To Örebro University

The interconnected crises of climate change, biodiversity loss and chemical pollution constitute nothing less than an existential threat to humanity. Yet the global community’s response is manifestly inadequate. Against this backdrop, this dissertation analyses climate obstruction through the notion of corporate environmental legitimation – a relational process that cultivates and maintains the perception that corporations are suitable, willing and capable environmental stewards, even within a structurally unsustainable economic system. Empirically, the dissertation presents a multi-method case study of the Ellen Mac-Arthur Foundation (EMF), a leading advocacy organisation promoting the concept of a circular economy. Combining critical political economy and critical discourse analyses, it examines EMF’s funding, networks, governance, policy engagement, promotional practices, dis-course and relations to EU circular economy policy. The findings shows that EMF plays a central role in promoting a version of the circular economy that reinforces corporate environmental legitimacy, and that this largely overlaps with the EU’s version. Corporate initiatives are framed as meaningful progress within content circulated through EMF’s promotional system, where different legitimation strategies interact to reinforce this message. Furthermore, the analysis shows that EMF mobilises the circular economy as a brand, directing public enthusiasm towards corporate “circular” initiatives. The dissertation argues that this constitutes an important mode of climate obstruction: messages that create the impression that adequate solutions are underway function as smoke and mirrors. They foster hope in incremental improvements while potentially crowding out attention to the urgent systemic transformations required to return humanity to a safe operating space.

Rapid changes in population health needs have increased the importance of understanding how healthcare institutions in low- and middle-income countries (LMICs) support the development of healthcare professionals’ competencies. While traditional training has primarily emphasised clinical and technical skills, healthcare systems in LMICs often face additional challenges related to limited resources, entrenched hierarchies, and complex cultural contexts. These conditions highlight the need to better understand the development of non-technical skills (NTSs), including communication, teamwork, leadership, and interprofessional collaboration.

This thesis examines how healthcare professionals understand and develop NTSs, with particular attention to leadership and interprofessional collaboration within healthcare settings in the Balkan region and in the context of international healthcare partnerships in Sweden. The study applies a designed experiential learning approach grounded in participatory and interactive educational practices. Kolb’s experiential learning theory serves as the overarching conceptual framework and is applied across individual, team, and organisational levels, with surveys-used to stimulate dialogue.

The thesis is presented as a compilation, consisting of an introduction and four empirical studies employing both qualitative and quantitative research methods. Data were collected through interviews, focusgroups, and observations.

The findings indicate that designed experiential learning approaches effectively foster the development and understanding of NTSs. These skills develop in ways that are closely shaped by organisational structures and cultural contexts. The results contribute to a growing body of evidence supporting experiential, participatory, and context-sensitive approaches in healthcare education and suggest that traditional dyadic mentoring alone is insufficient to support professional development in increasingly complex healthcare environments.

In public health crises, such as pandemics, nurses’ willingness to care for patients, even when risking one’s own well-being, is essential. The overall aim of this thesis was to explore nurses’ willingness to care, and experiences of nursing care and altruism, in relation to a pandemic. The thesis includes four qualitative studies, based on interviews with primarily intensive care nurses, in which the phenomenological hermeneutical method and systematic text condensation were used for analysis, and in addition, the concept of professional altruism was clarified. The composite findings of the qualitative studies (studies I–IV) showed that, before the COVID-19 pandemic, nurses felt ambivalence towards the phenomenon of altruism. When the COVID-19 pandemic hit Sweden, nurses showed great willingness to care for patients and displayed altruistic behaviour. Working at over full capacity took its toll, and nurses’ willingness to care started to diminish when personal consequences became apparent. They were pushed to the edge, and for some nurses their willingness to care was transformed into a need to prioritise oneself, and ultimately to resign from intensive care. The last study (Study V) considered professional altruism as a willingness to act for others, a moral orientation, a motivational force, an unwavering professional expectation, and a valued, yet challenged and sometimes rejected phenomenon. Supporting the willingness to care and professional altruism, as well as to prepare nurses and nursing students for future crises, including ethical decision-making in situations of scarce resources, is crucial to ensure effective health crisis management in the future.

Overall aim: The overall aim was to investigate and optimize inhaler use and patient education in COPD in primary health care.

Methods: Study I was an observational study assessing patients’ inhaler use. Study II was a non-randomized controlled clinical trial evaluating the effect of an additional inhaler education session. Study III was a register-based study identifying factors associated with patient education. Study IV was a qualitative focus group study exploring asthma/COPD nurses’ experiences of inhaler education.

Results: In Study I, 45% of the 183 patients made at least one inhaler error; 50% made errors related to device handling, 31% made errors relating to inhalation technique, and 19% made both types of error. In Study II, the education group showed a significant improvement in device-handling errors, after one revisit. Study III revealed that only 44% of patients with COPD had received patient education, primarily those with more severe disease. In Study IV, asthma/COPD nurses described inhalation treatment as challenging, and reported deficiencies particularly in continuing education, limited managerial support, and insufficient time for patient care.

Conclusions: The findings demonstrate a high prevalence of incorrect inhaler use among patients with COPD. Categorizing errors into device-handling and inhalation-technique errors may simplify assessment and facilitate more individualized, patient-centered education. Health care organizations should prioritize COPD and support asthma/COPD nurses through education, collaboration, and adequate time for patient care. Patient education, including inhaler training, should be initiated earlier and offered to more patients.

Bloodstream infection (BSI) is a life-threatening condition associated with high mortality. Pathogen identification is essential for patient management, yet blood culture (BC), the diagnostic gold standard, may require 2–5 days and has reduced sensitivity after antimicrobial treatment. This thesis investigates modern molecular DNA sequencing approaches for detecting bacterial DNA as a diagnostic alternative. In Study I, a cohort of 484 patients with suspected BSI was described, and prediction tools were applied to identify patients at high risk of BSI for subsequent studies. Study II used whole blood samples from 51 patients to detect bacterial DNA with short-read sequencing and showed low concordance with routine BC results. In Study III, a shotgun metagenomic workflow using the Nanopore platform was developed, and DNA extraction efficiency was evaluated in contrived samples. Bacterial DNA recovery was slightly higher in whole blood than in plasma, but no firm conclusion regarding the optimal sample matrix could be drawn. We also observed that extraction efficiency differed between bacterial species. Based on these methodological challenges and limited sensitivity, Study IV evaluated bacterial DNA enrichment prior to sequencing through short BC incubation (4 h) followed by targeted 16S gene sequencing in 161patients selected using the criteria defined in Study I. This workflow showed increased diagnostic performance, with sensitivity of 56.8% and specificity of 46.7% compared with BC. Additional bacteria with plausible clinical relevance were detected in a few patients with negative routine BC. Overall, the sequencing protocols evaluated in this thesis provided limited additional diagnostic value for BSI compared to BC. In comparison with the literature, we highlight methodological challenges to guide future research in this area.

This scoping review examines school-based programs aimed at promoting the adjustment and well-being of recently arrived immigrant adolescents. Following PRISMA-ScR guidelines, five databases (Medline, PsycINFO, CINAHL, SCOPUS, ERIC) were systematically searched for studies published since 2000, focusing on interventions implemented in formal school settings for recently arrived adolescents. The review identified 15 studies evaluating outcomes of 17 programs. Most school-based programs for recently arrived adolescents aimed to promote social-emotional well-being, mitigate mental health problems, strengthen resilience and social support, or address trauma-related symptoms. Around 50% of the reviewed programs had some effects on the intended outcomes. Despite some promising findings, the current literature has several limitations that limit the ability to draw robust conclusions. Future research needs to focus on understanding why, how, and for whom programs lead (or do not lead) to intended outcomes, and on developing effective programs that can be implemented using available resources at school. 

This article investigates how pressures toward constraint in higher education, conceptualized analytically as saturation, can be understood through the perspectives of communication professionals within Scandinavian higher education institutions (HEIs). Drawing on 29 qualitative interviews with communication professionals at twelve HEIs across Denmark, Norway, and Sweden, the study examines how these actors frame changing expectations around relevance, efficiency, and questions of institutional legitimacy. Rather than treating saturation as a purely structural condition, the article approaches it as an analytical concept emerging from how actors interpret and respond to converging pressures, highlighting the communicative practices through which universities navigate new constraints. The article contributes to higher education research by advancing understanding of saturation beyond labour market mismatch and capacity limits, showing how it is constructed through strategic communication. It also provides empirical insight into how universities respond to emerging pressures and offers a framework for understanding how professional staff interpret changing conditions in higher education systems under pressure.

The purpose of this dissertation is to contribute to knowledge about teachers’ didactic choices when teaching content to students with intellectual disabilities. To achieve this, the study examines how teachers choose and justify teaching content. The focus is on content that highlights the educational goals of democracy and sustainable development. The results from the sub-studies are developed into didactic principles that explain how teachers can make content decisions to prepare students with intellectual disabilities for adulthood in society and the world. Didactic theory helps explain teachers’ content choices by emphasizing the connections among teachers, students, and content, as well as the democratic purpose of education and the broader contexts of schooling, society, and global sustainability. Three studies were conducted: a scoping review (Study I) and two interview and observation studies (Studies II and III). Study I shows that teachers’ didactic choices are rarely explicitly discussed in prior research and that academic content often overlaps with life functional content. Study II identifies participation, independence, and communication as key reasons for selecting content for students with intellectual disabilities; these reasons are evident across school, societal, and global contexts. Study III shows that teachers incorporate the Sustainable Development Goals and sustainability concepts into their teaching. These choices help students develop skills that support a sustainable society and world – such as self-determination, codetermination, and solidarity. In the final chapter, the findings are synthesized into three didactic principles based on teachers’ choices: 1. Teaching should give students with intellectual disabilities the chance to become active agents in a democratic society. 2. Teaching should provide students with intellectual disabilities opportunities for communication, independence, and participation. 3. Teaching should enable students with intellectual disabilities to contribute to sustainable development.

Modern medicine relies heavily on effective antimicrobial therapies. However, rapid emergence and global spread of antimicrobial resistance (AMR) threaten the treatment and control of bacterial infections. Neisseria gonorrhoeae, the causative agent of gonorrhoea, has an extraordinary capacity to acquire and develop resistance to all antimicrobials introduced for treatment. The aim of this study was to gain insight into the evolution and global emergence of AMR over the past century. This thesis further investigates whether the observed genomic data patterns are sufficiently informative for accurate prediction of antimicrobial susceptibility that can directly inform treatment.

By applying whole-genome sequencing (WGS) to global, historical, and contemporary gonococcal collections spanning all continents and isolated from the pre-antibiotic era to the present day, several questions can be addressed. The data revealed that the global gonococcal population is divided into two distinct lineages with different evolutionary strategies. Furthermore, temporal analysis demonstrates that the modern N. gonorrhoeae is younger than previously presumed and that antimicrobial exposure has been a major driver of the evolution of this species.

The accumulated knowledge base of phenotypic and especially genomic AMR generated in this work is compiled and integrated within a dedicated analytical framework, SensiTyper, demonstrating that WGS-based approaches can infer antimicrobial susceptibility and recommend susceptibility-guided individualised treatment strategies.

This essay attempts to draw together a coherent theory of cognitive emergence from Jean-François Lyotard’s speculations on thinking machines and from his later writings on sense and aesthetic experience. The essay situates Lyotard’s arguments about cognition in the context of recent successes of large language model (LLM) artificial intelligence and the persistence of the so-called ‘symbol grounding problem’ in that field. Rather than tracing a developmental line from stimulus at the material substrate to symbolic reasoning, Lyotard envisions a recursive relationship between sense and reason, in which new forms of thought emerge from an agent’s sensuous experience of symbolic categories and logical constructs, even if that experience is just the suffering of waiting for an analogy to form or a paradox to resolve. It is through an agent’s sensuous experience of its own thoughts that it is able to form an internal model of itself inside of time—a crucial ingredient for genuine cognition.  

Given the intensifying environmental and sustainability crises that societies face, there is an urgent need for transformative change on all levels in society. A key question is what role education can, and should have, in a transformation toward sustainability. In this chapter we suggest one answer to this question by outlining a teaching approach in environmental and sustainability education (ESE) called critical pluralism. By drawing on the wider and established pluralistic tradition in ESE, we outline what a critical pluralistic approach could mean in teaching practice. Central to this approach is that arguments and opinions that are brought up in classrooms must be critically scrutinized by both the teacher and the students. However, discussing environmental and sustainability issues comes with didactic challenges for teachers, such as how to open for plural perspectives without giving into an “anything goes” relativism. In this chapter we address this and similar didactic challenges. In our closing discussion we argue that critical pluralistic ESE is not merely teaching strategy to implement but can be seen as a more fundamental approach to the relation between education, democracy and transformative change.

This report presents the development of a blueprint for a Master’s program tailored to public sector professionals without formal technical or programming backgrounds. The blueprint was developed as a part of WASP-ED (The Wallenberg AI and Transformative Technologies Education Development Program) work area 2. The program addresses the growing need for competencies among public servants leading or participating in AI-supported transformation in public organizations. It aims to equip domain experts with foundational AI knowledge and cross-disciplinary insight into the intersection of technology, society, and individuals. The goal is to bridge the knowledge gap between technical experts and sector-specific specialists.

The blueprint responds to knowledge gaps identified through a review of academic literature on AI system development and implementation in public organizations, recent reports from public and private entities, and complemented by empirical data collected through workshops, discussions, and interviews with key stakeholders. The suggested program is grounded in both theory and practice, drawing on the Sociotechnical AI systems perspective and the AI System Development Life Cycle (AI-DLC), designed specifically for public sector needs.

The initial version of the blueprint was shared with stakeholders including study directors, and others involved in developing Master's-level education in relevant disciplines. It was also evaluated against the WASP-ED curriculum and existing Master’s programs in Sweden.

Finally, there is a strong network of informatics researchers across Sweden — including at Örebro University (ÖRU), University of Gothenburg (GU), Linköping University (LiU), and Mid Sweden University — who possess relevant expertise, actively publish in this field, and maintain close collaborations with the public sector. These competencies and networks will play a key role in shaping and further developing the program.

The human intestinal micmbiome plays an important role in health due to the large number of beneficial effects related to the bacterial profile and the metabolites generated in the intestine. Arabinoxylans are compounds present in different cereals such as wheat and they can modulate the profile and functioning of some beneficial bacteria from human intestinal microbiota. In the present work, a colonic in vitro fermentation with human faecal inoculum was done using arabinoxylans extracted from Argentinian hard and soft wheat as substrates. Molecular size alteration of arabinoxylans were studied during fermentation and Lactobacillus and Bifidobacterium abundance as well as short chain fatty acids concentrations were determined. The arabinoxylans fermentation was proved to induce the growth of Bifidobacterium and the release of short-chain fatty acids. The speed and efficiency of fermentation were different for each of the arabinoxylan extracted from both wheat genotypes, perhaps because of differences in their chemical and physical structures. The consumption of water-extractable arabinoxylans (WE-AX) (either supplemented or enriched) to maintain the balance or modulate in a favorable way the profile of Bifidobacterium can be an important contribution to the human health.

Insomnia is a sleep disorder with a high prevalence in teenagers, which leads to suffering and an increased risk of future health problems. The Insomnia Severity Index (ISI) is a validated and well-researched self-report questionnaire that aims to quantifyinsomnia and its severity in adults. It is used in insomnia research and has been translated into many languages, including Swedish. The psychometric properties of the ISI have been thoroughly investigated for adults and to some extent on teenagers abroad, but a psychometric evaluation of a Swedish transla-tion of the ISI with young people in Sweden is still missing. The approach of this study was to evaluate the psychometric properties of the ISI in a sample of Swedish teenagers. A cross-sectional study with data from 205 participants from a larger study was conducted to ascer-tain the reliability of the scale and seek to operationalize its underlying components. The results indicate that the ISI is a reliable self-assess-ment scale for young people in a Swedish context. Furthermore, two components were distinguished by using principal component analysis. The components were operationalized into typical and atypical insom-nia symptoms for teenagers, respectively. The typical symptoms were difficulties falling asleep combined with daytime complaints, while atypical symptoms consisted of difficulties falling asleep after waking up during the night, and difficulties with premature awakenings early in the mornings. The conclusion is that a Swedish translation of the ISI has good psychometric properties in teenagers, and that the underlying structure of the ISI for teenagers differs from what studies on adults have found, and to some extent studies on foreign teenagers. This could reflect age-related differences in sleep, cultural differences or problem-atic phrasing of the Swedish version of the ISI.

Background: Insomnia is a common health complaint among adults and is associated with poor health. Sleeping patterns differ due to age, with younger adults having greater problems with sleep deprivation causing problems during daytime, and older adults having greater problems with nocturnal awakenings during nights. Screening tools are needed to initially capture insomnia symptoms in need for further assessment. The study aim was to psychometrically test and compare the original three item Minimal Insomnia Symptom Scale (MISS), revised version of MISS (MISS-R), and a four-item scale (MISS-4) on an adult population, by using the Rasch measurement model.

Methods: A cross-sectional design was used, and the sample consisted of participants in an online survey of a Citizen Panel. Out of the total sample (N = 1517), 784 (52%) persons with answers on four sleep items, age, and gender were included in this study. The majority of the sample were men (n = 536), 248 were women, and 52% were younger (16–59 years). Rasch model analyses focused on targeting, reliability, response category functioning, model fit, differential item functioning (DIF) and score transformation.

Results: The MISS and the MISS-R had good measurement properties. When adding the item “daytime disturbance” to the MISS, and creating MISS-4, the measurement properties deteriorated. When replacing the original MISS item “not rested by sleep” with the item “daytime disturbance” in the MISS-R, the measurement properties maintained like those of the MISS. There were, as expected, DIF between the younger (below 60 years) and older samples (60 years or older) in insomnia symptoms. The reliability for the MISS was 0.58 and the MISS-R 0.57.

Conclusions: The MISS and the MISS-R demonstrate good measurement properties as insomnia screening questionnaires. When using either the MISS or the MISS-R to identify potential insomnia, a ≥ 6 cut-off seems to be most appropriate. While psychometrically supported, determination of the optimal cut-score for clinical insomnia identification should consider gold standard diagnostic criteria in future studies.

Boken Smartare Jurist med Generativ Artificiell Intelligens av Magnus Kristoffersson utgör en praktiskt orienterad handbok för jurister och juriststudenter som vill använda generativa AI-verktyg på ett effektivare och säkrare sätt. Författaren introducerar KOMPASS-metoden – en akronym för Kontext, Objektiv, Material, Precisering, Arbetsmetod, Svarsformat och Säkring – som ett strukturerat ramverk för att formulera effektiva prompts i juridiskt arbete.

Boken redogör för hur stora språkmodeller fungerar, varför de tenderar att "hallucinera" och vilka risker detta medför för juridisk verksamhet. Med Microsoft Copilot som huvudexempel, men även med behandling av ChatGPT, Claude och Gemini, visar författaren hur GenAI-verktyg bäst används för att stödja form, struktur och disposition snarare än att okontrollerat producera texter med juridisk substans.

En central del av boken utgörs av QC-processen (Quality Control), en trestegsmodell för systematisk granskning och verifiering av AI-genererat innehåll. Författaren betonar genomgående att det professionella ansvaret alltid kvarstår hos användaren och att GenAI-verktyg inte kan ersätta den mänskliga juridiska bedömningen eller självständig rättskällesökning.

Boken riktar sig till såväl nybörjare som mer erfarna användare och syftar till att överbrygga klyftan mellan teknisk kapacitet och metodiskt ansvarsfull användning av AI i den juridiska domänen.

Traditional tabular augmentation methods, such as SMOTE and Gaussian sampling, treat features as generic vectors, disregarding the domain-specific constraints often present in scientific tabular data. This work introduces a domain-aware augmentation approach that leverages Large Language Models (LLMs) to encode scientific knowledge through policy generation. The effectiveness of this approach is demonstrated using a case study on geochemical compositions, where data must satisfy closure constraints and exhibit intrinsic correlations that geometric interpolation methods fail to preserve. Evaluated on an imbalanced geochemical rock classification dataset, the LLM-based augmentation achieves 95.74% accuracy and a 0.9544 macro-F1 score, outperforming SMOTE, Gaussian sampling, and no-augmentation baselines while requiring fewer synthetic samples.

In ecomusicology, important steps have been taken towards an understanding of music to be an activity produced by more-than-human relations. These are not just scholarly ideas. Rather, they have been present in fäbod culture for centuries. Swedish fäbod culture is a transhumance system where the farmers transfer their cows, goats, and sheep to a summer farm during the summer seasons. Historically, these farmers were women. The herders communicate with the animals, and with each other, through both vocal signals and the use of instruments. Previous studies and historical records show that fäbod farmers consider these signals as interspecies communication and acts of kinship. Starting from the view of the fäbod farmers, this thesis explores how vocal herding music is co-composed by more than humans in specific situations, and how vocal herding music in turn affect the very milieu that produced it. This thesis also suggests methods and theoretical approaches that allow for more-than-human participation in eco- and ethnomusicological studies. Drawing from 4 years of seasonal ethnographic fieldwork and participant observation, the results show that the rhythms, voices, and movements of animals and forests, partake in the organisation of singing and calling. A sung melody, a phrase, or a rhythm bare witness of the movements and vocalisations of the animals. By using, and elaborating on, the concepts mimesis and affective attunement, the thesis uncover show affect, as body sympathy or intensity, not only enables interspecies communication, but also affords humans and animals to attune to each other and through that process reach shared experiences where voices and rhythms synchronise (mimesis). The study also shows that by connecting to nature, kulning singers free themselves from normative ideas about what kulning is and should sound like; the forest becomes part of both resistance and creativity.

For traditional learning algorithms, inferring structured outputs, such as decisions that must satisfy domain-specific constraints, is not straightforward. A common challenge arises when elements in the output vector are interdependent in ways that standard learning algorithms or neural network activation functions cannot easily enforce. On the other hand, the fields of Operations Research and Discrete Optimization offer a complementary approach with the use of specific solvers or algorithms with techniques to ensure constraint satisfaction, generally formulated as Constrained Optimization Problems (COPs).

This thesis explores how COPs with predefined constraints can be integrated into learning frameworks by learning missing components of their cost functions through gradient-based optimization to ensure constraint satisfaction in learning tasks. Learning those missing components of the COP cost allows the inference process to use a suitable solver under the learned COP that ensures structured outputs, i.e., outputs satisfying those predefined constraints. The key challenge in this approach is the learning process itself, namely the difficulty of propagating gradients through the solver, which is often nondifferentiable or discrete. This tension motivates the central question of the thesis: How to learn the costs of the COP such that the observed decisions can be approximately recovered when applying the corresponding COP solver?

To address this question, we develop new methods for estimating or approximating gradients through specific classes of solvers while also leveraging existing techniques from differentiable optimization. These methods underpin novel deep learning frameworks that achieve structured inference by treating solvers as implicit layers in the learning process. This thesis is grouped into three learning paradigms: i) decision-focused learning, ii) imitation learning, and iii) structured generative modeling. In each paradigm, a neural architecture infers the cost parameters of the underlying COP. Consequently, the proposed methods not only preserve the expressiveness needed to capture complex relationships but, more importantly, guarantee feasibility at inference time.

Applications examined in this thesis are relevant to real-world decisionmaking and include improving the understanding of uncertainty estimation in conditional allocations through Portfolio Optimization Problems and contextaware inventory decisions via Inventory Optimization Problems; enhancing scalability in recovering transition costs in graphs from observed paths and cycles using Discrete Routing Optimization Problems, where large-scale experiments are conducted with real taxi and ship trajectory data; and improving the quality of control inputs in autonomous racing with simulation and hardware-implemented Model Predictive Controllers. Across all these applications, the proposed methods were successful and demonstrated consistent improvements over established baselines.

This article introduces the concept of epistemologies of communication – culturally and institutionally embedded ’ways of knowing’ that guide professional practice – to examine how communication professionals conceptualize and legitimize their work. We draw on the work by strategic communicators in Nordic higher education institutions (HEIs), a context simultaneously shaped by pressures for visibility and competitiveness, while also operating within the so called ‘Nordic’ model where ideals of egalitarianism, trust, and collaborative approaches are central. Drawing on 29 semi-structured interviews across Denmark, Norway, and Sweden, we perform an analysis of metadiscourses to explore epistemological foundations of communication. The findings reveal diverse metadiscourses that point to how strategic communication should be understood not only as an organizational function but also as a practice grounded in explicit and implicit epistemologies of communication – ways of knowing that shape communicative choices and strategic direction. The metadiscursive accounts also illuminate how professionals navigate tensions between Nordic values and market-oriented demands. The study suggests that there is not one Nordic way of strategic communication, but presents a characterization of the region in terms of its adaptability and contextual responsiveness, rather than uniformity

Rapporten ”Viktig fråga, men vem bryr sig?" handlar om hur reformer om jämställdhet och mångfald som en gång utvecklats för att åtgärda strukturella ojämlikheter mellan olika grupper i samhället, idag å ena sidan ifrågasätts och beskrivs som politisering som hotar konstnärlig frihet, yttrandefrihet och demokrati, och å andra sidan ses som en marknadsfördel. 

This umbrella literature review is driven by the pressing need for research to transform the elderly care sector, which is challenged by a rapidly ageing population and constrained resources. In this context, data-driven decision-making emerges as a promising approach to facilitate the necessary transformation. The aim this this review was to explore how socio-technical aspects are discussed in research on the use of data-driven decision-making in elderly care settings. Our analysis suggests that the majority of current research places disproportionate focus on technological development while neglecting the social, structural, and systemic contexts that are critical for successful and sustainable implementation. A more balanced and contextually aware approach—grounded in socio-technical systems thinking—is essential to advance the field and ensure that digital innovations truly enhance elderly care.

Unlocking the full societal potential of artificial intelligence demands a fundamental shift towards responsible and reproducible research. Understanding that PhD students are pivotal in conducting and reproducing experiments, we investigated the challenges of 28 AI PhD candidates from 13 European countries. We identify three critical areas where current practices fall short: (1) the findability and quality of AI resources such as datasets, models, and experiments; (2) the difficulties in replicating the experiments in AI papers; (3) and the lack of trustworthiness and interdisciplinarity. After uncovering some of the underlying reasons behind the challenges, we propose a combination of social and technical recommendations to overcome the identified challenges and foster a more transparent and reliable AI research ecosystem. Socially, we recommend the general adoption of reproducibility initiatives in AI conferences and journals, as well as improved interdisciplinary scientific collaboration, especially in data governance practices. On the technical front, we call for enhanced tools to better support versioning control of datasets and code, and a computing infrastructure that facilitates the sharing and discovery of AI resources, as well as the sharing, execution, and verification of experiments.

Automated task planning traditionally relies on manually generated domain models, creating bottlenecks in scalability and requiring extensive domain expertise. This paper presents a novel framework to automate the process of generating Planning Domain Definition Language (PDDL) domains and problem files by integrating Web Ontology Language (OWL) ontologies with Visual Language Models (VLMs). Our approach leverages the rich semantic structure of OWL ontologies to systematically define domain classes, predicates, and actions, while VLMs ground abstract ontological concepts in concrete visual observations—automating the generation of instance‑specific planning problems. The proposed framework transforms ontological knowledge into PDDL domain files through a mapping algorithm that preserves semantic relationships and logical constraints. The VLM performs visual scene analysis to identify relevant objects, attributes, and spatial configurations for generating initial states, while natural language instructions are used to derive goal states.We evaluate the framework across multiple planning domains, demonstrating that it generates syntactically correct and semantically coherent PDDL domain and problem files directly from OWL ontologies, camera images, and natural language inputs. The resulting files are comparable in quality to those manually generated by planning experts and outperform previous automated systems in terms of semantic fidelity and adaptability.

This article brings together societal debates on digitalisation and analysis of governmental policy on older people, through the theoretical frame of hegemonic digitalisation, and the empirical case of Finnish policy. With changing demographics in many countries, old age has gained high-profile focus in governmental policy in recent years, which has in turn become increasingly constructed in relation to digitalisation, often via an interventionist logic that positions new technologies as solutions to problems of ageing. In this analysis, material-discursive power relations are highlighted, with policy understood as material-discursive in its formation, form and effects. Specifically, the article critically examines: the conceptualisation of hegemonic digitalisation, often assumed, even across political differences, as an efficient solution to declining resources within neoliberal governance, with reference to policy on/and older people; how digitalisation and older people are constructed in governmental policy documents in Finland; and the applicability of hegemonic digitalisation to Finnish policy on older people. The article concludes with discussion of the implications that follow for everyday life, policy, social analysis and social theory. Regarding theoretical implications, with hegemonic digitalisation, the hegemony of digitalisation is emphasised, as in national policy contexts, rather than whole-society or global hegemony seen as digital. Additionally, tensions persist between foregrounding hegemonic digitalisation, and the co-constitution of older people and digitalisation policy. Specificities of ageing and older age impinge on and present challenges for both analysis of digitalisation and social theory. Theorising materiality-discursivity is a fertile perspective for studies on age, policy and digitalisation, including how future-orientation figures in policy and research.

Employers in Sweden are mandated to take active measures to prevent discrimination against sexual minorities. While it is important that relevant measures are taken, knowledge is lacking about cisgender lesbian, gay, and bisexual (LGB) people’s minority stress experiences at Swedish workplaces. The present work is based on a thematic analysis of interviews with 53 cisgender LGB participants, focusing on how they experienced and dealt with minority stress experiences at work. Results are drawing on the minority stress model and illustrate experiences of distal minority stress due to a heteronormative work climate. This distal stress led to proximal stressors, such as constantly being on guard. Participants took considerable responsibility for others’ feelings, felt a responsibility to educate on LGBTQ issues, and sometimes engaged in formal policy work to improve workplace conditions. The study points at the importance of shifting the burden of workplace minority stress from individual LGB people to employers.

Recent advances in treatment effect estimation (TEE) leverage deep neural networks to learn latent representations that balance treatment and control groups. However, despite their empirical success, we know remarkably little about the learned representations themselves—what structure they capture, why they lead to improved performance, and how far we can trust them. Existing evaluations focus almost exclusively on causal outcome metrics such as PEHE or ATE bias, leaving the representational mechanisms largely unexplored. This lack of understanding limits our ability to explain why models perform well, to identify when they fail, and to design more interpretable or trustworthy causal systems. In this work, we present a comprehensive analysis for representation-level evaluation of TEE methods. We benchmark CFR, DRCFR, TEDVAE and modern disentanglement-driven models using a unified suite of quantitative and graphical measures. Quantitatively, we assess AUC(T | Z) for confounding separability, Mutual Information (MI) retention for informativeness, MMD2 and SW1 for balance, CH-ratio for compactness, and Participation Ratio (PR) for effective dimensionality. Graphically, we analyze bias removal through embedding visualizations, kNN-mixing histograms, covariance heatmaps, and outcome-smoothness histograms—providing an interpretable view of latent geometry. Our analysis shows that modern VAE-based disentangling models yield more structured, balanced, and informative representations, which also correlate strongly with causal reliability. We provide a practical way to see inside TEE representations—revealing hidden biases, interpreting latent behavior, and understanding when conventional embedding properties (such as neighborhood preservation) may actually hinder bias mitigation. Overall, our findings emphasize that evaluating how models learn is essential to understanding why they perform well, paving the way toward more interpretable and trustworthy causal inference.

Monitoring of gas emissions is critical for ensuring safety, efficiency, and compliance in industrial plants. Approaches based on stationary sensor networks leave areas of the plant unmonitored and lack adaptability to changing conditions. Combining stationary with mobile sensors on quadruped robots offers a robust approach with improved spatial resolution, adaptability and responsiveness. To streamline the integration of such systems, a unified and open software framework is required. We propose to abstract the low-level implementations, e.g., the robot control layers, using the REST API to facilitate the implementation of high-level tasks such as data visualization and AI-based analysis, thus enabling scalable monitoring solutions. An indoor gas release experiment was conducted, which showed the benefits of the quadruped robot-based and the combined sensing approach.

Gas leaks in industry and nature can cause harmful effects on the environment and human health. Knowing how much gas is emitted over time helps to assess potential damage, track pollution sources, and develop effective mitigation strategies. To address this challenge, we propose to estimate the source's release rate based on measured gas concentrations in a cross section of the gas plume in the down-wind regions of the source. By combining wind information and the 2-D gas distribution in the cross section plane, we can infer the flow of gas through this plane, which is equal to the release rate of the source. We propose a Tunable Diode Laser Absorption Spectroscopy sensor (TDLAS) for remote, open-path gas sensing. By combining multiple TDLAS measurements with a gas tomography reconstruction algorithm, we obtain a 2-D map of gas distribution. This letter introduces an improved novel approach for gas tomography by incorporating prior model assumptions into the algorithm. Our method significantly enhances the accuracy and robustness of release rate estimates. We validate our approach through wind tunnel experiments, demonstrating that our novel estimation method produces precise and reliable release rate estimations for methane gas. The results further encourage exploring how 3-D gas tomography can improve our release rate estimations in the future.

In a world marked by democratic decline, the aim of this paper is to summarize research on digital activism in undemocratic countries. We examine how the field of ICT4D can engage with digital activism research in such contexts by highlighting both the role of technology in enabling activism and the profiles of those conducting this research. Based on a scoping review, we find that the main enabler of technology use in undemocratic countries is its capacity to support on-the-ground organizing and to sustain network of activists together in a community. The major constraint, however, is government surveillance and censorship. We conclude by arguing that future studies should involve activists in both conducting and reflecting on the research, in order to generate meaningful and practice-oriented insights. Inspiration can be drawn from action-oriented and practitioner-focused methodologies, including—but not limited to—those used in information systems. Furthermore, future research should explore technological challenges in greater depth, particularly issues related to security and anonymity. Addressing these challenges requires the involvement of information systems researchers with expertise in security and privacy.

Gas distribution mapping (GDM) is essential for industrial safety and environmental monitoring, as it enables real-time hazard detection and air quality assessment. Traditional GDM methods, such as kernel-based techniques, struggle to reconstruct complex gas plume dynamics accurately. While deep learning has shown promise for GDM, two critical challenges hinder its practical use: the scarcity of available training data and the incompatibility of conventional architectures with irregular sensor layouts. To address these limitations, we propose GNN-DM, a graph neural network-based model for GDM that incorporates the relational structure of sensor networks to infer high-resolution maps from minimal, irregular inputs. The model is pretrained on synthetic gas dispersion data generated from measured wind data and fine-tuned on two industrial datasets collected on a ferry car deck and in a hot rolling mill. Compared with established GDM techniques, GNN-DM achieves higher accuracy on synthetic and real-world data, highlighting the potential of graph-based learning for practical gas mapping applications.

This study explores disability representation in the transnational television crime drama Midsomer Murders(1997-present). It focuses on Lana who has a partial right arm and how she is positioned as a central figure in the Christmas 2023 episode. The analysis reveals a complex and contradictory pattern of interwoven ableist and anti-ableist ideas. Even though negative stereotypes are avoided, Lana’s portrayal is simultaneously visually and emotionally conditioned to avoid alienating audiences. The study speaks to critical scholars across disciplines with a broad interest in the fictional television crime drama and its abilities to communicate social (diversity) issues.

What could happen in the future if AI technologies became sentient like human beings, thus capable of suffering and feeling existentially alienated? The question warrants serious consideration, especially in light of recent advances in biocomputing, where small human brain organoids have been cultivated that display a rudimentary form of sentience. Currently, researchers aim a tcreating complex networked interfaces where several brain organoids are interconnected with each other, to real-world input/output sensors, and to machine-learning models. This constitutes a first step in the development of an ‘organoid intelligence’ (OI) that could potentially improve future decision-making and be much more energy efficient. But what are the ethical stakes of bringing new sentient and intelligent creatures to life to work for the benefit of humanity? In the present study, my aim is to highlight the latest progress in biocomputing and to examine the hypothetical risks of creating OI using a ‘speculative design fiction’ approach. My main argument is that we should not disregard the possibility that these efforts could accidentally cause the phenomenal experience of suffering to spill over from biology into technology, which could thus have tragic consequences. I will present three different, but partly overlapping, scenarios that illustrate how OI might experience and respond to an alienated labor situation. In the conclusion, I will assess a few solutions meant to prevent OI suffering and the importance of expanding our moral circles.

This paper presents the co-designing process of the StandByMe 2.0 project, which aims to counte ronline and offline gender-based violence (GBV) through youth actions in Hungary, Italy, Poland, and Slovenia. The main goal of the project is to develop a digital platform to educate and empower young people to recognise and combat GBV. Employing a systematic and iterative user-centred development approach, the paper outlines steps of forming an expert team, defining the problem, incorporating evidence-based intervention methods, gathering feedback from educators, and integrating gamification. The findings indicate that the platform’s activities could raise awareness, foster empathy, and promote action; however, activities requiring complexity and sensitivity require human interaction. This underscores the importance of human monitoring in gamified interventions addressing social issues like GBV. In the future, the project aims to enhance the platform by incorporating additional gamification mechanisms and involving young people in evaluating its effectiveness against GBV.

We tested how individuals' dispositional perspective-taking impacts the effectiveness of an intergroup perspective-taking intervention aimed to reduce anti-Gypsyism in Hungary. For the intervention, we adapted a role-playing book with a Roma adolescent protagonist (called the 'Gypsy Maze'). We predicted that the intervention would be more effective in reducing prejudice for stronger perspective-takers, compared to weaker, for whom it may even be counterproductive, potentially triggering victim blaming. We conducted a pilot field study among high school students and two experiments among university students (N = 982). In Study 1, we found that dispositional perspective-taking qualified the effect of the intervention. While strong perspective-takers endorsed less prejudice following the intervention versus control conditions, weak perspective-takers endorsed more. In Study 2, we aimed to replicate Study 1 and test underlying explanations, however, the intervention was ineffective in reducing prejudice among both weak and strong perspective-takers. These results draw attention to the limits of perspective-taking interventions and highlight the importance of both societal-contextual boundaries and personalizing intergroup interventions. Please refer to the Supplementary Material section to find this article's Community and Social Impact Statement.

Misrecognition describes everyday practices that deny the autonomy of minority members to define who they are and instead impose identities that may diverge from their own sense of self. Being misrecognized is particularly relevant for the historically marginalized Roma people, whose national belonging is repeatedly questioned despite centuries of co-existence and citizenship. Our aim was to understand whether the experience of misrecognition, along with discrimination, would predict identification patterns that represent an obstacle to dual identification among Roma people in three East-Central European countries: Hungary, Romania and Serbia. We collected data among Roma participants online and face-to-face with convenience sampling (N = 1,325). Latent class analysis revealed three similar classes based on national and ethnic identification scores in all subsamples: (1) disidentification, (2) Roma identification and (3) dual identification. Logistic regression analysis showed that misrecognition and discrimination predicted stronger Roma identity than dual identification in Hungary and Serbia. However, misrecognition predicted stronger dual identification in Romania, possibly as a reaffirmation strategy in response to misrecognition. Our results show that misrecognition can add to our understanding of minority group members' identification with the superordinate category of the nation as well as subgroup ethnic minority identity, and this connection could be key for advancing Roma inclusion.

Gender-based violence is widely recognized as a significant societal issue, necessitating a concerted effort to raise awareness about its manifestations and consequences. Serious games have been often used to address the topic, particularly among the younger population. On the other hand, the use of gamification (i.e., integrating game elements to enhance user engagement with the subject) to raise awareness of gender-based violence remains unexplored. To address this gap, we developed a gamified platform aimed at raising awareness of gender-based violence and in particular of the role of gender stereotypes, in young adults. This paper presents the findings of our preliminary mixed-methods evaluation and a comparison between two versions of the platform: one employing cooperative gamification, and the other one based on an individual game modality. While the quantitative results indicate that the current platform may not be sufficient to have a positive effect in participants' awareness of gender-based violence and attitudes toward bystander behavior, qualitative analyses yield crucial insights into participants' positive evaluation of the cooperative mode. These analyses provide valuable information for future developments of the platform, and contribute to the exploration of gamification applied to sensitive topics.

The concept of microaggressions alerts us how majority group members' everyday behaviour can impact minorities negatively. Recently, some researchers have questioned the criteria for identifying microaggressions and rejected the concept's utility. We maintain that attending to minorities' everyday experiences is important and illustrate this through a three-phase study with Roma in Hungary. First, we conducted interviews exploring their everyday interactional experiences (Phase 1, N = 17); second, Roma participants filmed (naturally occurring) interactions with majority group members (Phase 2, N = 10); third, we showed such filmed interactions to Roma focus groups and recorded their discussions (Phase 3, N = 28). Analysing these discussions, we focused on how the experience of surveillance when shopping (even when manifested in apparently helpful attention from shop assistants) impacted participants in ways that majority group members likely have little awareness of. Specifically, participants reported their need to (a) reflect on (and manage) their emotional reactions; (b) weigh a variety of strategic considerations as to how to respond; and (c) engage in in-the-moment interpretation as to the nature of the interaction. Such experiences negatively impact the use of public space and illustrate the value of adopting the minority's vantage point concerning the identification of microaggressive treatment.

"Brilliance," a state of extreme intellectual ability, is stereotypically associated with men but not women. Research finds that portrayals of brilliance as a prerequisite for success contribute to women's underrepresentation in certain academic fields and high-level positions. In this work, we examined whether gender roles contribute to the perception of women as less brilliant. In four preregistered experimental studies (N = 920), we tested whether brilliance deviates from ascribed and prescribed gender roles more for women than for men and whether such deviation places women who display their brilliance at a higher risk of experiencing backlash. In Study 1, an average intelligent and a brilliant man were rated as more similar on gender-specific traits than an average intelligent and a brilliant woman. In Study 2, while intelligence and gender individually influenced prescriptions of masculinity and femininity, their interaction did not support larger differences for female targets, indicating a lack of differential expectations by gender and intelligence. Study 3 showed that brilliant women are more likely to experience backlash at work than brilliant men, while Study 4 demonstrated that while brilliance enhances professional desirability across genders, it decreases social desirability, suggesting social costs that could affect workplace dynamics. Our results support that brilliance can be considered a form of gender-role deviance for women and might lead to a backlash. This underscores the need for policies to counteract gendered stereotypes of brilliance, which hinder women's career advancement and contribute to the gender gap in the workplace.

Gender-based violence (GBV) remains a critical human rights issue, deeply rooted in gender inequality and affecting individuals globally. The current study evaluated the user experience of a gamified platform designed to raise awareness about GBV. Gamification, using game elements in non-game environments. has been proven to promote online and offline learning, but its effectiveness has yet to be tested in the case of sensitive educational material. We explored how the platform motivated and engaged users through two versions: individual and cooperative. Using the MEEGA360 scale for user experience and the Geneva Emotion Wheel for emotions, 40 users were randomly assigned to one of the versions. Results showed the platform was well-received, with users finding it enjoyable, user-friendly, and effective in facilitating discussions on GBV. Positive emotions like involvement, amusement, and interest were common, though negative emotions like irritation and anger also appeared. Despite these positive outcomes, the platform faced usability challenges and requests for more complex activities and detailed feedback. The cooperative version scored higher on social interaction but did not significantly outperform the individual version. Further research is needed to explore these differences and improve the platform's effectiveness in GBV prevention.

Anti-Gypsyism is a deeply entrenched form of ethnic bias in Europe, characterised by realistic conflict perceptions, social norms approving bias, unacknowledged historical victimisation, and denial of cultural autonomy. Mainstream prejudice reduction interventions may have limited applicability to anti-Gypsyism, because prejudice is normative, hence people are not motivated to change their views; interventions focusing on increased liking, but not competence perception consolidate status differences; stereotypes are resistant to change because of the overlap with low SES stereotypes and assimilation of counter-stereotypical members. With the help of a meta-analysis with 169 effect sizes from 18 countries in 14 languages we showed that contact frequency was indeed only associated with affective (liking), but not cognitive and behavioural prejudice. However, we found the expected negative association between positive contact and prejudice. Eurobarometer data also showed the salience of counter-normative experiences, as friendships had a stronger impact in countries with higher overall means of anti-Gypsyism.

The coronavirus pandemic posed a major challenge to mental health. Existing evidence shows that COVID-19 is related to poor emotional well-being, particularly among women. However, most work on the subject uses single-country samples, limiting the ability to generalize the disparity or explain it as a function of societal variables. The present study investigates the expression of positive and negative emotions during the pandemic as a function of gender and across 24 countries (N= 49,637). Strong gender differences emerged across countries, with women reporting more negative emotions (anxious, depressed, nervous, exhausted) and less positive emotions (calm, content, relaxed, energetic) than men. The gender gap in positive emotions was significantly wider in countries higher in individualism and narrower in countries higher in power distance. For instance, differences in emotions were larger in Western countries high in individualism, such as the USA, the UK, Italy, and France, and smaller in countries with higher collectivism and power distance, such as China, Malaysia, and South Korea, with a few exceptions like Japan and Brazil. These gender differences across countries were not explained by country-level gender inequalities indicators (GGGI and GII). Interestingly, the national severity of the pandemic, an epidemiological factor, reduced gender differences in positive emotions. These results underscore the importance of considering cultural and national factors when assessing gender differences in well-being.