International Journal of Human-Computer Studies最新文献

{"title":"Self-avatar-supported observational learning: Designing and evaluating VR-based physical exercise tutorial systems for older adults","authors":"Ruitong Che ,&nbsp;Jiaan Li ,&nbsp;Chi Deng ,&nbsp;Qian Mao ,&nbsp;Jeffrey C.F. Ho ,&nbsp;Fiona Nah ,&nbsp;Calvin K.L. Or ,&nbsp;Hailiang Wang","doi":"10.1016/j.ijhcs.2026.103826","DOIUrl":"10.1016/j.ijhcs.2026.103826","url":null,"abstract":"<div><div>The global trend toward longer life spans presents an opportunity to promote active and healthy aging. Physical exercises like Qigong support holistic well-being by integrating physical, cognitive, and emotional health. However, traditional programs lack adaptability to accommodate age-related changes in physical and cognitive abilities, which can limit accessibility and engagement for older adults. Virtual Reality (VR) offers a novel solution by creating immersive, customizable environments. In our study, we designed a VR-based physical exercise tutorial (VRPET) system and assessed the efficacy of using an adaptive self-avatar (i.e., a virtual representation of the user) to enhance user exercise performance and attitudes, while examining its impact on perceived workload. We conducted a two-phase mixed-methods investigation: (1) A formative phase involving focus group interviews (n=14), a consultation with a Qigong master on movement standardization, and a heuristic evaluation (n=5) to establish design requirements; (2) A user study (n = 30) that compared the self-avatar versus non-self-avatar conditions to assess their effects on perceived workload, exercise performance, and attitude metrics. Despite a significant dip in exercise performance (<em>p</em>=0.03) and a non-significant increase in perceived workload (<em>p</em>=0.58), participants expressed a preference for the self-avatar’s real-time feedback when scaffolded appropriately. Multi-modal analysis revealed auditory cues as most effective, followed by tactile and visual feedback. Based on these findings, we propose the ACT Framework (Adaptive, Cultural, Targeted) for developing age-appropriate VR exercise systems. Furthermore, we distill our iterative process into a tripartite validation workflow, advocating for a methodology that harmonizes user desirability, expert safety, and HCI usability. These evidence-based insights advance the design of therapeutic VR interventions that can support healthy aging populations.</div></div>","PeriodicalId":54955,"journal":{"name":"International Journal of Human-Computer Studies","volume":"212 ","pages":"Article 103826"},"PeriodicalIF":5.1,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147799131","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Guilty apology and trust repair in generative artificial intelligence: the role of mind perception","authors":"Fang Xu ,&nbsp;Zihui Yuan ,&nbsp;Siyu Jin ,&nbsp;Gengfeng Niu ,&nbsp;Zongkui Zhou","doi":"10.1016/j.ijhcs.2026.103813","DOIUrl":"10.1016/j.ijhcs.2026.103813","url":null,"abstract":"<div><div>With the rapid proliferation of Generative AI (GenAI) in daily life, errors and service failures are inevitable, making effective trust repair a pressing concern. This study, grounded in the theory of mind perception, examines the influence of guilty apologies on trust repair after AI errors and the underlying mechanisms (perceived experience and perceived agency). Through two human-AI interaction experiments, the results show that: Experiment 1 found that, compared with neutral apologies, guilty apologies not only significantly promoted trust repair but also enhanced perceived experience, with no significant effect on perceived agency. Experiment 2 further manipulated individuals’ perceived experience and found that it positively influenced trust repair following AI errors. In the high perceived experience condition, guilty apologies were more effective in trust repair than neutral apologies, while in the low perceived experience condition, the effect of guilty apologies on trust repair was not significant. The MMM design confirmed the mediating path of “guilty apology → perceived experience → trust repair.” This study deepens the theoretical understanding of human-AI trust repair and provides empirical support for the emotional design of AI dialogue systems, showing that apologies with a sense of “guilt” effectively enhance user tolerance and relationship repair.</div></div>","PeriodicalId":54955,"journal":{"name":"International Journal of Human-Computer Studies","volume":"212 ","pages":"Article 103813"},"PeriodicalIF":5.1,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147799130","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Trust the Explanation or my Expectation? Effects of Output Accuracy and Explanations on Expectation Violations and Trust in AI-Supported Decisions","authors":"Tim Hunsicker ,&nbsp;Isabel Duhl ,&nbsp;Pascal Haubert ,&nbsp;Linda Onnasch ,&nbsp;Markus Langer","doi":"10.1016/j.ijhcs.2026.103775","DOIUrl":"10.1016/j.ijhcs.2026.103775","url":null,"abstract":"<div><div>Systems based on Artificial Intelligence (AI) increasingly support decision-making, but their outputs may be inaccurate. Prior research has suggested that explanations might help detect inaccuracies, aiding successful human-AI interaction. This study investigates how the accuracy of system outputs influences users’ trust, trusting behavior, and trustworthiness perceptions, the role of expectation violations in this process, and how explanations for the system outputs influence these effects. In an online study with a 2(explanation vs. no explanation) × 2(accurate vs. inaccurate outputs) between-within design, 218 participants evaluated six job applicants. They received CVs and algorithmic evaluations of applicants’ suitability. For three applicants, outputs were accurate; for the other three, outputs reflected a 40% lower suitability than their true suitability. Half of the participants received explanations. Accurate outputs led to higher trustworthiness, trust, and trusting behavior than inaccurate outputs. Expectation violation fully mediated how accuracy affected trust and trustworthiness, and partially how accuracy influenced trusting behavior. Moreover, there was a significant interaction between explanations and output accuracy concerning trusting behavior: when outputs were accurate, explanations had little effect on trusting behavior; however, when outputs were inaccurate, explanations led to stronger trusting behavior, as participants less strongly deviated from the inaccurate outputs. We conclude that users are able to deviate from inaccurate outputs, and we highlight the importance of expectation violations in this regard. However, our findings also show possible detrimental effects of explanations as they can increase the decisional weight of inaccurate outputs instead of facilitating the detection of inaccuracies.</div></div>","PeriodicalId":54955,"journal":{"name":"International Journal of Human-Computer Studies","volume":"211 ","pages":"Article 103775"},"PeriodicalIF":5.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147601743","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sense of agency in cooperative gestures: The effect of hierarchical roles","authors":"Xinpei Zheng ,&nbsp;Olivier Chapuis ,&nbsp;Ouriel Grynszpan","doi":"10.1016/j.ijhcs.2026.103803","DOIUrl":"10.1016/j.ijhcs.2026.103803","url":null,"abstract":"<div><div>Cooperative gestures are gestures performed by multiple users to facilitate collaboration on large interactive surfaces, such as tabletops. While previous research suggested that such gestures enhance user experience, they overlooked important aspects of user experience and, in particular, the users’ sense of agency. This paper considers two types of cooperative gestures: (i) egalitarian, where both users have the same role, and (ii) hierarchical, where one user (the leader) leads the action, and the other user (the follower) follows the leader’s actions. In a user experiment, we evaluate these types of cooperative gestures by measuring the participants’ sense of agency using intentional binding and judgment of control. Results show a decrease in the sense of agency for the followers. By contrast, leaders and users with egalitarian roles maintained their agency at the implicit sensorimotor level. We conclude with “implications for design” of these findings.</div></div>","PeriodicalId":54955,"journal":{"name":"International Journal of Human-Computer Studies","volume":"211 ","pages":"Article 103803"},"PeriodicalIF":5.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147601746","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Acted sleepy speech vs. real sleepy speech: Human perception and machine prediction for sleepiness estimation","authors":"Jihye Moon ,&nbsp;Youngsun Kong ,&nbsp;Jeffrey Bolkhovsky ,&nbsp;Yashvi Gupta ,&nbsp;Ki H. Chon","doi":"10.1016/j.ijhcs.2026.103795","DOIUrl":"10.1016/j.ijhcs.2026.103795","url":null,"abstract":"<div><div>As agentic AI systems increasingly operate in high-stakes human-centered environments, their ability to detect physiological states such as sleepiness is critical for mitigating health and safety risks. Accurately estimating sleepiness from speech is essential for managing these risks, including adverse health outcomes, reduced productivity, and occupational safety hazards. For example, chronic sleep deprivation increases the risk of cardiovascular disease and impairs cognitive performance. Therefore, accurately determining sleepiness holds significant value for both health monitoring and workplace risk management. Since speech serves as a primary interface for human-AI interaction, estimating sleepiness from speech has emerged as an efficient way to mitigate these risks through AI. However, current machine learning (ML) studies have struggled to estimate sleepiness from speech, showing weak Spearman correlations (below 0.40) with perceived sleepiness levels (ground-truth sleepiness). As ML performance depends on high-quality training data, obtaining speech from noticeably sleepy individuals is essential for its improvement. However, collecting such data is risky and costly, as it requires physiologically sleep-deprived speakers with at least 24 h of prolonged wakefulness. Given that humans can mimic sleepy speech, this paper proposes that acted sleepy speech can be an effective surrogate for modeling sleepiness patterns. Our study demonstrates two key findings: First, human listeners rated acted sleepy speech as significantly sleepier than real sleepy speech obtained from a 25-hour sleep deprivation protocol, as confirmed by Welch’s <em>t</em>-test. Second, ML models trained on acted speech significantly outperformed those trained on real sleepy speech, achieving a 0.57 correlation with perceived sleepiness levels and 0.83 accuracy in detecting cognitive impairments in sleep-deprived individuals awake for 25 h, even with fewer samples and different lexical transcripts, indicating robustness to lexical variation. This human-centered, ethical, and efficient approach demonstrates that acted sleepy speech can advance real-world speech-based sleepiness estimation and cognitive impairment detection systems. Ultimately, it offers a promising pathway to real-world solutions for reducing sleepiness-related risks in the workplace, improving health outcomes, and enhancing the ability of human-centered agentic AI to support daily human tasks.</div></div>","PeriodicalId":54955,"journal":{"name":"International Journal of Human-Computer Studies","volume":"211 ","pages":"Article 103795"},"PeriodicalIF":5.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147657581","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sentiment-based robot sonification can enhance comprehension, emotional meaning and engagement in the fairy tale listening: A preliminary study","authors":"Yeaji Lee ,&nbsp;Sophie Thompsen ,&nbsp;Ziming Fang ,&nbsp;Myounghoon Jeon","doi":"10.1016/j.ijhcs.2026.103805","DOIUrl":"10.1016/j.ijhcs.2026.103805","url":null,"abstract":"<div><div>Sonification provides highly informative data, but its interpretation can vary depending on the context and the audience’s capability. Specifically, sonification can play a key role in storytelling experience. The current study examined how sentiment-based sonification vs. classical music vs. no sonification influence the experience of robot storytelling, especially in terms of audience members’ comprehension, emotional meaning, and engagement in such an environment. We analyzed three fairy tales using narrative analysis principles and a sentiment analysis algorithm to design sentiment-based sonification based on classical music. Thirty-four participants experienced three stories with emotion-reflecting sonification, classical music, or no sonification (within-subjects design) with a humanoid robot, NAO, as a storyteller. The results showed that the ratings for comprehension, emotional meaning, and engagement were significantly improved in the sonification condition compared to the other conditions. Implications are discussed with various models and theories, including the ITPRA theory of expectation, the capacity model of attention, and the limited capacity theory, among others.</div></div>","PeriodicalId":54955,"journal":{"name":"International Journal of Human-Computer Studies","volume":"211 ","pages":"Article 103805"},"PeriodicalIF":5.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147657580","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Editorial for the special issue on child-centred AI","authors":"Jun Zhao ,&nbsp;Grace Lin ,&nbsp;Jason Yip ,&nbsp;Zhen Bai ,&nbsp;Ayca Atabey ,&nbsp;Ge Wang ,&nbsp;Kaiwen Sun","doi":"10.1016/j.ijhcs.2026.103777","DOIUrl":"10.1016/j.ijhcs.2026.103777","url":null,"abstract":"","PeriodicalId":54955,"journal":{"name":"International Journal of Human-Computer Studies","volume":"211 ","pages":"Article 103777"},"PeriodicalIF":5.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147747291","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exploring a user-centered approach for movement-based features in interaction design","authors":"Antonio Escamilla ,&nbsp;Javier Melenchón ,&nbsp;Carlos Monzo ,&nbsp;Jose Antonio Morán ,&nbsp;Juan Pablo Carrascal","doi":"10.1016/j.ijhcs.2026.103802","DOIUrl":"10.1016/j.ijhcs.2026.103802","url":null,"abstract":"<div><div>Far beyond very fine-grained, accurate data, movement-based interaction design often benefits from working with higher-level features to create engaging interactions. Yet, movement-based frameworks most often focus on the detection strategy without providing clarity on how to use such data for the design of user interactions. This purely technical approach makes it challenging for practitioners to explore feature extraction technology as a design material. This paper presents an approach that considers designers’ perspectives and attitudes about using movement-based features to propose a set of <em>designer-interpretable</em> descriptors and enhance their ability to use human motion in interaction design. In addition, a computational prototype is utilized to visually present the features and help designers better understand movement content. The investigation into the effects of <em>designer-interpretable</em> features on the concept ideation and design of motion-based interactions followed a user-centered approach, and its validity was established through a qualitative study conducted with interaction designers. Semi-structured interviews following a creative practice exercise were employed to evaluate how the computational prototype influenced interaction designers’ processes, allowing them to reflect on their experience and compare it to previous approaches when designing movement-based interactions. The study determined that interaction designers were able to leverage features to identify detection capabilities and enrich the ideation process. Moreover, feature visualization provided further insights into the characteristics of movement, which helped practitioners understand the interaction opportunities that come with it.</div></div>","PeriodicalId":54955,"journal":{"name":"International Journal of Human-Computer Studies","volume":"211 ","pages":"Article 103802"},"PeriodicalIF":5.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147601745","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Children’s communication repairs with AI versus human partners","authors":"Zhixin Li ,&nbsp;Trisha Thomas ,&nbsp;Chi-Lin Yu ,&nbsp;Ying Xu","doi":"10.1016/j.ijhcs.2026.103800","DOIUrl":"10.1016/j.ijhcs.2026.103800","url":null,"abstract":"<div><div>Children’s interactions with artificial intelligence (AI) are growing, yet communication breakdowns—instances where mutual understanding fails—remain a challenge, especially for young children. While generative AI shows promise in engaging children in open-ended conversations, how children navigate and repair these communication breakdowns remains unclear. This study compares how 78 children, aged four to eight years, managed communication breakdowns and repair strategies while co-creating stories with an AI agent (powered by a large language model) versus a human counterpart. Results reveal that the type of conversational partner—human or AI—significantly influenced children’s repair behaviors. Children experienced more communication breakdowns when interacting with the AI partner but attempted repairs more frequently with the human counterpart. Misunderstandings and mishearings are the most frequent causes, with clarification requests as the primary repair strategy in both cases. However, when interacting with the AI, children were more likely to go along with the conversation flow to compensate for AI errors, even when the dialogue deviated from their intended meaning—a pattern not observed with human partners. Additionally, children’s social perceptions of their partner, especially beliefs about emotional capacity and their psychological closeness to their partner, influenced repair attempts. This study expands research on children’s conversational repairs with AI, shedding light on the role of social dynamics in shaping these interactions.</div></div>","PeriodicalId":54955,"journal":{"name":"International Journal of Human-Computer Studies","volume":"211 ","pages":"Article 103800"},"PeriodicalIF":5.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147601744","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Under what influence: Measuring AI influence to fit user profiles in decision-making","authors":"Andrea Campagner ,&nbsp;Caterina Fregosi ,&nbsp;Chiara Natali ,&nbsp;Federico Cabitza","doi":"10.1016/j.ijhcs.2026.103763","DOIUrl":"10.1016/j.ijhcs.2026.103763","url":null,"abstract":"<div><div>Artificial Intelligence (AI) has become a pivotal tool in augmenting human decision-making across various domains, yet its influence on user decisions often lacks comprehensive evaluation. While technical performance metrics such as accuracy and efficiency dominate AI design, integrating human-centered approaches that consider trust and reliance remains underexplored. This study addresses the knowledge gap in understanding how AI systems influence decision-making quality, calibrated to user profiles, including their expertise, skills, professional role, confidence, and reliance tendencies.</div><div>We present a novel and comprehensive metric framework for evaluating AI influence, emphasizing behavioral patterns and measurable improvements in decision outcomes beyond simple alignment with AI recommendations. The framework is applied to four medical domain case studies—MRI, ECG, X-ray, and ENDO – with user groups spanning specialists, sub-specialists, and trainees. Results reveal that while human and AI systems achieve high agreement rates (up to 81%), AI influence on decision quality varies significantly. Notably, X-ray decision-making showed the highest influence index (0.27), while MRI decisions exhibited substantial self-anchoring bias (6.94), undermining the potential positive impact of AI. Influence metrics unveiled nuances missed by agreement scores, highlighting domain-specific biases and opportunities to optimize AI-human interaction.</div><div>This research underscores the necessity of adapting the type of AI system and affordances to user characteristics and attitudes of reliance to foster calibrated trust and improve decision outcomes. Our findings inform the design of AI systems that better support diverse user needs and align with human decisions, driving progress toward human-centered AI integration in high-stakes domains.</div></div>","PeriodicalId":54955,"journal":{"name":"International Journal of Human-Computer Studies","volume":"210 ","pages":"Article 103763"},"PeriodicalIF":5.1,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147402120","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}