Authentic Listening | Emotional Connection: UMSI Research Roundup

University of Michigan School of Information faculty and PhD students are creating and sharing knowledge that helps build a better world. Here are some of their recent publications. 

Publications 

Authentic listening as relational praxis: theorizing from interpersonal, journalistic, ethnographic, and community work to a reflexive framework

Annals of the International Communication Association, March 2026

Jeffrey Lane, Sanna Ala-Kortesmaa, Chelsea Peterson-Salahuddin, Calandra Lindstadt, Mohan J Dutta

This essay advances the concept of authentic listening as a multidimensional and multilevel process of relational, political, and ethical labor. Rather than conceptualizing authenticity as an individual trait or expressive ideal, as is often the case in communication research, we reframe it as a co-constructed achievement that emerges through listening, and is therefore shaped by power, positionality, and the conditions of reception. Drawing on four cases across different domains (interpersonal identity construction, Black feminist journalism, rural Extension-based health engagement, and urban ethnographic research), we trace three interrelated dimensions: authenticity as relational co-construction; authentic listening as emotional, ethical, and epistemic labor; and authentic listening as infrastructure for trust with community partners. Across these sites, listening is not a neutral behavior as it operates as a structure of relationship through which identities are affirmed, knowledge is legitimized, and voice is either redistributed or constrained. We contribute both conceptual tools and a reflexive, five-part framework to assess when listening practices foster inclusion and when they reproduce harm. In doing so, we call for a shift in communication studies of authenticity: from theorizing performance to theorizing reception, and from valuing expression to practicing accountability.

Deep-learning-based de novo discovery and design of therapeutics that reverse disease-associated transcriptional phenotypes

Cell: A Cell Press Journal, March 2026 

Jing Xing, Mingdian Tan, Dmitry Leshchiner, Mengying Sun, Mohamed Abdelgied, Li Huang, Shreya Paithankar, Katie Uhl, Rama Shankar, Erika Lisabeth, Bilal Aleiwi, Tara Jager, Cameron Lawson, Ruoqiao Chen, Matthew Giletto, Reda Girgis, Richard R. Neubig, Samuel So, Edmund Ellsworth,  Xiaopeng Li, Mei-Sze Chua, Jiayu Zhou,  Bin Chen

Identifying drugs that reverse disease-associated transcriptomic features has been widely explored for drug repurposing, but its potential for de novo drug discovery remains underexplored. Here, we present gene expression profile predictor on chemical structures (GPS), a deep-learning-based drug discovery platform, guided by transcriptomic features, that screens large compound libraries and optimizes lead molecules. We first develop a model that captures transcriptomic perturbation signatures solely from chemical structures and deploy it to library compounds. We refine scoring methods and employ a tree-search method for optimization. By incorporating structure-gene-activity relationships, we uncover drug mechanisms from transcriptomic data. We evaluate GPS across multiple diseases and conduct extensive validation in two cases. In hepatocellular carcinoma, we discover two unique compound series with favorable cellular selectivity and in vivo efficacy. In idiopathic pulmonary fibrosis, we identify one repurposing candidate and one novel anti-fibrotic compound by reversing gene expression of multiple distinct cell types derived from single-cell transcriptomics.

A Qualitative Analysis of User Experiences of a mHealth Self-Care Intervention for Care Partners of Individuals with Traumatic Brain Injury

Archives of Rehabilitation Research and Clinical Translation, March 2026 

Rongqi Bei, Christopher M. Graves, Sung Won Choi, Angelle M. Sander, Madison J. Fansher, Jennifer A. Miner, Zhenke Wu, Predrag Klasnja, Mark W. Newman, Noelle E. Carlozzi

Objective: To identify challenges and solutions for conducting mobile health (mHealth) research in a rehabilitation care partner population that involved personalized self-care push notifications and self-monitoring data collection.

Design: Qualitative semi-structured debriefing interviews with participants who completed a six-month app-based self-care intervention involving self-monitoring alone or self-monitoring plus personalized push notifications. Interviews were guided by a data-prompted approach to aid participants’ recall.

Setting: Two academic medical centers.

Participants: 36 care partners of individuals with complicated mild to severe TBI requiring assistance with daily activities.

Interventions: The CareQOL app is a mHealth intervention that delivers personalized self-care push notifications based on daily self-monitoring data, including self-reported caregiver strain, depression, and anxiety, plus Fitbit®-tracked sleep and step data.

Main Outcome Measures: Qualitative interview data about participants’ experiences with, impressions of, and suggestions for the CareQOL app’s: (1) self-monitoring components (i.e., daily questions, health data dashboard); (2) personalized self-care notifications, including overall experience and perceptions of selected messages.

Results: Deductive thematic analysis was used to uncover participants’ experiences with the CareQOL app and identify considerations across intervention and study design. Intervention-wise, push notifications personalized solely based on passive sensing and self-report can be misaligned with participants’ needs, reducing support relevance; limited contextual cues in dashboard displays constrained meaningful self-reflection. Study-wise, technical breakdowns disrupted engagement during the extended study duration; data-prompted interviews effectively supported recall, with participant preferences informing methodological considerations.

Conclusions: Findings identified critical considerations and practical recommendations for designing personalized mHealth interventions for self-care and evaluating them in longitudinal randomized controlled trials.

Closing the digital divide for hemodialysis patients: implementing technology training and support in a digital patient activation intervention

Journal of the American Medical Informatics Association, February 2026

Tiffany C Veinot, Megan Wickens, Edward Hennessey, Marissa Argentina, Kara Eggebrecht, Alicia Zerkle, Vu Le, Kelli Collins Damron, Lisa Velez, Jennifer Bragg-Gresham, Sarah L Krein, Dinesh Chatoth, Michael Heung, Brenda Gillespie, Barbara Murphy, Kai Zheng, Rajiv Saran

Objectives: To detail patient challenges, and how technology support addressed them, in a remote patient activation intervention for hemodialysis patients (n = 93) from trained patient mentors (n = 26). 

Materials and Methods: Using digital divide theory-derived codes, content analysis of: technology support program delivery data, hemodialysis clinic staff interviews, and support staff reflection papers. Descriptive statistics from postintervention mentee/mentor surveys. 

Results: All mentees and 46.2% of mentors received support. Motivational access was targeted with explanations, rapport, and support availability. Study-provided, data-capable tablets enhanced material access, but internet access barriers persisted. Skills access was addressed by training; password-related challenges initially dominated. For usage access, on-demand technology support was balanced by engagement support: proactive prementoring session calls and login monitoring. 

Discussion: Interventionists should examine internet coverage in targeted areas, potentially using multiple carriers. A balance between password usability and security is required. Engagement support may be needed. 

Conclusion: Technology support can close patient digital divides.

How Teacher Educators Adapt Debugging Instruction for Novice Teachers in K-12 Professional Development Practices

SIGCSE TS 2026: Proceedings of the 57th ACM Technical Symposium on Computer Science Education V.1, February 2026 

Tamara Nelson-Fromm, Aadarsh Padiyath, Mark Guzdial

Computing education faces a unique challenge when teaching debugging skills to teachers who do not have a formal computer science background but who will need to learn programming to teach CS courses. While most debugging research about post-secondary learners focuses on preparing students for the technology industry, teacher professional development (PD) programs often serve a different population: teachers who are simultaneously learning computing knowledge and the pedagogical skills to teach it effectively. Through semi-structured interviews with seven facilitators of computing PD programs, this study explores how experienced PD facilitators approach the concept of debugging and instruct teachers in the process of debugging. We use reflexive thematic analysis to show how teacher PD differs from debugging recommendations in post-secondary CS: Rather than focusing on understanding the root causes of errors, facilitators scaffold the teachers' process of identifying and locating bugs to aid teachers in more quickly producing working programs, which they see as important for supporting teachers' confidence. This practical approach acknowledges the time constraints of PD workshops. Finding and understanding bugs is the most difficult part of the debugging process and is something post-secondary students struggle with even after completing one or two semesters of CS courses. These insights challenge assumptions about debugging pedagogy and highlight the need to carefully consider when new CS teachers should learn error analysis skills.

Reflecting on Thematic Analysis in Computer Science Education Research: A Field Guide for Researchers and Reviewers

SIGCSE TS 2026: Proceedings of the 57th ACM Technical Symposium on Computer Science Education V.1, February 2026 

* Best Paper Award

Aadarsh Padiyath, Tamara Nelson-Fromm

Thematic analysis is an increasingly popular method in computing education research; however, widespread methodological confusion undermines its potential. For example, notions of objectivity do not make sense with reflexive approaches, and evidence of saturation is not required for thematic analysis. This position paper details how thematic analysis evolved from Braun and Clarke's influential 2006 work into an umbrella method encompassing three general approaches with corresponding epistemologies: coding reliability (positivist), reflexive (interpretivist), and codebook (hybrid) thematic analysis. Each has different goals and assumptions, but researchers often inadvertently mix incompatible elements. We then present our personal journeys of learning about thematic analysis and finally dissect common confusing claims in our field's publications and peer reviews. Our goal is for the field of computing education research to move towards a ''knowing'' practice. By clarifying thematic analysis approaches and providing guidance for authors and reviewers, we hope to help the field of computing education research better understand this popular method.

The relationship between social media use and perceived social support, loneliness, and emotional connection across 7 countries

Journal of Quantitative Description: Digital Media, February 2026

Brenna Davidson, Nicole Ellison, William Tov, Ryan Ritter

Social media plays a central role in shaping interpersonal connections, yet its association with socioemotional well-being remains widely debated. This study examines the relationship between social media use, loneliness, emotional connection, and perceived social support across seven diverse countries using data from the 2022 Gallup/Meta State of Social Connections Survey. To address inconsistencies in prior research, we employ Specification Curve Analysis (SCA) to assess how different measurements and analytical choices shape how these relationships are reported. Findings suggest limited and context-dependent relationships, with social media use showing a small but positive association with emotional connection and perceived social support, while its link to loneliness remains weak. These results highlight the contextual and methodological complexities of studying social media’s association with socioemotional well-being and emphasize the need for diverse measures and expanded analytical

approaches to fully capture these multifaceted relationships, ultimately enhancing our understanding of how digital platforms intersect with human well-being.

How We Define Privacy Literacy: Teaching Experiences & Challenges of Community-Engaged Privacy Educators

Proceedings on Privacy Enhancing Technologies Symposium, January 2026

Tanisha Afnan, Sheza Naveed, Griffin Christie, Jackie Hu, Byron M. Lowens, Allison McDonald, Florian Schaub

This study examines the pedagogical approaches and experiences of community-engaged educators—individuals who teach privacy, online safety, or security to specific communities through community organizations, companies, or local institutions, such as libraries. We draw on interviews with 21 such educators across the United States and find that, unlike some privacy and security advice that may emphasize knowledge retention of common skills and strategies, these educators prioritized teaching for independent decision-making. Our participants conceptualized privacy literacy as a process for taking informed action, and, from their insights, we identified five core competencies of privacy literacy: (1) data fluency, (2) account security, (3) fraud detection, (4) information vetting, and (5) surveillance capitalism. Notably, these competencies integrate privacy, security, and online safety concepts into privacy literacy—reflecting an increasingly integrated threat landscape. Embedded within the communities they serve, these educators shared their deep understanding of their students’ needs, which varied dramatically, and shared ways in which they tailored their programming accordingly. However, educators also shared significant teaching constraints, including limited time, resources, and organizational support. We discuss the implications of our findings for privacy literacy and for supporting community-engaged privacy literacy efforts.

Pre-prints, Working Papers, Articles, Workshops and Talks

FlyAOC: Evaluating Agentic Ontology Curation of Drosophila Scientific Knowledge Bases

arXiv, February 2026 

Xingjian Zhang, Sophia Moylan, Ziyang Xiong, Qiaozhu Mei, Yichen Luo, Jiaqi W. Ma

Scientific knowledge bases accelerate discovery by curating findings from primary literature into structured, queryable formats for both human researchers and emerging AI systems. Maintaining these resources requires expert curators to search relevant papers, reconcile evidence across documents, and produce ontology-grounded annotations - a workflow that existing benchmarks, focused on isolated subtasks like named entity recognition or relation extraction, do not capture. We present FlyBench to evaluate AI agents on end-to-end agentic ontology curation from scientific literature. Given only a gene symbol, agents must search and read from a corpus of 16,898 full-text papers to produce structured annotations: Gene Ontology terms describing function, expression patterns, and historical synonyms linking decades of nomenclature. The benchmark includes 7,397 expert-curated annotations across 100 genes drawn from FlyBase, the Drosophila (fruit fly) knowledge base. We evaluate four baseline agent architectures: memorization, fixed pipeline, single-agent, and multi-agent. We find that architectural choices significantly impact performance, with multi-agent designs outperforming simpler alternatives, yet scaling backbone models yields diminishing returns. All baselines leave substantial room for improvement. Our analysis surfaces several findings to guide future development; for example, agents primarily use retrieval to confirm parametric knowledge rather than discover new information. We hope FlyBench will drive progress on retrieval-augmented scientific reasoning, a capability with broad applications across scientific domains.

MedViz: An Agent-based, Visual-guided Research Assistant for Navigating Biomedical Literature 

arXiv, January 2026

Huan He, Xueqing Peng, Yutong Xie, Qijia Liu, Chia-Hsuan Chang, Lingfei Qian, Brian Ondov, Qiaozhu Mei, Hua Xu

Biomedical researchers face increasing challenges in navigating millions of publications in diverse domains. Traditional search engines typically return articles as ranked text lists, offering little support for global exploration or in-depth analysis. Although recent advances in generative AI and large language models have shown promise in tasks such as summarization, extraction, and question answering, their dialog-based implementations are poorly integrated with literature search workflows. To address this gap, we introduce MedViz, a visual analytics system that integrates multiple AI agents with interactive visualization to support the exploration of the large-scale biomedical literature. MedViz combines a semantic map of millions of articles with agent-driven functions for querying, summarizing, and hypothesis generation, allowing researchers to iteratively refine questions, identify trends, and uncover hidden connections. By bridging intelligent agents with interactive visualization, MedViz transforms biomedical literature search into a dynamic, exploratory process that accelerates knowledge discovery.

Stochastic Modeling of BMP Heterodimer-Receptor Interactions Shows Emergence of Low-Pass Filtering Behavior

bioRxiv, October 2025

Nissa J. Larson, Aasakiran Madamanchi, Linlin Li,  David M. Umulis

In developing tissues, signal transduction from morphogen gradients conveys positional information to cells, resulting in cell specification and differentiation. One such morphogen is bone morphogenetic protein (BMP), of the TGF-β superfamily, whose signaling network is highly conserved across many species. In Danio rerio (zebrafish), this signaling pathway directs dorsoventral axis formation during early embryogenesis. Many of the molecules that play a role in this network are well-understood; however, the mechanisms through which they achieve noise attenuation and gradient robustness have not been fully defined. Specifically, the heterodimer-heterotetramer complex has been shown to be required for signal transduction[1], but current understanding and modeling of the BMP membrane receptors at this stage has not given any insight into evolutionary drivers of the requirement. In this study, we develop a stochastic model of receptor oligomerization with the published reports of binding kinetics of BMP ligand-receptor interactions to mechanistically assess zebrafish phenotype variability related to the distributions of noise and stochasticity. We can also analyze time-dependent signaling and frequency metrics that are not available in traditional, deterministic modeling. Fast Fourier Transform and cumulative energy spectral density visualization show that the heterodimer-heterotetramer complex may function as part of a low-pass filter mechanism in the dorsal-ventral axis formation process, specifically tuned to the noise of the system. Under dynamic conditions such as the mid-blastula transition (MBT), wherein the morphogen gradient rapidly changes shape, established metrics of noise and information transduction, such as coefficient of variation and mutual information, overlook important temporal effects that may be particularly relevant during development. As the BMP signaling pathway is highly conserved and has been implicated in human bone growth and wound healing, its study in simpler systems stands to accelerate our comprehension of BMP network structure and molecular mechanisms with potential application in regenerative medical studies.

RELATED

Keep up with research from UMSI experts by subscribing to our free research roundup newsletter!