The Empty Corner
What a leadership exercise revealed about uncertainty, expertise, and the capacity we were never taught to build.
There were about fifty of us in the room.
Clinicians, administrators, healthcare leaders with decades of experience between them. People who had spent their careers making decisions under pressure, communicating difficult information, guiding organizations and patients through conditions of radical uncertainty. We had gathered for a leadership development day, and we had just completed an exercise: a series of questions about our habits of mind, our instinctive responses to challenge, our personal strengths.
The strengths were things you'd expect in a room like that. Optimism. Resilience. Creativity. Adventurousness. And one more: Tolerance for Ambiguity.
When the scoring was complete, we were each asked to move to the area of the room that corresponded with our primary strength.
Not one person stood in the corner designated for tolerance for ambiguity.
Not one.
I remember the pull of it — a quiet, energetic whisper that I have learned to trust enough to investigate. And when I took a moment to look around the room, I wondered whether I was the only one standing somewhere safe while some part of me had already decided that it was time to move towards uncertainty.
I have thought about that room many times since. Fifty people who had chosen careers in healing — in the careful, skilled, daily work of accompanying other human beings through their most uncertain and frightening moments — and not one of us had identified our own tolerance for not-knowing as a strength. Not one of us had named ambiguity as something we do well, something we lean into, something we might even lead from.
That empty corner is what I want to talk about.
The medical system — and by extension the broader institutional cultures that medicine has shaped — is built on an architecture of certainty. Diagnosis. Protocol. Evidence-based intervention. The language of clinical life is the language of knowing: assessment, findings, conclusions, recommendations. The clinician who projects confidence is trusted. The one who expresses uncertainty makes patients anxious. And so, early and thoroughly, clinicians learn to perform certainty — even when the clinical reality can be far more probabilistic than the performance suggests.
Diagnosis is rarely as clean as a name implies. Treatment response can be individual and unpredictable. The body, and the mind, resist the precision that the system promises. And yet the performance continues, because the patient in the room needs to believe that someone knows. And so someone performs knowing.
This is not unique to medicine. It is the logic of most institutions that carry high stakes and hierarchical trust. The attorney, the executive, the senior partner — each operates within a culture that has quietly agreed: certainty is what authority looks like. To say I don't know is to risk the room. And so not-knowing becomes something to be managed, minimized, concealed.
But there is a cost to this. And it runs deeper than we usually acknowledge.
The data from that leadership day suggested something uncomfortable: that the path toward a career in medicine may carry a selection bias against people who are comfortable with uncertainty. That the very qualities that draw someone toward clinical work — the drive to understand, to diagnose, to help, to fix — may be, simultaneously, a drive away from ambiguity. We are drawn to medicine, at least in part, because we want to know.
Because not-knowing is uncomfortable, and we have found a vocation that promises, if we study hard enough and think clearly enough, to give us answers.
And then we stand in that room together, and see it reflected back at us all at once.
The research contributes to some concern about the findings in that room. Low tolerance for ambiguity is associated with higher rates of physician burnout, increased anxiety, higher rates of diagnostic testing, and reduced satisfaction. [1] It predicts discomfort with dying patients, greater rigidity in clinical reasoning, and reduced empathy. [2] These are not trivial stakes.
But here is where I want to complicate the story — because the reflex to simply increase uncertainty tolerance, as though more is always better, misses something important. Philosophers and medical ethicists have arrived at a more nuanced position: neither high nor low tolerance for uncertainty is inherently good or bad. A clinician who is too comfortable with not-knowing may delay necessary action, miss the moment when ambiguity has resolved into something that requires a clear and urgent response. The capacity to sit with uncertainty is not the same as the capacity to recognize when it's time to move.
What we are really talking about is not a dial to turn in one direction. It is a form of discernment — the ability to read a situation accurately, to distinguish between uncertainty that calls for patience and uncertainty that calls for decision. The virtues that support this kind of clinical judgment are not tolerance per se, but courage, diligence, and curiosity: the willingness to stay present with complexity, keep investigating, and act decisively when the moment demands it. [3]
The goal is to become calibrated — accurate in your assessment of what you know, what you don't, and what the situation actually requires.
Medicine itself offers a natural experiment here, because not all specialties are built the same way. The classic research on this divided clinical practice along a fault line: surgery and its subspecialties scored lowest on tolerance for ambiguity, while psychiatry, radiology, and anesthesiology scored highest. [4] Family medicine has long been associated with greater comfort in unstructured, undifferentiated presentations — patients who walk in without a clear diagnosis already attached, problems that resolve slowly or not at all, plans that have to flex as new information arrives. The reasoning was straightforward: a field structured around ambiguity will, over time, either select for people who tolerate it or train that tolerance into the people who stay.
It is a tidy story, and like most tidy stories, the evidence is now more complicated than it first appeared. A 2025 reexamination, using the same instruments as the original studies, found that specialty choice explained less than two percent of the variance in uncertainty tolerance — and concluded that the link between the two may be more myth than established fact. [5] Other findings cut in both directions: residency training itself seems to build tolerance over time regardless of specialty, with family practice residents becoming measurably more comfortable with ambiguity between their first and third years. [6]
What this tells us is more useful than a clean correlation would have been. Tolerance for ambiguity is not simply a trait that sorts people into the right specialty before they ever begin. It is something that training — the right kind of training, sustained over years — appears able to build, in people who did not start out with it. The surgeon and the psychiatrist may begin in different places, but both are shaped by what their work asks of them, repeatedly, over time.
Research shows that clinicians with higher tolerance for ambiguity experience significantly lower rates of burnout. The capacity to remain present in uncertainty is not only intellectually generous. It is protective. And leaders who cannot tolerate ambiguity tend to make two characteristic errors: they decide too quickly, collapsing complexity into premature certainty — or they freeze, unable to act without the certainty they were trained to expect. Neither serves the people they lead.
What the research also makes clear is that this capacity is not fixed. It can be taught. Studies across medical schools have found that humanities-based curricula — narrative medicine, art-based observation, literature, ethics — significantly build tolerance for ambiguity. [7] Simulation programs that deliberately introduce irresolution have produced deeper clinical reflection. These are increasingly recognized as core competencies in medical education: the kind of training that produces not just technically capable clinicians, but clinicians who can remain present in the places where technical capability runs out.
The key is not teaching tolerance of uncertainty as passive endurance. It is teaching it as active engagement — the skilled, courageous practice of staying in contact with complexity long enough for something true to become visible. Not-knowing, held with integrity and skill, is itself a form of leadership. The leader who can say I am working with incomplete information — I will stay, pay attention, and revise my understanding as new things become visible is not demonstrating weakness. They are demonstrating something rarer, and more trustworthy, than performed confidence.
But most of us in that room did not receive formal training in any of this. The humanities curricula, the reflective simulations — they weren't there when we trained. We were handed the architecture of certainty and told to inhabit it. And now, decades in, we are being asked to renovate from the inside.
So what, practically, can a seasoned clinician actually do?
Mindfulness-based practice. Mindfulness-based interventions for healthcare professionals have shown consistent effects: reduced emotional reactivity, greater capacity to remain present under pressure, and increased tolerance for uncertainty in both personal and professional contexts. This is not about meditation as a wellness add-on — it is neurological change over time. [8] Uncertainty, after all, is not only a cognitive state. The body holds it first — in held breath, in tightened shoulders — long before the mind finds language for what it's already sensing. A growing number of mindfulness programs are now designed specifically for clinicians, many offering CME credit, and are worth seeking out as continuing education that addresses not just knowledge, but the practitioner behind the knowledge.
A coaching approach — for yourself and your team. Coaching may be one of the most underutilized tools for building uncertainty tolerance in clinical and leadership contexts — precisely because it is designed, at its core, to work in unresolved space.
The International Coaching Federation's core competency model describes the coaching relationship in terms that map directly onto what the research says builds uncertainty tolerance. A coach is required to embody a coaching mindset — remaining open, curious, flexible, and client-centered, with an ongoing reflective practice and the ability to regulate one's own emotional responses. A coach maintains presence in a way that explicitly includes being comfortable working in a space of not-knowing, and creating space for silence, pause, and reflection rather than rushing toward resolution. Curiosity is not a personality trait you either have or don't — it is a practice, and the coaching relationship makes it a discipline. These are not incidental features of good coaching. They are the thing itself.
For leaders, individual coaching builds this capacity in two directions simultaneously. The coaching relationship becomes a practice environment — a place where uncertainty can be named, examined, and held without the pressure to perform certainty, and where saying I don't know becomes available in high-stakes moments because it has been practiced in lower-stakes ones. The skills the coach models become available to the leader: powerful questioning that opens rather than closes thinking; active listening that attends to what is not being said as much as what is.
For teams, a leader who brings a coaching approach to their work creates psychological safety around not-knowing. When a leader asks what are we not seeing here? rather than projecting certainty they don't feel, they signal that uncertainty is a shared condition to be navigated together — not a failure to be concealed. This is, in the language of the ICF competency model, cultivating trust and safety. The research on uncertainty tolerance consistently points to normalization as one of the most powerful interventions available: when not-knowing is treated as a legitimate part of clinical and organizational life rather than a deviation from competence, its grip loosens. A coaching approach — whether received individually or practiced within a team culture — creates exactly this environment. It does not eliminate uncertainty. It changes our relationship with it.
Deliberate engagement with the arts and humanities. The evidence is specific: engagement with narrative, visual art, literature, and music builds the cognitive flexibility that underlies uncertainty tolerance. It is not the content that matters so much as the practice of staying with something open-ended — resisting the pull toward resolution, holding multiple interpretations without collapsing them into one. Narrative medicine has been shown to build exactly this capacity in clinicians at any career stage. Reading fiction. Looking carefully at a painting. These are not recreation. They are training.
None of these eliminate the discomfort of not-knowing. That is not the goal. The goal is calibration — an accurate, honest relationship with the limits of your own knowledge, and the confidence to act wisely within those limits rather than performing certainty you don't feel.
I want to return to that room one final time.
What struck me most was not the absence — not one person — but what the absence revealed. Not a failure of character. A failure of formation. We had been trained, selected, and socialized into a professional culture that made uncertainty something to overcome rather than something to develop a relationship with. The corner was empty not because we lacked the capacity, but because no one had ever told us that capacity was worth building.
That is changing. Slowly, unevenly, and not fast enough — but it is changing. The question for those of us already deep in our careers is whether we wait for the system to catch up, or whether we begin, now, the work of growing something that was never formally cultivated in us.
Tolerance for ambiguity is not a fixed trait. It is a capacity. And like all capacities, it responds to attention.
The clinician who can do this is a better clinician. The leader who can do this is not a more uncertain leader. They are a more present one.
What if it's not certainty that heals — but presence?
References
Hillen, M. A., Gutheil, C. M., Strout, T. D., Smets, E. M. A., & Han, P. K. J. (2017). Tolerance of uncertainty: Conceptual analysis, integrative model, and implications for healthcare. Social Science & Medicine, 180, 62–75. https://doi.org/10.1016/j.socscimed.2017.03.024
Kvale, J., Berg, L., Groff, J. Y., & Lange, G. (1999). Factors associated with residents' attitudes toward dying patients. Family Medicine, 31(10), 691–696.
Reis-Dennis, S., Gerrity, M. S., & Geller, G. (2021). Tolerance for uncertainty and professional development: A normative analysis. Journal of General Internal Medicine, 36(8), 2408–2413. https://doi.org/10.1007/s11606-020-06538-y
Geller, G., Faden, R. R., & Levine, D. M. (1990). Tolerance for ambiguity among medical students: Implications for their selection, training and practice. Social Science & Medicine, 31(5), 619–624. https://doi.org/10.1016/0277-9536(90)90098-D
Wegwarth, O., Pfoch, M., Spies, C., Möckel, M., Schaller, S. J., Wehler, M., & Giese, H. (2025). Tolerance for uncertainty and medical students' specialty choices: A myth revisited. Medical Education, 59(8), 833–841. https://doi.org/10.1111/medu.15610
DeForge, B. R., & Sobal, J. (1991). Intolerance of ambiguity among family practice residents. Family Medicine, 23(6), 466–468.
Mangione, S., Chakraborti, C., Staltari, G., Harrison, R., Tunkel, A. R., Liou, K. T., Cerceo, E., Voeller, M., Bedwell, W. L., Fletcher, K., & Kahn, M. J. (2018). Medical students' exposure to the humanities correlates with positive personal qualities and reduced burnout: A multi-institutional U.S. survey. Journal of General Internal Medicine, 33(5), 628–634. https://doi.org/10.1007/s11606-017-4275-8
Epstein, R. M. (2017). Attending: Medicine, mindfulness, and humanity. Scribner.
