Mentorship is often portrayed as the invisible engine of graduate education, the quiet force that turns anxious first-year students into confident researchers. But what happens when the mentors and the students being mentored hold fundamentally different ideas about what mentorship should actually look like? A new study published in Biomedical Engineering Education tackles that question in one of the most complex training environments in modern science: a multi-institutional Engineering Research Center, or ERC, funded by the U.S. National Science Foundation, where biomedical engineering students work across laboratories, departments, and universities that each operate under their own unwritten rules. The findings reveal a striking pattern of agreement on the big picture and deep misalignment on the details, a combination that researchers say could reshape how interdisciplinary research centers think about supporting the next generation of scientists.
The study, conducted by Hwangbo Bae and Joi-Lynn Mondisa of the University of Michigan, used an approach called interpretative phenomenological analysis, a qualitative method designed to capture how people make sense of their own lived experiences. The researchers interviewed 20 participants from a single ERC in its sixth year of operation: 13 student trainees and 7 faculty mentors drawn from partner institutions that included two predominantly white institutions and one Hispanic-serving institution, all classified as R1 doctoral universities with very high research activity. Rather than analyzing mentor-mentee pairs directly, which would have risked identifying individual students in a center of modest size, the team developed group-level themes for each population separately and then compared them, a design choice that protected anonymity while still exposing where the two groups’ perceptions converged and collided.
The methodological toolkit was unusually rich for a mentoring study. Student trainees completed pre-interview surveys and drew sociograms, visual maps of their mentoring networks in which the people placed in the innermost circle were those they found most supportive or influential. During hour-long semi-structured interviews, participants used these drawings as prompts to explain the nature, frequency, and value of each relationship. Faculty mentors, by contrast, were asked about their mentoring practices and strategies rather than being asked to draw, a decision the researchers made to keep faculty engaged and to avoid prompting them to present their mentoring as uniformly supportive. Reflection notes written immediately after each interview captured contextual details, and the analysis proceeded iteratively from exploratory notes to personal experiential themes and finally to group experiential themes that could be compared across the two populations.
The theoretical backbone of the study was Expectancy-Value Theory, a motivational framework holding that people engage most deeply in activities when they believe they can succeed at them and when they perceive those activities as valuable. Applied to mentorship, the theory suggests students invest in mentoring relationships when they expect those relationships to advance their academic and professional goals and when the interactions themselves feel worthwhile. Faculty, meanwhile, often see their role as professional stewardship, preparing the next generation of scholars, but they report time constraints and institutional pressures that can limit their availability. That tension, the researchers suspected, might be where expectations begin to drift apart, and the data largely confirmed it.
Three domains of shared expectation emerged with remarkable consistency. The first was the mentor as connector. In a conventional single-lab setting, collaboration opportunities depend heavily on a faculty member’s personal network, but ERC mentors leveraged the center’s multi-institutional structure to deliberately link trainees with researchers, resources, and projects across campuses. One faculty mentor described encouraging students to maintain multiple small collaborations alongside their main project, complete with structured design reviews, precisely to build skills the mentor considered essential in the real world. Students echoed this, describing how mentors arranged introductions to professors outside their own labs, bridged connections they would have found intimidating to initiate themselves, and, in one postdoc’s words, wondered aloud whether principal investigators should more intentionally broadcast their trainees’ expertise so collaboration opportunities could find them.
The second shared domain was mentorship that extends beyond research supervision. Both groups emphasized career development and psychosocial support as central to the mentoring relationship, not peripheral to it. Students described mentors who checked on their mental health and financial well-being, offered tailored advice on post-graduation pathways, and encouraged them to learn skills like proper software documentation even when those skills would never appear in a publication. One trainee described a mentor who looked up the start date of Ramadan each year and proactively offered to adjust meeting schedules, a small gesture the student found deeply respectful. Another recounted a painful episode of racial discrimination in their own building, being mistaken for a food delivery worker, and a mentor who responded concretely by arranging a staff jacket typically reserved for professors, after which the incidents stopped. The researchers noted that such support buffered the emotional toll of discrimination and contributed directly to the trainee’s persistence and confidence.
The third common domain was the collaborative lab environment, in which mentorship was deliberately distributed rather than concentrated in a single advisor. Mentors described cultivating open, inclusive group cultures where students felt safe speaking up, celebrating every accomplishment from a first undergraduate poster to a new patent in shared group chats. Senior students and postdocs served as accessible, hands-on sources of guidance, particularly in large labs where the principal investigator’s time was stretched thin. Mentors also described handing real responsibility to senior trainees, explicitly telling them they had the reins to guide newer students, a gesture those trainees read as a powerful signal of trust that strengthened both the lab community and their own development as emerging mentors.
Yet beneath this consensus, two significant misalignments surfaced. The first concerned mentor accessibility. Faculty mentors described mentorship as a dynamic process that should shift over time, moving from close, hands-on guidance for novices toward greater autonomy as mentees matured. Many students, however, even experienced ones, still craved direct engagement. One of the most senior graduate students in a large lab described her relationship with her mentor as formal, saying she saw him as a boss, hesitated to seek help because she viewed his time as too valuable, and preferred to figure problems out herself or ask someone else. Crucially, the data showed that limited availability did not automatically read as inattention: another doctoral student acknowledged her mentor’s many responsibilities but expressed complete confidence that a single message would guarantee a meeting. What mattered, the researchers concluded, was not the frequency of meetings but the mentor’s demonstrated commitment and the quality of each interaction.
The second misalignment centered on mentee independence, and it exposed the most philosophically fraught question in the study. Both groups valued autonomy, but navigating it proved genuinely difficult. Faculty mentors admitted uncertainty about when to intervene, with one describing the dilemma candidly: too much guidance might hinder independent thinking, but too little leaves students feeling unsupported, and he had not yet figured out the right approach. The same mentor noted that his strongest trainees were generally not the ones he had guided heavily early on, yet some struggling students later told him earlier guidance would have helped. Another mentor, who prided himself on giving students space to make choices and mistakes, acknowledged that he had recently failed to recognize a student who was not progressing. Students, for their part, sometimes wanted more structured scaffolding when facing unfamiliar cross-disciplinary tasks, suggesting that in interdisciplinary environments independence cannot simply be granted but must be intentionally scaffolded.
The study’s implications reach well beyond a single research center. The authors argue that ERCs function less like traditional advisor-centered models and more like mentoring ecosystems, interconnected networks in which no single mentor fulfills every role and support flows from a constellation of faculty, peers, postdocs, and structured programs such as student leadership councils and journal clubs. Notably, the center’s written mentoring guidelines and Best Practices Manual were almost never invoked by participants, who described practices grounded instead in individual style, an observation the researchers link to broader arguments that training resources do not translate into daily practice without systems of implementation and accountability. Their recommendation is deceptively simple: early, intentional, structured conversations in which mentors articulate their long-term goal of developing independent researchers and mentees share their goals, concerns, and preferred modes of guidance. In settings where no single set of institutional norms governs mentoring, such explicit alignment may be the difference between a mentoring ecosystem that thrives and one where students quietly hesitate at the door of a busy professor’s office, unsure whether they are allowed to knock.
Subject of Research: Mentoring expectations and experiences of faculty and students in a multi-institutional biomedical engineering research center
Article Title: Examining the Alignment and Misalignment of Mentoring Expectations Between Faculty and Students of a Multi-institutional Biomedical Engineering Research Center
Article References: Examining the Alignment and Misalignment of Mentoring Expectations Between Faculty and Students of a Multi-institutional Biomedical Engineering Research Center. (n.d.). https://doi.org/10.1007/s43683-026-00254-5
Image Credits: AI Generated
DOI: 10.1007/s43683-026-00254-5
Keywords: mentorship, biomedical engineering, Engineering Research Center, graduate education, interdisciplinary research, faculty mentoring, student trainees, Expectancy-Value Theory, qualitative research, interpretative phenomenological analysis, STEM education, mentoring ecosystems
News Source: Richard Spencer. (October 13, 2026). When Mentors and Students Disagree: Inside the Hidden Fault Lines of Biomedical Engineering Training. Scienmag.
