceptance of Algorithmic Advice
(2026-01-06) Schecter, Aaron; Bogert, Eric; Lauharatanahirun, Nina
This study explores the influence of individual differences in uncertainty tolerance and source preference on the acceptance of algorithmic advice. By integrating economic lottery choice tasks with decision-making experiments, we examine how comfort with uncertainty shapes behavioral tendencies toward algorithmic or human advisors. Across three tasks varying in objectivity-—crowd counting, remote associates, and caption generation-—we identify an overall algorithmic appreciation effect, where participants favor algorithmic advice, particularly in objective tasks. Importantly, the degree of algorithmic appreciation is contingent upon uncertainty tolerances and source preferences. Further, we identify significant heterogeneity over tasks. Our findings suggest that individual psychological factors, such as a preference for non-social uncertainty sources, predict algorithmic advice acceptance.
Even Better than the Real Thing: How Imperfection Shapes Trust and Engagement with Digital Humans
(2026-01-06) Tang-Ear, Johnny; Seymour, Mike; Dennis, Alan; Yuan, Lingyao; Hardy, Catherine
This study challenges the assumption that more realism in digital humans always leads to greater trust and engagement. Using eye-tracking and post-exposure surveys, we compared viewer responses to three video presenters: a highly realistic digital human, a real human, and an imperfect altered human, represented by a real presenter altered to have unblinking eye contact. While participants rarely noticed visual imperfections consciously, the human with subtle flaws led to significantly greater trust and willingness to pay. The imperfect video outperformed the fully realistic, unaltered human video, suggesting that perfect realism may not always be best. These findings offer important implications for the design of AI-driven digital humans, highlighting that strategic imperfection can enhance authenticity, trust, and engagement in customer interactions. Moreover, the results contribute new empirical insights into the Uncanny Valley theory, suggesting that user affinity and trust may peak not at perfect realism, but can peak at a point just prior to the full realism.
Evolving Intention in Digital Activities and Its Relationship to Productivity and Immersion
(2026-01-06) Kim, Jiyoon; Toyama , Kentaro
User intention plays a key role in digital activity, but it is less studied than other aspects of technology use such as productivity. We report on a mixed-methods study with university students in two countries (United States and South Korea) in which two distinct forms of intention emerged: the degree of (1) "starting intention" involved in beginning an activity, and the subsequent (2) "continuing intention" to continue the activity. Predictably, an activity's continuing intention often matches its starting intention, and in work activities, higher overall intention correlates with productivity. However, unexpected findings also surfaced: for example, intention's relationship with productivity varies by activity type; breaks started with low intention can be immersive; and work activities started with low intention are sometimes considered productive. These and other findings add to our understanding of digital productivity and recreation, and suggest new designs for technologies to support digital well-being.
Simplicity Gone Wrong: Revisiting Implicit Designs for Older Adults
(2026-01-06) Wang, Hong-Lin; Choudhry , Abhinav; Adler, Rachel F.; Zhou, Kyrie Zhixuan
This study investigated how removing implicit design elements, such as gesture-based or symbol-based interactions, affects the usability and learnability of mobile applications for older adults. We used Line, a popular social media platform in Taiwan, for a case study and created a prototype version with implicit designs removed for comparison. We collected data from 16 older adult participants and primarily focused on their verbal responses and task completion performance. Our findings suggest that replacing symbol-based designs with text can improve user experience, while replacing gesture-based designs did not result in better usability. We also found that how the user explores the application greatly impacts whether implicit design affects their experience. This study provides insights into how users interact with different kinds of implicitness and provides a case study of a less frequently reviewed application.
Introduction to the Minitrack on Human-Computer Interaction
(2026-01-06) Jenkins, Jeffrey; Nah, Fiona Fui-Hoon; Valacich, Joseph; Schneider, Christoph

Browse

Recent Submissions