The study arrives against a backdrop of dramatic change. A UK-wide survey of 1,041 undergraduates in early 2025 found that 88 percent of students were using generative AI for assessment purposes, up from just 53 percent a year earlier. In healthcare, generative AI is already being explored for diagnosis support, disease risk prediction, patient communication and documentation automation. That raises an obvious question for medical educators: if tomorrow’s doctors will work in AI-enabled hospitals, shouldn’t medical schools be teaching them how to use these systems wisely today?

To find out what is really happening on the ground, the researchers ran an exploratory sequential mixed-methods pilot. They began with three focus groups, one each with preclinical students, clinical students and faculty members, and used reflexive thematic analysis of the transcripts to build a questionnaire. That survey was then distributed to all undergraduates on the five-year MBChB programme and to teaching staff at the school, yielding valid responses from 29 students and 32 faculty members between January and February 2025. The team anchored their analysis in established technology acceptance frameworks, the Technology Acceptance Model and the Unified Theory of Acceptance and Use of Technology, which hold that perceived usefulness and perceived ease of use, shaped by factors such as training and support, drive whether people actually adopt a new tool.

The headline finding is that generative AI is already woven into medical education, but thinly and unevenly. Across all respondents, average use frequency was just 1.08 on a scale from 0 (never) to 4 (daily), equivalent to rare use. Nevertheless, 58.6 percent of students and 40.7 percent of faculty reported using generative AI to some extent in a medical education context. Students most commonly deployed it to summarise specific topics, cited by 76.5 percent of student users, and to generate practice questions, exam stations or simulated medical cases, each reported by 41.2 percent. Faculty leaned on it mainly for improving their own writing, at 53.8 percent, and preparing teaching materials, at 38.5 percent. Notably, no faculty member reported using it to write feedback on student assignments or to mark them.

Confidence, however, lags far behind adoption. Respondents rated their overall self-confidence in AI-related knowledge and skills at just 1.38 on a 0 to 4 scale, sitting between slightly confident and moderately confident, with no statistically significant difference between students and staff. Among those who had never used the technology, the two groups diverged sharply in their reasons: 58.3 percent of non-using students said they simply did not trust generative AI, a figure significantly higher than the 15.8 percent of faculty who gave the same answer. Faculty non-users instead pointed to practical barriers, with 47.4 percent reporting no relevant training, 42.1 percent saying they had no time to explore the tools, and roughly a third of both groups unsure about the regulations governing acceptable use.

When it came to what the technology can do, respondents on average slightly agreed that generative AI has genuine capabilities in medical education, with time-saving the most strongly endorsed item at a mean of 3.92 on a five-point agreement scale. Students were significantly more positive than faculty about AI’s potential to make educational resources more financially accessible, echoing wider research suggesting that AI-generated simulations and virtual patients could lower cost barriers for schools and learners. But many participants insisted that effectiveness depends entirely on how the tools are applied. One faculty respondent argued that the question is less about the capabilities of generative AI and much more about the design of the processes it is involved in, noting that a badly designed learning activity remains badly designed whatever technologies are brought to bear on it. Another observed that educators and students have started using it without necessarily understanding it or having guidelines and boundaries for appropriate use.

The concerns were louder than the enthusiasm. On average, respondents agreed that generative AI carries substantial limitations, challenges and ethical risks, with a mean of 3.93, and inaccuracy, lack of transparency and bias topped the list. Faculty held significantly stronger concerns than students overall, and were particularly worried about inequitable student access to the technology and the opacity of training data and algorithms. Several respondents warned that AI could become a crutch that erodes creativity, discourages independent study and prevents students from developing problem-solving skills. One student captured the unease vividly, describing generative AI as having passed the point of being a tool and become instead a machine that carries out tasks for us rather than with us, with the potential to damage much of what makes human learning and even culture special. Others flagged the technology’s shallow grasp of scientific literature and its limited emotional intelligence, with one respondent noting that AI used exclusively for medical education would not effectively teach empathy and humane clinical practice, although some participants conceded that newer models show improving sensitivity on sensitive topics.

Assessment emerged as the most contested battleground. Many respondents said they could not yet judge whether AI’s impact on medical education was positive or negative, and both groups perceived a slightly negative effect on most aspects of the curriculum, with the exception of curriculum content, teaching methods and career preferences. Faculty worried that students could use AI to write assessments and go undetected, undermining the validity of qualifications in a profession where knowledge is a key requisite for being a doctor, and one student claimed that every assessment without intense invigilation or tab-blocking was already at risk. Suggestions for reform included making assessments more authentic and rethinking how critical thinking can be promoted and evaluated in ways that cannot be circumvented by AI, though one faculty member cautioned that academics feel unconfident and unequipped to design authentic assessments that use AI explicitly. Another respondent offered a provocatively optimistic spin, arguing that some of the problems AI poses for assessment are overdue spurs to innovate beyond long-standing, lazy approaches to examining students.

One of the study’s most striking results concerns gender. Male respondents used generative AI significantly more often than females, reported markedly higher confidence in their AI knowledge and skills, expressed fewer concerns about its limitations and ethical challenges, and held more positive attitudes towards its impact on medical education. The researchers note that such differences echo previous findings of a gender gap in AI adoption and argue that larger studies should investigate whether the technology’s growing embedment in medical education risks creating unequal access and unequal benefit across subgroups of future doctors.

What both groups agreed on, emphatically, was the need for support. Students most wanted open communication with educators about AI use, cited by 62.1 percent, followed by clearer institutional regulations and guidance, at 58.6 percent, and student training sessions, at 55.2 percent. Faculty demands were even more emphatic: 90.6 percent called for training for themselves, and majorities also wanted open dialogue with students, clearer policies, student training and advice from national advisory bodies, with faculty expressing significantly greater need for these resources than students did. Several faculty respondents went further, describing a cultural problem in which some colleagues deny that students are using AI at all and design assessments for a context that no longer exists, while stigma against students who use AI may push them to misuse it secretly rather than engage with it openly. The study’s authors conclude that ensuring ethical, effective and equitable use of generative AI in medical education will require strategic support from schools, universities and national regulators, and they call for national-scale research to track how this rapidly shifting landscape evolves. As one faculty respondent put it, generative AI is one of the biggest challenges to the way universities operate in a long time, and it will be the leadership and vision of institutions that determine whether its disruptive injection of change turns out to be for the good or destructive.

Subject of Research: Student and faculty experiences and perceptions of generative artificial intelligence in UK medical education

Article Title: Exploring student and faculty experiences and perceptions of generative artificial intelligence in medical education

Article References: Ding, H., & Homer, M. (2026). Exploring student and faculty experiences and perceptions of generative artificial intelligence in medical education. Discover Education, 5(1), Article 832. https://doi.org/10.1007/s44217-026-02106-4

Image Credits: AI Generated

DOI: 10.1007/s44217-026-02106-4

Keywords: generative artificial intelligence, medical education, undergraduate medical students, faculty perceptions, technology acceptance model, UTAUT, assessment integrity, AI training, gender differences, ethical challenges, UK medical school, mixed methods

The research team, led by Junfang Shi, Zeyu Li, and Peifen Ma of the School of Nursing at Lanzhou University and the Second Hospital of Lanzhou University, set out to systematically map why some patients embrace these bedside terminals while others abandon them. Their approach was deliberately grounded in evidence rather than speculation. In September 2024, the researchers conducted semi-structured interviews with thirteen hospitalized inpatients aged 22 to 73 years, nine men and four women, all of whom had prior experience using the Smart Bedside Station system at their hospital. Participants were recruited through purposive sampling, meaning the team deliberately selected patients with direct hands-on experience of the technology rather than relying on convenience. To structure the inquiry, the researchers turned to one of the most influential frameworks in technology adoption research: the Unified Theory of Acceptance and Use of Technology, or UTAUT.

UTAUT, first formalized by Venkatesh and colleagues in 2003, distills decades of technology acceptance research into four core constructs that predict whether a person will actually use a new system. Performance expectancy captures the belief that the technology will help the user accomplish something valuable. Effort expectancy reflects how easy the user perceives the system to be. Social influence describes the pressure or encouragement a user feels from important others, such as clinicians or family members. Facilitating conditions encompass the environmental supports, training, and infrastructure that make use possible. The Lanzhou team used these constructs to design their interview guide and then applied a hybrid thematic analysis, combining deductive coding against the UTAUT framework with inductive coding that allowed unexpected themes to surface from the raw interview data. What emerged was a remarkably detailed picture of the psychology of bedside technology adoption: ten distinct facilitators and fifteen distinct barriers to system use.

The facilitators tell a clear story about what makes patients reach for the bedside screen. Eleven of the thirteen participants first learned about the system from nurses, with only two discovering it through self-exploration, and nurse instruction and encouragement emerged as a dominant social influence driving initial engagement. Trustworthy educational videos delivered through the terminal supported performance expectancy, giving patients confidence that the health information they were watching was clinically reliable rather than internet noise. Transparent access to personal expenses and medical reports strengthened the perception that the system offered real, tangible value. Convenient call functions and entertainment options rounded out the appeal. Meanwhile, a simple interface reduced effort expectancy, making the system feel approachable even to patients who were not technologically confident. In UTAUT terms, the system’s clinical utility, nursing guidance, and environmental supports were doing the heavy lifting of adoption.

But the barriers, fifteen of them, reveal why sustained use collapses so quickly. The most fundamental obstacle was limited perceived added value: many patients simply saw the bedside terminal as redundant when they already carried a personal smartphone capable of similar functions. Treatment-time constraints meant that hospital schedules packed with medications, procedures, and rounds left little cognitive or temporal room for exploring a new interface. Information and feature overload overwhelmed users who confronted too many options at once. Delayed or lagging updates frustrated patients who expected the responsiveness of consumer technology. Interface misperception led some users to misunderstand what the system could actually do. And perhaps most tellingly, fear of misoperation or unintended charges made patients hesitant to touch anything at all, worried that a wrong tap might trigger a fee or disrupt their care. Older age and lower educational attainment further compounded these frictions, echoing a well-documented digital divide in eHealth research.

The usage statistics embedded in the study underscore how shallow engagement often was. Use was short-term for most participants, and only 15.4 percent of the sample reported using the system for more than thirty days. This is a critical finding for anyone designing patient-facing hospital technology. Initial exposure, often brokered by a nurse during admission, was sufficient to spark curiosity, but without perceived added value beyond the patient’s own phone, and without system reliability and workflow compatibility, that curiosity rarely matured into habitual use. The pattern mirrors findings from other health information technology evaluations, including mixed-method studies of health information exchange in England and qualitative research on digital devices for home use, where technology acceptance hinges on the same balance of usefulness, ease, social endorsement, and environmental support.

What makes this study technically important is its theoretical mapping. By anchoring facilitators and barriers to the four UTAUT constructs, the researchers transformed anecdotal patient complaints into a structured framework that hospital administrators and software designers can act upon. Nurse instruction and encouragement map to social influence, suggesting that nursing workflows should formally incorporate bedside system onboarding rather than leaving it to chance. Trustworthy video education and transparent expense access map to performance expectancy, indicating that content quality and data transparency are not cosmetic features but core adoption drivers. A simple interface maps to effort expectancy, reinforcing the principle that patient-facing clinical systems must be radically simpler than consumer apps. Facilitating conditions, from training to system maintenance, emerged as the environmental scaffolding that determines whether any of the other drivers can operate at all.

The barriers, mapped against the same framework, expose the cognitive and workflow burdens that constrain adoption. Information overload and feature overload suggest that designers should resist the temptation to pack every possible function into the bedside terminal, a lesson consistent with research on information retention in first-time hearing aid users, where multimedia education outperformed dense verbal counseling. Fear of misoperation and unintended charges points to a trust problem that requires explicit reassurance, perhaps through clear on-screen confirmation of free versus paid functions. Lagging updates and system frictions indicate that facilitating conditions were not fully met, meaning that maintenance, content freshness, and technical support are as important as interface design. Treatment-time constraints remind us that patient engagement technology must fit within the rhythms of clinical care, not compete against them.

The implications ripple outward to the broader smart hospital movement. As hospitals worldwide invest in digital transformation, from electronic health records to IoT-enabled patient monitoring platforms, the Lanzhou findings offer a cautionary technical lesson: hardware installation is not adoption. A bedside terminal bolted to a wall is merely infrastructure; adoption requires nursing guidance, perceived clinical utility, cognitive simplicity, and system reliability. The researchers conclude that addressing barriers and enhancing system features is essential to improve patient engagement and satisfaction, ultimately facilitating wider adoption of Smart Bedside Station systems in smart hospitals. For a field racing to digitize the patient experience, the message is clear. The most important interface in a smart ward is not the screen at all. It is the conversation between a nurse and a patient, and the technology that survives will be the technology that nurse chooses to recommend.

Subject of Research: Facilitators and Barriers of Using Smart Bedside Station Systems Among Hospitalized Patients: A Qualitative Study Based on UTAUT

Article Title: Facilitators and Barriers of Using Smart Bedside Station Systems Among Hospitalized Patients: A Qualitative Study Based on UTAUT

Article References: Shi, J., Li, Z., Dai, Y., Chen, X., Yang, L., Wang, X., & Ma, P. (2026). Facilitators and Barriers of Using Smart Bedside Station Systems Among Hospitalized Patients: A Qualitative Study Based on UTAUT. Journal of General Internal Medicine. https://doi.org/10.1007/s11606-026-10663-5

Image Credits: AI Generated

DOI: 10.1007/s11606-026-10663-5

Keywords: smart bedside station, UTAUT, patient engagement, smart hospital, health information technology, qualitative research, Facilitators, Barriers, Smart, Bedside, Station, Systems

The research team built its conceptual model on two well-established theoretical pillars. The first is the Unified Theory of Acceptance and Use of Technology, known as UTAUT, which explains technology adoption through four core determinants: performance expectancy, the belief that a tool improves one’s performance; effort expectancy, the perceived ease of using it; social influence, the sense that important others expect one to use it; and facilitating conditions, the perception that infrastructure and support exist to make use possible. The second pillar is Social Cognitive Theory, which frames human behavior as the product of personal beliefs, environmental factors, and actions interacting in a continuous loop. By fusing these frameworks, the researchers could trace a full causal pathway: from the psychological and contextual drivers of generative AI adoption, through actual use of the tools, and onward to observable innovative behavior, with mediating variables operating in the middle of the chain.

Data came from an online survey administered in China, which yielded 291 valid responses from Generation Z students in higher education. The sample size is modest by big-data standards but well suited to the analytical strategy the team deployed. Ethics approval was granted by the Academic Committee of the School of Economics and Management at the Institute of Disaster Prevention in February 2025, and informed consent was obtained from all participants, who were told their involvement was voluntary and that they could withdraw at any time. The survey instrument captured students’ perceptions across the UTAUT determinants, their actual generative AI use, their knowledge-sharing habits, their creative self-efficacy, and their self-reported innovative behavior, allowing the researchers to model the entire adoption-to-innovation pathway in a single framework.

The analytical approach is one of the study’s most distinctive technical contributions. Rather than relying on a single statistical technique, the team executed a dual-stage analysis combining partial least squares structural equation modeling, or PLS-SEM, with an artificial neural network, or ANN. PLS-SEM is a variance-based technique that estimates networks of simultaneous regression equations linking latent constructs, which are abstract psychological variables like performance expectancy that cannot be observed directly but are inferred from multiple survey indicators. This makes PLS-SEM ideal for testing whether a hypothesized causal structure fits the data. Its limitation, however, is that it assumes the relationships between constructs are linear and additive. Human psychology rarely cooperates with such assumptions, and that is precisely where the neural network enters the picture.

Artificial neural networks are machine-learning models inspired loosely by the architecture of biological brains. They consist of layers of interconnected nodes that transform inputs into outputs through weighted connections and nonlinear activation functions, allowing them to capture complex, nonlinear patterns that classical regression equations miss. In this study, the ANN served as a complementary second stage: after PLS-SEM confirmed which pathways were statistically significant, the neural network ranked the relative importance of the predictors of innovative behavior without imposing linearity constraints. This dual-stage design, pairing a theory-driven structural model with a data-driven learning model, is increasingly seen as best practice in behavioral research, because it lets researchers test causal theory while also letting the data reveal nonlinear effects and importance rankings that the structural model alone would conceal.

The PLS-SEM results affirmed all four UTAUT determinants as positive drivers of generative AI adoption. Performance expectancy, effort expectancy, social influence, and facilitating conditions each significantly increased the likelihood that students adopt these tools, which is consistent with the broader technology-acceptance literature but notable in the generative AI context, where the tools are conversational, iterative, and qualitatively different from earlier educational software. More consequentially, actual generative AI use significantly enhanced students’ innovative behavior, confirming that adoption is not merely a matter of convenience but a genuine catalyst for generating, promoting, and realizing new ideas in academic work. The pathway from use to innovation, in other words, held up under rigorous structural testing.

The study then probed what operates between use and innovation, and two mediators emerged. Knowledge sharing, the extent to which students exchange ideas, insights, and information with peers, was identified as a positive driver of innovative behavior and a mediator of the relationship between generative AI use and innovation. Creative self-efficacy, a student’s belief in their own capacity to produce creative outcomes, played the same dual role. Both findings enrich the understanding of these mediators in educational settings: generative AI use appears to stimulate innovation partly by opening channels of knowledge exchange and partly by strengthening students’ confidence in their own creative abilities. The tools do not simply hand students finished answers; they seem to activate social and psychological processes that make students more inclined and more able to innovate.

The ANN analysis delivered the study’s most striking technical insight. When the neural network ranked the predictors of innovative behavior, creative self-efficacy emerged as the pivotal factor, underscoring its central role in fostering innovation among Generation Z students. This nonlinear importance ranking complements the structural findings: while PLS-SEM confirmed that generative AI use, knowledge sharing, and creative self-efficacy all significantly predict innovative behavior, the ANN revealed that the personal belief in one’s own creativity carries exceptional weight in the innovation equation. For educators, this suggests that the psychological lever matters as much as the technological one, and that interventions aimed at building students’ creative confidence may amplify the innovative payoff of generative AI adoption more effectively than simply providing access to the tools.

The implications for universities are actionable and timely. The confirmation that facilitating conditions positively influence adoption means that infrastructure, institutional support, and accessible training are prerequisites, not afterthoughts. The mediating roles of knowledge sharing and creative self-efficacy suggest that universities should design learning environments where generative AI use is paired with collaborative exchange and confidence-building exercises, rather than treated as a solitary activity. If the goal is to cultivate innovative behavior, the study indicates that effective integration requires attention to the full psychological chain: expectancy beliefs that motivate adoption, actual use that catalyzes innovation, and the social and self-belief mechanisms that carry the innovative impulse from tool to outcome.

The research also carries honest limitations that temper its conclusions. The cross-sectional survey design captures perceptions at a single point in time, so causal claims rest on the theoretical model rather than longitudinal observation, and self-reported measures of innovative behavior may not fully align with objectively assessed creativity. The sample of 291 students in China, while adequate for the analytical techniques, may not generalize to all Generation Z populations or all educational systems. The authors note that the study advances the burgeoning literature on generative AI in education by highlighting its capacity to cultivate innovative behavior among students and by offering insights for its effective integration into learning environments. As universities continue to debate the role of these tools, this dual-stage analysis offers a technically rigorous data point: the question is no longer only whether students will use generative AI, but whether institutions will build the conditions, the sharing cultures, and the creative confidence that turn that use into genuine innovation.

Subject of Research: The impact of generative AI adoption on the innovative behavior of Generation Z students in higher education

Article Title: Generative AI adoption and innovative behavior among Gen Z students: a deep-learning based dual-stage PLS-SEM-ANN analysis

Article References: Zhang, J., Quoquab, F., Xu, Z., Zhang, J., Mohammad, J., & Yan, B. (2026). Generative AI adoption and innovative behavior among Gen Z students: a deep-learning based dual-stage PLS-SEM-ANN analysis. Humanities and Social Sciences Communications. https://doi.org/10.1057/s41599-026-09524-7

Image Credits: AI Generated

DOI: 10.1057/s41599-026-09524-7

Keywords: generative AI, Generation Z, higher education, innovative behavior, UTAUT, Social Cognitive Theory, PLS-SEM, artificial neural network, creative self-efficacy, knowledge sharing, technology adoption, China

That adoption gap is the focus of a Research Topic published in Frontiers in Medical Engineering titled Bridging the Gap: Understanding Adoption and Acceptance of Emerging Healthcare Technologies, introduced by an editorial from Xianghui Peng of Penn State Erie and Victor Prybutok of the University of North Texas. Six studies, spanning Kuwait, China, and the United States and covering everything from pharmacist-facing mHealth applications to Facebook support groups for chronic lung disease, collectively map the behavioral, psychological, organizational, and social determinants that decide whether a healthcare technology lives or dies in the real world. Their combined message is sobering and empowering at once: acceptance is not a switch that flips when a product launches, but a multidimensional process shaped by trust, anxiety, cultural stigma, cognitive load, and social context.

The theoretical scaffolding for most of this work descends from the Technology Acceptance Model, originally formulated by F. D. Davis in 1989, and its successors the Unified Theory of Acceptance and Use of Technology and the consumer-oriented UTAUT2. These frameworks model adoption as a function of constructs such as performance expectancy, the belief that a technology will help the user accomplish something valuable, and effort expectancy, the perceived ease of use. What distinguishes the new research is how often those standard models had to be augmented with context-specific constructs drawn from complementary theories, including the Health Belief Model, Social Exchange Theory, and the Stressor-Strain-Outcome framework. Methodologically, the studies lean on quantitative survey research analyzed with structural equation modeling and its partial least squares variant, a statistical technique that tests networks of hypothesized causal pathways between latent psychological constructs simultaneously rather than one relationship at a time.

On the provider side, AlGhannam and colleagues examined pharmacists’ acceptance of mobile health applications within Kuwait’s government health sector using the mobile-specific extension of the Technology Acceptance Model. Their cross-sectional survey found that compatibility with existing workflows, performance expectancy, personal innovativeness, and effort expectancy all significantly predicted behavioral intention to use mHealth tools. Crucially, performance expectancy acted as a critical mediator, a statistical pathway through which compatibility and ease of use exerted their influence, meaning pharmacists embraced apps primarily when those apps demonstrably made their professional work better. The team also uncovered a chain of mediating effects linking job-related mental demands, professional identity, and perceived social support to adoption intention, underscoring that a clinician’s decision to embrace a new tool is embedded in psychological and social circumstances far beyond the app’s interface. For ministries of health attempting to scale mHealth in resource-constrained environments, these findings suggest deployment strategies must attend to workload and professional identity, not merely technical training.

Patient-facing studies reveal equally complex dynamics. Huang and colleagues investigated older adults’ adoption of online medical services in China using a modified UTAUT2 model enriched with technology anxiety, trust, and perceived risk. Drawing on 379 valid questionnaires analyzed through structural equation modeling, they found that performance expectancy, effort expectancy, social influence, and trust were positive drivers of adoption, while technology anxiety and perceived risk acted as significant barriers. Perhaps most surprising, facilitating conditions, meaning the availability of support and resources, and price value showed no significant effect at all, challenging assumptions routinely imported from mainstream consumer technology research where cost and convenience loom large. Moderating analyses showed that age, education level, and health status all reshaped these relationships, a reminder that older adults are not a monolithic bloc. As populations age worldwide, designing digital health systems that genuinely serve this heterogeneous group may require abandoning one-size-fits-all assumptions entirely.

A second patient-centered study by Boyce and colleagues widened the lens to online health communities, examining information-seeking behavior among patients with Chronic Obstructive Pulmonary Disease who use Facebook groups. Integrating the Health Belief Model, Social Exchange Theory, and the Technology Acceptance Model within a partial least squares structural equation modeling framework with 155 participants, the study demonstrated that ease of internet use and perceived utility were significant determinants of active participation, while emotional support and perceptions of self-worth played meaningful complementary roles. This positions online peer communities as a distinct and historically underappreciated modality of health technology adoption, one whose value flows as much from human connection as from information exchange. The implications cut both ways: platform designers can engineer for lower friction and richer reciprocity, while clinicians can treat these communities as legitimate extensions of the care ecosystem rather than digital curiosities.

Artificial intelligence introduces its own ambivalence. Russell and Bode conducted a mixed-methods analysis of high-engagement public discourse on X, formerly Twitter, from early 2025, analyzing 496 posts in English and Spanish through a pioneering application of the UTAUT lens. They identified a central paradox: public discourse reflects deep frustration with traditional mental health care, citing prohibitive costs and perceived inefficacy, while AI is valued not as a therapeutic equivalent but as a pragmatic tool offering accessibility, anonymity, and scalability. Simultaneously, AI is critiqued for its inability to replicate human empathy. Spanish-language discourse particularly highlighted how AI’s anonymity functions as a countermeasure to cultural stigma surrounding mental health, suggesting acceptance mechanisms can be deeply culture-specific. The authors’ most provocative conclusion is that AI in mental health is being adopted to patch systemic failures rather than because of its technological merit, an insight with profound ethical and policy weight for anyone designing hybrid human-AI care systems.

On the system side, Jia and colleagues studied user avoidance of intelligent customer service in pharmaceutical e-commerce through the Stressor-Strain-Outcome framework, using SmartPLS to analyze 418 valid responses. Their model showed that system overload, information overload, and service overload amplify users’ emotional stress, which in turn drives avoidance behavior, with strong explanatory power indicated by R-squared values ranging from 0.450 to 0.586. The finding exposes a mechanism often overlooked in adoption research: poorly designed intelligent systems do not merely fail to attract users, they actively repel them. In a digital pharmacy context, where an avoided chatbot may mean an unanswered question about dosing or drug interactions, this dynamic carries downstream consequences for medication adherence and patient safety. Adoption research, in other words, needs a science of rejection as rigorous as its science of intention.

Rounding out the collection, Ji and colleagues applied real-time motion-tracking and ergonomic analysis to competitive archery, the first such use of the technology in that sport. Thirteen participants performed tasks with varying bow draw weights and target positions, generating precise recordings of human movement that revealed gender-differentiated postural patterns and quantified spinal force exposure. Though distinct in its clinical engineering orientation, the study aligns thematically with the collection’s core concern: emerging technologies improving health-relevant outcomes beyond traditional clinical settings, and the promise of non-invasive, real-time sensing as a scalable approach to injury prevention and occupational health.

Across all six contributions, cross-cutting lessons crystallize. Context shapes adoption profoundly, with drivers and barriers varying by population, culture, and technology type. Negative behaviors, including avoidance, resistance, and ambivalence, deserve as much scholarly attention as positive intentions, since the field’s historical emphasis on behavioral intention risks undertheorizing non-adoption. Standard acceptance frameworks remain powerful but demand contextual augmentation with constructs like trust, anxiety, cognitive overload, stigma, and emotional support that were absent from the productivity-oriented settings where those models were born. And consistently, the evidence points toward user-centered design as the critical lever for closing the gap, whether by reducing cognitive burdens in intelligent customer service or building intuitive platforms for chronic disease communities. The editors call for longitudinal studies capturing actual use rather than stated intention, cross-cultural comparisons, and hybrid methodologies, urging that as generative AI, wearable biosensors, and ambient care platforms proliferate, the questions raised here will only grow more urgent for designers, implementers, and policymakers alike.

Subject of Research: Adoption and acceptance of emerging healthcare technologies including mHealth, AI systems, and online health communities

Article Title: Editorial: Bridging the gap: understanding adoption and acceptance of emerging healthcare technologies

Article References: Peng, X., & Prybutok, V. (2026). Editorial: Bridging the gap: understanding adoption and acceptance of emerging healthcare technologies. Frontiers in Medical Engineering, 4, Article 1898286. https://doi.org/10.3389/fmede.2026.1898286

Image Credits: AI Generated

DOI: 10.3389/fmede.2026.1898286

Keywords: healthcare technology adoption, technology acceptance model, UTAUT, mHealth, artificial intelligence, online health communities, digital health, technology anxiety, user avoidance, pharmacists, older adults, mental health

The research focused on a course that would seem an unlikely candidate for virtual worlds: Hospitality Cost Control, a subject rooted in numbers, recipes, and profit margins rather than flashy simulations. Yet Wang argued that the way students encounter that content matters as much as the content itself. Conventional video platforms present a grid of faces and a shared screen, a format that can flatten the social cues and spontaneous interactions that make classroom learning feel alive. Metaverse platforms such as Gather.Town take a different approach, rendering a two-dimensional pixel-art campus through which students move avatars, walk into virtual rooms, and trigger video and audio connections only when their avatars stand near classmates or instructors. The result is a hybrid of video conferencing and a video game that mimics the spatial logic of a real building.

To test whether that spatial logic translates into better learning, Wang recruited two intact classes of Hospitality Cost Control students at a university in southern Taiwan, a total of 96 participants. Forty-six students formed the control group, receiving their online instruction through Google Meet, while fifty students formed the treatment group, learning in Gather.Town. The design was deliberately conservative. Both groups began with the same face-to-face instruction phase and sat the same midterm examination, which served as a pretest to establish a baseline. Only then, during weeks eleven through seventeen of the semester, did the groups diverge into their respective online formats. At the end of the term, all students took a final exam and completed a cognitive load assessment, while the metaverse group additionally answered a survey based on the Unified Theory of Acceptance and Use of Technology, a widely used framework known as UTAUT.

The outcome on achievement was clear. Students taught in the metaverse environment scored significantly higher on the final examination than their counterparts in the video-conference group, even after accounting for their midterm baseline. In practical terms, the virtual campus did not merely make online lessons more entertaining; it measurably improved how much cost-control knowledge students retained and applied. Just as important was what the study did not find. The cognitive load assessment showed that the richer, more interactive environment did not impose a heavier mental burden. That distinction matters, because educational technologists have long worried that immersive platforms, for all their appeal, can overwhelm learners with novelty, navigation, and interface management, leaving less working memory available for the actual subject matter. The Taiwanese results suggest that a well-designed metaverse classroom can add engagement without adding strain.

The second half of the study tackled a different question: why do students embrace or reject such platforms? Here Wang turned to UTAUT, the model developed by Venkatesh and colleagues in 2003 that has become a workhorse of technology-adoption research. UTAUT proposes that four constructs shape a person’s intention to use a new technology: performance expectancy, the belief that the tool will improve one’s performance; effort expectancy, the belief that it will be easy to use; social influence, the sense that important others expect one to use it; and facilitating conditions, the perception that infrastructure and support are available. Behavioral intention, in turn, is expected to drive actual use behavior, moderated by the resources and support at hand.

Using structural equation modeling, a statistical technique that tests networks of hypothesized relationships among latent variables simultaneously, Wang found that all four UTAUT antecedents exerted significant positive influences on the hospitality students’ behavioral intention to keep using the metaverse platform. Students who believed Gather.Town would help them learn, who found it easy to navigate, who felt encouraged by peers and instructors, and who perceived adequate technical support were all more likely to intend to use it. The model then showed that facilitating conditions and behavioral intention together significantly predicted actual use behavior. In other words, wanting to use the platform was not enough on its own; students also needed the practical scaffolding of reliable equipment, connectivity, and institutional backing to convert intention into genuine engagement.

These findings arrive at a moment when the hospitality and tourism industries are themselves experimenting with virtual worlds, from virtual hotel tours to metaverse-based destination marketing, and when hospitality educators face pressure to prepare students for technology-rich workplaces. Prior research has documented growing interest in metaverse applications across hospitality and tourism, but rigorous classroom experiments comparing metaverse instruction against standard video conferencing have remained scarce. By anchoring the comparison in a real course, with intact classes, a pretest, a posttest, and a validated adoption model, the study offers a template that other institutions can adapt. It also speaks to the broader debate about post-pandemic online learning, where the question is no longer whether to teach online but how to do so without sacrificing the social and spatial richness of campus life.

The cognitive load result deserves particular attention from instructional designers. Cognitive load theory holds that working memory is a limited resource and that learning suffers when extraneous load, the mental effort imposed by the interface rather than the content, grows too large. Many immersive technologies flunk this test, dazzling students while burying the lesson. Gather.Town’s design, which layers video conferencing onto a simple, navigable map, appears to hit a sweet spot: it supplies spatial presence and serendipitous interaction, the ingredients missing from grid-based calls, while keeping the controls intuitive enough that students do not burn mental bandwidth figuring out where to click. The UTAUT results echo this interpretation, since effort expectancy emerged as a significant driver of intention, suggesting that ease of use was not a barrier for this cohort.

The study is not without boundaries. It involved two classes at a single university in southern Taiwan, all studying one hospitality subject, so generalizing to other disciplines, cultures, or platform designs requires caution. The treatment group alone completed the UTAUT survey, meaning the adoption analysis rests on fifty respondents and describes users of one specific metaverse platform rather than metaverse instruction in general. The quasi-experimental design, while strengthened by the midterm pretest and intact-class structure, cannot match the causal purity of randomized assignment. Wang also notes that the datasets generated during the study are available from the corresponding author upon reasonable request, and the research received ethical approval from the National Cheng Kung University Human Research Ethics Committee, with written informed consent obtained from all participants.

Even with those caveats, the implications are hard to ignore. Higher education invested heavily in video-conference teaching during the pandemic and has largely reverted to it as the default online format, yet student dissatisfaction with that format is well documented. This study suggests that the next generation of online instruction may look less like a conference call and more like a campus, a persistent virtual space where students wander, bump into classmates, and learn by moving through an environment rather than staring at a grid. For hospitality educators, whose field depends on spatial service, atmosphere, and human interaction, the metaverse may be an especially natural fit. For everyone else, the Taiwanese experiment offers a measured but provocative data point: the technology that students accept, and that genuinely lightens rather than loads their minds, may be the one that finally makes online learning feel less like a workaround and more like a place.

Subject of Research: Metaverse-based online instruction and technology acceptance in higher education hospitality training

Article Title: Metaverse adoption in hospitality education: using the UTAUT framework for online instruction environments

Article References: Wang, C.-J. (2026). Metaverse adoption in hospitality education: using the UTAUT framework for online instruction environments. Current Psychology, 45(19), Article 1569. https://doi.org/10.1007/s12144-026-10142-4

Image Credits: AI Generated

DOI: 10.1007/s12144-026-10142-4

Keywords: metaverse, UTAUT, hospitality education, online instruction, Gather.Town, cognitive load, learning achievement, technology acceptance, structural equation modeling, higher education, virtual learning environments, Current Psychology

The review, led by Xiaolin Wang of Peking University School and Hospital of Stomatology together with colleagues from the National Cancer Center in Beijing, was registered in PROSPERO and searched PubMed, Embase, Web of Science, and CINAHL from their inception through 7 September 2025. Two reviewers independently carried out study selection, data extraction, and quality assessment, and the team employed a best-fit framework approach: they mapped findings deductively onto the Unified Theory of Acceptance and Use of Technology, a widely used model of technology adoption, while also coding inductively for concepts that emerged from the data themselves. Eleven studies spanning multiple countries made the final cut, and confidence in the resulting findings was rated moderate to high using the GRADE-CERQual framework.

What emerged from the synthesis was not a single barrier or a single driver but five interrelated themes that together explain the fate of mHealth in head and neck cancer care. The first is perceived clinical value: patients, caregivers, and clinicians engage with digital tools when they can see a tangible benefit, whether that is earlier detection of complications, better symptom control, or more efficient communication with the care team. Where that value is ambiguous or invisible, enthusiasm evaporates quickly. The second theme concerns usability and burden, a factor that carries particular weight in this disease.

Head and neck cancer and its treatment impose a distinctive symptom burden. Radiation-based therapy commonly triggers oral mucositis, a painful inflammation of the mouth and throat lining, while surgery and radiotherapy frequently impair swallowing, speech, and shoulder function. Patients may struggle with dry mouth, neck stiffness, disfigurement, and profound fatigue. Against that backdrop, an app that demands lengthy data entry, precise touchscreen manipulation, or sustained visual attention can become yet another burden rather than a relief. The synthesis makes clear that usability is not a cosmetic concern for this population; it is a determinant of whether the technology is used at all.

The third theme identified by the reviewers is professional and social endorsement and support. Patients are far more likely to adopt and sustain use of a mobile health tool when clinicians actively recommend it, integrate it into consultations, and respond to the data it generates. Caregivers, who often shoulder much of the day-to-day care in head and neck cancer, act as both facilitators and gatekeepers, helping patients navigate the technology while also drawing support from it themselves. Social reinforcement, including encouragement from peers and family, further shapes whether a tool becomes part of daily routine or a forgotten icon on a home screen.

The fourth theme, system readiness and workflow integration, shifts the lens from the individual to the institution. Even a well-designed, clinically valuable app will falter if hospitals lack the infrastructure to receive its data, if responsibilities for responding to patient-reported symptoms are unclear, or if the tool sits awkwardly alongside existing electronic health records and clinical routines. The review underscores that mHealth adoption is not merely a patient-side decision but a system-side commitment, requiring organizational buy-in, training, and clear pathways for acting on the information the technology produces.

Perhaps the most striking insight of the synthesis is its fifth theme: needs and perceptions are not static. The cancer care trajectory, from diagnosis through intensive treatment into survivorship or palliative care, acts as a key moderator of engagement. A patient in the throes of radiotherapy may crave daily symptom check-ins and rapid clinical feedback, while the same person months later, in recovery, may value rehabilitation exercises, peer connection, or long-term surveillance instead. Tools designed for one phase of the journey often fail to flex to the next, and the review suggests that this mismatch between evolving needs and rigid technology is a major reason engagement decays over time.

The theoretical scaffolding of the study matters here. The Unified Theory of Acceptance and Use of Technology, and its later meta-analytic refinements, predict that performance expectancy, effort expectancy, social influence, and facilitating conditions jointly determine both the intention to use a technology and the behavior itself. By anchoring the synthesis in this framework while remaining open to emergent concepts, the researchers were able to translate abstract adoption theory into concrete, cancer-specific mechanisms. The result is a map that developers, clinicians, and health system planners can actually navigate: demonstrate value, minimize burden, mobilize professional and social champions, and engineer the system so that the digital tool has a genuine home in the workflow.

The stakes are considerable. Head and neck cancer accounts for a substantial share of the global cancer burden, with GLOBOCAN 2022 estimates documenting rising incidence and mortality across many regions, and the disease is notorious for treatment-related morbidity that persists for years. Prophylactic swallowing interventions, for instance, are known to improve outcomes but demand sustained patient effort at precisely the time when patients feel worst. Mobile platforms, including apps that coach swallowing exercises and wearable devices that track function, could close that gap, and prior studies of tools such as swallowing therapy applications have shown promising feasibility. But feasibility in a pilot study is not the same as sustained adoption in the messy reality of patients’ lives, which is exactly the terrain this synthesis illuminates.

The review also speaks to a broader conversation in digital health. The World Health Organization’s 2019 guideline on digital interventions for health system strengthening called for rigorous evaluation of whether such tools actually deliver benefit, and subsequent reviews across chronic disease and cancer survivorship have repeatedly found that user engagement is the weak link. By focusing specifically on the lived experiences of patients, caregivers, and clinicians in head and neck cancer, and by assessing confidence in each finding rather than simply aggregating quotes, the Chinese team has produced evidence that is both methodologically disciplined and clinically actionable. Their conclusion is measured but firm: acceptance and sustained use of mHealth in this field are shaped by a dynamic interplay among perceived value, usability, endorsement, and system readiness, with the trajectory of care moderating everything. For a field that has often treated app adoption as a one-time event, that reframing, engagement as a moving target that must be re-earned at every stage of the cancer journey, may prove to be the study’s most enduring contribution.

Subject of Research: Factors shaping acceptance and sustained use of mobile health technology in head and neck cancer care

Article Title: Experiences of mHealth use in head and neck cancer care: a thematic synthesis guided by the Unified Theory of Acceptance and Use of Technology

Article References: Wang, X., Zhou, N., Yang, Y., & Zhang, J. (2026). Experiences of mHealth use in head and neck cancer care: a thematic synthesis guided by the Unified Theory of Acceptance and Use of Technology. Supportive Care in Cancer, 34(10), Article 1015. https://doi.org/10.1007/s00520-026-11253-1

Image Credits: AI Generated

DOI: 10.1007/s00520-026-11253-1

Keywords: mHealth, head and neck cancer, UTAUT, qualitative evidence synthesis, digital health, symptom monitoring, swallowing rehabilitation, patient engagement, health informatics, supportive care, caregivers, technology acceptance

The research arrives at a pivotal moment for Bangladesh. Mobile financial services such as bKash and Nagad now process more than USD 20 billion in transactions every month, and smartphone coverage reached roughly 70 percent in 2023, creating an enormous trove of digital transaction data. Yet the banking sector remains hampered by persistent inefficiencies: slow loan processing, weak monitoring, and a non-performing loan ratio of 8.16 percent in 2023. While digital payments have flourished, the integration of FinTech into core loan management, including origination, tracking, and recovery, has lagged far behind, leaving a critical gap between the country’s payment revolution and its lending operations.

To bridge that gap, the research team extended the classic UTAUT framework, originally developed by Venkatesh and colleagues, by adding Loan Performance as a final endogenous variable. The original model posits that Performance Expectancy, the belief that technology improves job performance; Effort Expectancy, the perceived ease of use; and Social Influence, the pressure from peers, managers, and regulators, jointly shape Behavioral Intention to adopt a technology. Behavioral Intention and Facilitating Conditions, which cover infrastructure, training, and technical support, then drive actual Use Behavior. The researchers argue this original formulation fits institutional banking better than the consumer-oriented UTAUT2, because bank adoption decisions hinge on efficiency and performance rather than hedonic motivation or price value.

The data were collected through structured questionnaires administered both physically and electronically, targeting bank officials, loan officers, IT staff, and FinTech users across all major divisions of Bangladesh, including Dhaka, Chattogram, and Khulna. Of 440 initial surveys, 354 valid responses were retained, an 80.45 percent retention rate that comfortably exceeds methodological minimums for partial least squares structural equation modeling. Respondents were experienced and well qualified: nearly half had more than fifteen years in the industry, and most held graduate degrees or professional banking credentials. All constructs were measured on seven-point Likert scales adapted from validated prior instruments and refined through a pilot test with thirty respondents.

The statistical analysis, conducted with SmartPLS and verified with STATA, revealed a clear and statistically significant causal chain. Performance Expectancy exerted a strong effect on Behavioral Intention with a path coefficient of 0.381, while Effort Expectancy was the single most powerful adoption driver at 0.438, underscoring that in a market with uneven digital literacy, user-friendly systems matter enormously. Social Influence contributed a smaller but still significant 0.193. On the next stage of the chain, Facilitating Conditions dominated with a coefficient of 0.615, and Behavioral Intention added 0.380 in predicting actual Use Behavior. Most strikingly, Use Behavior predicted Loan Performance with a coefficient of 0.898, confirming that the benefits of FinTech materialize only through consistent, effective use rather than mere intention.

The explanatory power of the model was remarkable across every stage. Behavioral Intention accounted for 82.9 percent of the variance, Use Behavior for 88.8 percent, and Loan Performance for 80.7 percent, all well above the 0.75 threshold conventionally regarded as strong. Measurement quality checks were rigorous: indicator loadings, composite reliability values between 0.880 and 0.942, and average variance extracted figures above 0.50 confirmed convergent validity, while heterotrait-monotrait ratios all fell within the acceptable 0.90 limit. Harman’s single-factor test and full collinearity assessments indicated that common method bias was not a serious concern, and robustness checks re-estimating the model without high-collinearity items produced consistent results.

The theoretical contribution is substantial. Most previous UTAUT research stopped at intention or usage, and most FinTech performance studies examined macro-level indicators such as profitability without grounding them in adoption theory. By linking individual-level acceptance constructs to institutional loan outcomes through a behavior-to-performance pathway, the study converts UTAUT into a performance-oriented institutional model. The mediation analysis is particularly instructive: Behavioral Intention affects Loan Performance indirectly, only through Use Behavior, demonstrating that strategic financial gains require the transformation of positive attitudes into daily operational practice.

The findings also carry important cautions. The authors emphasize that FinTech adoption exposes banks and borrowers to significant risks, including phishing, identity theft, ransomware, fraudulent loan applications, and algorithmic bias that could disadvantage rural borrowers, women, and small enterprises with thin credit files. Data privacy, transparency in automated lending decisions, responsible affordability assessments, and accessible grievance mechanisms are essential to sustaining trust. Bangladesh’s substantial Islamic banking sector adds a further layer, since digital lending products must comply with Shariah principles and offer clear contractual terms. Regulatory foundations such as Bangladesh Bank’s ICT Security Guidelines, mobile financial services regulations, and updated electronic know-your-customer requirements provide a starting framework, but regulatory clarity for emerging tools like peer-to-peer lending and blockchain-based credit remains uncertain.

For policymakers and bank leaders, the practical implications are direct. Because Effort Expectancy proved so influential, banks should invest in intuitive, well-tested lending platforms and pair them with digital literacy programs for staff and customers. The dominance of Facilitating Conditions argues for sustained spending on IT infrastructure, reliable connectivity, and technical support, especially in rural branches, alongside cybersecurity measures such as multi-factor authentication, encryption, real-time fraud detection, and rapid incident response. Social Influence, though weaker, can be harnessed through leadership-led digital transformation targets, industry-wide standards, and visible success stories that normalize FinTech-based lending.

The study does have limitations that temper interpretation. It relies on self-reported, perceptual measures of loan performance, since bank-level financial data are confidential, and it treats all banks as operating under comparable conditions despite real differences between public, private, and specialized institutions. Its cross-sectional design cannot capture how FinTech benefits accumulate over time, and moderating factors such as organizational culture, customer trust, and regulatory constraints were not modeled explicitly. Future research using objective indicators such as non-performing loan ratios and recovery rates, along with longitudinal and moderated designs, would strengthen the evidence. Even so, the Bangladesh experience offers a compelling template for other developing economies: locally adapted digital innovation, when paired with infrastructure, training, and prudent regulation, can convert technology acceptance into faster, safer, and more inclusive lending.

Subject of Research: FinTech adoption and bank loan performance in Bangladesh using the UTAUT model

Article Title: The role of FinTech in enhancing bank loan performance in Bangladesh through the UTAUT model

Article References: Rana, M., Shah Jalal, M., Kamruzzaman, M., & Islam, M. A. (2026). The role of FinTech in enhancing bank loan performance in Bangladesh through the UTAUT model. Discover Global Society, 4(1), Article 254. https://doi.org/10.1007/s44282-026-00573-6

Image Credits: AI Generated

DOI: 10.1007/s44282-026-00573-6

Keywords: FinTech, Bangladesh, UTAUT, bank loan performance, digital lending, structural equation modeling, behavioral intention, use behavior, financial inclusion, cybersecurity, mobile financial services, emerging economies

The research, published in Smart Agricultural Technology, was led by Sabrina Gulab, Hanan Ishaque, and Guillaume Lhermie, who surveyed 587 commercial crop and livestock producers operating farms larger than 700 acres across the three Prairie provinces. After restricting the analysis to producers who had adopted at least one digital agricultural technology, or DAT, the final analytical sample comprised 392 operations. The team used a binary logit regression model to identify which factors were associated with what they call data-use intensity, defined as the degree to which producers integrate technology-generated data into day-to-day farm management decisions. Respondents were classified as making high use of data or low or no use, and the model estimated how behavioral, psychological, and institutional variables shifted the probability of falling into the high-use category.

The theoretical foundation of the study is an extension of two influential frameworks in technology acceptance research: the Technology Acceptance Model, or TAM, and the Unified Theory of Acceptance and Use of Technology, known as UTAUT. Both frameworks were originally designed to explain why people adopt technology in the first place, emphasizing constructs such as perceived usefulness, effort expectancy, and facilitating conditions. The authors argue that these models fall short in the post-adoption phase, where the relevant question is not whether a farmer expects a tool to perform, but whether the continuous stream of data it generates is perceived as actionable, trustworthy, and valuable for ongoing decisions. They systematically remapped UTAUT constructs to post-adoption equivalents: performance expectancy became the perceived usefulness of data insights, facilitating conditions became data handling support, and social influence was extended to capture emerging norms around data sharing and collective contribution to agricultural data ecosystems.

The descriptive findings alone are sobering for the digital agriculture industry. Only 40.3 percent of surveyed producers reported high use of data in farm decision-making, while 54 percent reported low use and 6 percent reported no use at all. In other words, a majority of farmers who had already paid for and installed digital technologies were not meaningfully integrating the resulting information into their management choices. Producers operating larger farms showed higher levels of data integration than those on smaller operations, a pattern the authors attribute to economies of scale, where data-driven management produces clearer efficiency gains on complex, resource-intensive enterprises.

The regression results sharpen the picture considerably. Perceived usefulness of data emerged as the single strongest predictor: a one-unit increase in the belief that data helps optimize resources and inputs was associated with a 20.4 percentage point higher probability of high data use. Openness and proactive behavior came next, with roughly a one-standard-deviation increase associated with a 14.9 percentage point rise in the probability of high data use. Producers who actively seek information about new technologies, experiment with unfamiliar tools, and attend workshops appear to sustain engagement with data long after the initial purchase. Altruism also played a meaningful role, with producers comfortable sharing farm data with providers, researchers, companies, and government institutions showing an 8.7 percentage point higher probability of high data use, consistent with the idea that contributing to a shared data ecosystem both reflects and reinforces deeper engagement with one’s own data.

On the negative side, risk perception significantly suppressed data engagement. Producers who felt the risks of digital technologies outweighed their benefits had an 8.1 percentage point lower probability of high data use. Notably, the study uncovered a moderating effect: the negative influence of risk perception was stronger among high adopters who used many technologies, suggesting that farmers with the most at stake operationally and financially are also the most sensitive to concerns about data privacy, system reliability, and technological risk. Lack of digital literacy was another substantial barrier, associated with a 9.8 percentage point lower probability of high data use. Drawing on cognitive load theory, the authors argue that when interpreting complex dashboards and agronomic recommendations exceeds a producer’s cognitive resources, the data simply never makes it into the decision-making process.

Perhaps the most counterintuitive finding concerns social support. Producers who frequently consulted neighbors when they ran into difficulties with their technologies were less likely to use data intensively, with each unit increase in reliance on peer consultation associated with a 4.3 percentage point decrease in the probability of high data use. The authors interpret this as evidence that dependency-oriented support may substitute for independent engagement: farmers with lower technological self-efficacy lean on neighbors rather than developing the autonomous problem-solving skills needed to work with data themselves. This challenges a long-standing assumption in agricultural extension literature that peer networks uniformly accelerate technology uptake, and it suggests that building independent digital competence may matter more than facilitating help-seeking.

Equally striking is what did not predict data use. Concerns around data governance, including privacy, ownership, and commercial exploitation of farm data, showed no statistically significant association with data-use intensity, nor did trust in technology providers or the lack of facilitating conditions such as technical support and system interoperability. These factors dominate the adoption literature and are frequently cited in policy debates, yet in this sample of commercially established producers they did not shape what happened after adoption. The authors propose two explanations: established commercial producers may already operate within sufficiently developed governance arrangements, or governance concerns may function primarily as barriers at the adoption stage rather than during ongoing data integration. Either way, the finding underscores a stage-based view of technology engagement in which different barriers operate at different points along the adoption continuum.

The implications extend beyond farm gates. High-quality, farm-level data are the raw material for the next generation of AI-enabled agriculture, feeding the predictive models and recommendation systems that promise precision forecasting and optimized input use. If most adopters leave their data unexamined, both current returns on technology investment and future innovation pipelines are constrained. The authors argue that policy should therefore prioritize behavioral and cognitive enablers alongside infrastructure spending: strengthening digital literacy, reducing perceived risk through experiential learning, and framing data sharing as a contribution to the collective resilience of the agricultural community. Crop producers, who showed a 29.3 percentage point lower probability of high data use than livestock producers, may deserve particular attention, since livestock systems tend to depend on continuous monitoring that makes data value more immediately visible. As governments pour money into rural broadband and digital agriculture subsidies, this study offers a clear warning: the bottleneck is no longer the technology itself, but the human capacity and motivation to turn its output into decisions.

Subject of Research: Post-adoption data use behavior among commercial crop and livestock producers using digital agricultural technologies in Western Canada.

Article Title: Beyond adoption: Understanding data use behavior in digital agriculture

Article References: Gulab, S., Ishaque, H., & Lhermie, G. (2026). Beyond adoption: Understanding data use behavior in digital agriculture. Smart Agricultural Technology, 15, Article 102546. https://doi.org/10.1016/j.atech.2026.102546

Image Credits: AI Generated

DOI: 10.1016/j.atech.2026.102546

Keywords: digital agriculture, data use behavior, precision agriculture, technology adoption, UTAUT, data governance, digital literacy, farm decision-making, risk perception, altruism, Western Canada, agricultural policy

The study, led by Agostinho Sousa Pinto of the Polytechnic Institute of Porto together with colleagues in Portugal, Spain and Bangladesh, surveyed 202 university students and analysed their responses using partial least squares structural equation modelling, or PLS-SEM. This statistical technique allows researchers to test how multiple latent factors, such as trust or performance expectancy, interact to shape an outcome like the intention to use a technology. The team built their model on the extended unified theory of acceptance and use of technology (UTAUT), a framework that has become the workhorse of technology adoption research since its introduction in the early 2000s.

What sets this work apart is the addition of geopolitical concern to the model. The construct captures worries about national security risks, the strategic implications of relying on a Chinese AI system, and the broader question of technological dependency. The results showed that these concerns significantly shaped students’ trust in DeepSeek and their privacy perceptions, and in doing so indirectly dampened adoption intentions. In other words, students did not simply ask whether the tool worked well; they also asked who built it, where their data might end up, and what it means for their country’s digital sovereignty to depend on a foreign power’s AI infrastructure.

Despite these anxieties, the overall picture was surprisingly favourable. Students rated DeepSeek positively for reasons that reflect the realities of studying in a developing economy: the model is free to use, shows an element of cultural alignment that Western tools often lack, and offers a way to reduce dependence on Western technology giants. The researchers describe DeepSeek’s approach as a form of frugal innovation — delivering capable AI at minimal cost — and argue that this model provides a blueprint for emerging economies seeking affordable access to frontier technology.

On the technical side, the analysis identified performance expectancy and facilitating conditions as the key drivers of adoption. Performance expectancy, the belief that the tool improves learning and research productivity, was a strong predictor of behavioural intention, while facilitating conditions — the availability of infrastructure and support, such as reliable internet access — also played a significant role. Effort expectancy, how easy students find the tool to use, contributed as well. By contrast, social influence, the pressure or encouragement from peers and instructors, and hedonic motivation, the sheer enjoyment of using the technology, had negligible effects. This suggests Bangladeshi students approach AI tools pragmatically rather than socially or recreationally.

The mediating role of trust deserves particular attention. The study found that geopolitical concerns did not simply add or subtract directly from adoption intentions; instead, they worked indirectly by eroding trust and amplifying privacy concerns. This pathway aligns with earlier research showing that trust is a critical mediator in AI acceptance, and it highlights how geopolitical sentiment can undermine even a technically capable and free product. For policymakers in the Global South, the implication is that promoting AI adoption is not only a matter of building infrastructure but also of addressing public anxieties about data flows and foreign dependency.

The timing of the research is significant. DeepSeek made global headlines with its efficient mixture-of-experts architecture and its reasoning-focused R1 model, demonstrating that frontier-level AI could be developed and deployed at a fraction of the cost of Western equivalents. For countries like Bangladesh, which sit outside the major AI powers and often cannot afford premium subscriptions to Western services, such models are attractive precisely because they lower the barrier to entry. The Bangladeshi press has even framed DeepSeek’s rise as a call for the country to invest in retaining its own AI talent rather than watching it emigrate.

At the same time, the study is candid about the risks. Security analysts have raised questions about data governance in Chinese AI services, and the researchers note that privacy risks remain real even when users view a tool favourably. Bangladesh’s internet infrastructure also presents a constraint: reliability and speed gaps can undermine facilitating conditions and therefore suppress adoption, no matter how capable the underlying model is. The authors argue that policymakers and AI developers must address both geopolitical sentiments and infrastructure shortfalls if adoption is to grow.

Broader lessons extend well beyond Bangladesh. As the supply of globally available AI models diversifies, adoption decisions in the Global South increasingly reflect a calculus of sovereignty, cost and cultural fit rather than a simple race to the most powerful model. The researchers emphasise the need for culturally congruent, sovereignty-sensitive AI tools — systems that respect local data concerns while remaining affordable. Their study is, to their knowledge, among the first to formally integrate geopolitical factors into an established AI adoption framework, opening a line of inquiry that seems likely to grow as AI becomes an arena of great-power competition.

For educators and developers watching the generative AI boom, the message is clear: in emerging markets, the winning formula combines genuine usefulness, minimal cost, dependable infrastructure and attention to the political anxieties that surround foreign technology. DeepSeek’s popularity among Bangladeshi students shows that even in a field dominated by Silicon Valley narratives, the Global South is charting its own course through the AI revolution — one weighed down by real concerns, but propelled by the promise of accessible intelligence for all.

The theoretical lineage of the framework used in the study is worth unpacking. UTAUT emerged from a synthesis of eight earlier acceptance models and was designed to explain a large share of the variance in behavioural intention across workplace technologies. Its later extensions added constructs such as hedonic motivation, price value and habit, reflecting the shift from mandatory enterprise systems to consumer-facing tools. The Bangladeshi study pushes this evolution further by treating geopolitics as a measurable latent variable rather than background noise, an approach that acknowledges how international relations now shape everyday software choices in ways the original model’s authors could not have anticipated.

The methodological choices also merit attention. Partial least squares structural equation modelling is particularly suited to exploratory research where a new construct is being introduced, because it places fewer demands on sample size and distributional assumptions than covariance-based approaches. The researchers followed established practice by assessing the measurement model for reliability and validity before testing the structural paths, drawing on widely cited criteria for convergent and discriminant validity. A sample of 202 students is modest but adequate for this technique, and the reliance on self-reported intentions means the findings describe attitudes rather than observed long-term behaviour, a limitation common to the adoption literature.

The technical backdrop to DeepSeek’s appeal lies in its architectural efficiency. The model family employs a mixture-of-experts design, in which only a subset of parameters activates for any given query, cutting computational cost substantially compared with dense models of similar capability. Its reasoning model demonstrated that reinforcement learning could elicit strong chain-of-thought performance, and independent evaluations have since tested such models on engineering tasks with encouraging results. For students in low-income settings, this efficiency translates directly into free or near-free access to tools that would otherwise sit behind subscription paywalls.

Bangladesh’s broader digital context helps explain the pattern of results. The country has invested heavily in digital government services and mobile connectivity, yet internet quality remains uneven, and the study’s emphasis on facilitating conditions echoes real infrastructure constraints documented in connectivity indices. Prior research on AI literacy among South Asian students, including library and information science cohorts in Bangladesh, India and Pakistan, has shown uneven familiarity with AI systems, suggesting that effort expectancy and support structures matter as much as raw capability. The pragmatic orientation of the students surveyed, with social influence playing little role, fits a picture of adoption driven by tangible academic need rather than fashion.

The geopolitical dimension resonates with precedents elsewhere in the region. Debates over the TikTok ban in India illustrated how digital sovereignty concerns can override consumer popularity, and public opinion research has documented shifting views of China and the United States among South Asian populations. The study’s finding that geopolitical sentiment operates through trust rather than directly suggests a subtle mechanism: users may continue to value a tool’s usefulness while quietly discounting their confidence in it, a state of ambivalence that could shift rapidly with news of data mishandling or regulatory change.

Future research could extend the framework in several directions. Longitudinal designs would reveal whether geopolitical concerns harden or soften as familiarity grows, and comparative studies across countries with different alignments could test whether the construct behaves consistently. Sampling instructors, administrators and policymakers alongside students would broaden the picture, and behavioural measures such as actual usage logs would strengthen inference. As AI models multiply and great-power competition intensifies, the integration of geopolitical constructs into adoption theory offers a template for understanding how the next generation of digital tools will be welcomed, resisted or renegotiated across the Global South.

Subject of Research: Adoption of the DeepSeek large language model among university students in Bangladesh, analysed with an extended UTAUT framework incorporating geopolitical concern.

Article Title: Exploring DeepSeek Adoption in Higher Education in Bangladesh: A UTAUT-Based Approach

Article References: Pinto, A. S., Abreu, A., Cota, M. P., Paiva, J., & Biswas, M. S. (2026). Exploring DeepSeek Adoption in Higher Education in Bangladesh: A UTAUT-Based Approach. Frontiers of Digital Education, 3(2), Article 16. https://doi.org/10.1007/s44366-026-0090-2

Image Credits: AI Generated

DOI: 10.1007/s44366-026-0090-2

Keywords: DeepSeek, generative AI, UTAUT, Bangladesh, higher education, geopolitical concern, PLS-SEM, technology adoption, digital sovereignty, Global South, trust and privacy, frugal innovation