Building Learning Health Systems requires the combination of information, regulatory, and cultural infrastructures that create communities focused on changing health outcomes through the application of quality improvement methodology, focused data collection, closed feedback loops, and community‐participatory techniques. Accomplishing the vision of the Learning Health System relies on building robust infrastructures, and teaching a wide variety of stakeholders to participate in these novel socio‐technical systems. In this commentary, I draw on empirical examples from fieldwork with Learning Networks to describe how social scientists view culture and what this concept might hold for learning health sciences.

Patent applications provide insight into how inventors imagine and legitimize uses of their imagined technologies; as part of this imagining they envision social worlds and produce sociotechnical imaginaries. Examining sociotechnical imaginaries is important for emerging technologies in high-stakes contexts such as the case of emotion AI to address mental health care. We analyzed emotion AI patent applications (N=58) filed in the U.S. concerned with monitoring and detecting emotions and/or mental health. We examined the described technologies' imagined uses and the problems they were positioned to address. We found that inventors justified emotion AI inventions as solutions to issues surrounding data accuracy, care provision and experience, patient-provider communication, emotion regulation, and preventing harms attributed to mental health causes. We then applied an ethical speculation lens to anticipate the potential implications of the promissory emotion AI-enabled futures described in patent applications. We argue that such a future is one filled with mental health conditions' (or 'non-expected' emotions') stigmatization, equating mental health with propensity for crime, and lack of data subjects' agency. By framing individuals with mental health conditions as unpredictable and not capable of exercising their own agency, emotion AI mental health patent applications propose solutions that intervene in this imagined future: intensive surveillance, an emphasis on individual responsibility over structural barriers, and decontextualized behavioral change interventions. Using ethical speculation, we articulate the consequences of these discourses, raising questions about the role of emotion AI as positive, inherent, or inevitable in health and care-related contexts. We discuss our findings' implications for patent review processes, and advocate for policy makers, researchers and technologists to refer to patent (applications) to access, evaluate and (re)consider potentially harmful sociotechnical imaginaries before they become our reality.

Purpose Electronic flashcards (EFs) are a widely used learning resource in medical education. This scoping review provides a comprehensive overview of the existing empirical research on EFs and a direction for future research on EF interventions. The authors organized the reviewed articles into 4 nonmutually exclusive categories: development, delivery, utilization, and associated outcomes. Method In this scoping review, search queries were composed for 6 databases (PubMed, Embase, Education Resource Information Center, Web of Science, PsychInfo, and the Cumulative Index to Nursing and Allied Health Literature). The database search was conducted on April 12, 2024, for articles published before this date using the following keywords (as well as other keywords): anki, quizlet, firecracker, osmosis, flashcard, spaced repetition, spaced interval training, spaced interval learning, spaced test, and parallel curriculum. Two reviewers screened the studies for eligibility and performed data extraction between August 12, 2024, and September 26, 2024. Results Sixty-four studies were included in the review. Most articles were published in the last 6 years (2019-2024) (49 [77%]), in the United States (45 [7%]), and in the context of undergraduate medical education (42 [66%]). The studies predominantly focused on the utilization (51 [78%]) and associated outcomes (38 [59%]) of EFs. In contrast, EF development (12 [19%]) and delivery methods (16 [25%]) were less frequently examined. Conclusions This review of the rapidly increasing body of research on EFs suggests broad use and satisfaction among health professions learners; however, much work remains to be done to systematically understand EF development and delivery. Future research directions may consider more systematic investigations of how decisions made during the development and delivery of EFs affect downstream use and learning outcomes. Addressing these gaps will deepen the understanding of how EFs affect medical education pedagogy and facilitate the informed integration and refinement of these tools within curricula.