Learning in Humans versus Hierarchies


Mihal Emberton, MD, MPH, MS1

Perm J 2020;24:19.083 [Full Citation]

E-pub: 12/18/2019

Editor’s note: Please also see: Unconscious Bias Is a Human Condition and Etiology of Belief–behavior Systems and Hierarchies


Organizations worldwide as well as scholars in business, psychology, medicine, and the social sciences are racing to identify the rules that drive organizational innovation vs status quo, the rules that drive employee engagement vs oppression, and the rules that drive effective leadership vs micromanagement.  This phenomenon is burgeoning because relying on instinctive leadership alone leaves organizations susceptible to the possibility of ineffectual leadership in their hierarchies. To identify these rules, we first need to connect the patterns of how humans learn and solve problems and the patterns of leadership in hierarchies to find the pathways that drive organizational learning vs conformity. Only then can we identify the rules for those pathways, describing how and why learning and growth occurs in any organization or hierarchy. With this operationalized framework, we can now teach supervisors how to recognize the clues and evidence that result from our belief-behavior systems to subsequently convert conformity processes into learning processes, generating organizational innovation and growth as well as employee engagement, loyalty, and trust.


Scholars and intellectuals in the fields of business, psychology, medicine, and the social sciences agree that organizations must engage the employees who are closest to the work in organizational problem solving to drive organizational success and sustainability.1-5 Many of the terms used to describe these patterns of learning and engagement include servant leadership,6 growth mindset,7 kaizen and lean,8 scrum,9 pockets of good practice,10 feedforward,11 agile,12 humble leadership,13 psychological safety,14 and Theory Y.15 On the other hand, what has been lacking in the literature thus far is the identification of how the patterns of learning and the patterns of hierarchies intersect (Figure 1), as well as the identification of the rules for these patterns (Table 1).

The first fundamental rule is that human beings love to find patterns and identify the rules for those patterns; finding the rules and patterns to solve problems and puzzles brings internal, fundamental joy. Just observe yourself and fellow humans: My 7-year-old daughter lights up when she identifies the patterns that allow her to consistently trap an opponent in games of strategy, and I passionately and ardently sorted these fascinating patterns until they fit together precisely. Therefore, we don’t have to teach people how to discover or solve problems; we merely need to create and maintain the environments that support such discovery. Many physicians, for example, sought out the field of medicine because finding the patterns and rules related to health and disease elicits excitement and meaning, but all too often the hierarchies in which we work suppress the processes of discovery, leading to burnout and even despair.16 How and why does this happen?

To understand the burnout and despair that can happen in any organization or hierarchy, we next need to understand the fundamental rules of hierarchies. Hierarchies learn differently from individuals because unlike individuals, who possess all the resources and knowledge for their own learning, hierarchies endow the resources and tools for learning to the supervisors, whereas the knowledge and insights for learning are owned by the subordinates. Additionally, supervisors control the discovery process of subordinates depending on what unconscious belief the supervisors carry when presented with a new idea, new information, or new situation. As supervisors, if we have a democratic unconscious belief, then our feelings, thoughts, and behaviors drive collaborative problem solving with our subordinate. If, however, as supervisors, we have an autocratic unconscious belief, then our feelings, thoughts, and behaviors ensue to force our subordinate to conform to our understanding of the problem, and because we unconsciously believe that we have the right answer, we subsequently believe that no learning is needed.

And finally, we need to understand the fundamental rules that apply to us as supervisors. As supervisors, if we unconsciously believe that we may not have all the knowledge that is needed to solve a problem, a democratic unconscious belief, we will instinctually seek out knowledge from others, interpreting that information as necessary, helpful, and fascinating in enhancing our insights. If, however, as supervisors, we unconsciously believe that we have all the knowledge that is needed to solve a problem, an autocratic unconscious belief, then others’ ideas or experiences are interpreted as unnecessary, worthless, and disruptive. Our democratic and autocratic beliefs are totally unconscious and because of that, when we have toggled into an autocratic belief-behavior system, we are all too often completely and tragically unaware. Additionally, when we demonstrate the feelings, thoughts, and behaviors that we are functioning in our autocratic belief-behavior system, we then often misinterpret the clues and evidence of such as insurgency on the part of our subordinate rather than a direct consequence of the conformity process that our beliefs, feelings, thoughts, and behaviors are demanding. And, as supervisors, we can all toggle between democratic and autocratic unconscious beliefs at any moment.

Interestingly, Douglas McGregor’s XY Theory also recognizes that supervisors toggle between autocratic (X-Theory) and democratic (Y-Theory) behaviors.14 McGregor, however, was thinking that external characteristics such as the type of worker or the work environment is what leads a supervisor to toggle into a certain behavior pattern and that it is a conscious decision14 rather than recognizing that it is the unconscious democratic or autocratic belief of the supervisor that causes the democratic vs autocratic feelings-thoughts-behaviors cascade.

And, yes, although these rules and pathways apply to an organization’s supervisor-subordinate relationships, they compellingly apply to all hierarchical relationships such as parent-child, teacher-student, mentor-mentee, physician-patient, and legislator-constituent. Additionally, restorative justice, overcoming unconscious bias, and the Montessori method are also ideologies that have astutely encouraged supervisors to maintain democratic, not autocratic, belief systems; outlined how to engage in collaborative, not coercive, interactions; and emphasized learning processes over conformity processes to produce just solutions and intellectual development. And, a hierarchy composed entirely of democratically minded supervisors defines social justice and true democracy.



Supervisors interact with subordinates by allowing subordinates to have differing levels of participation in problem solving (engagement-transparency) or by imposing differing levels of control (deprivation-domination), and each spectrum is guided by the acuity of a situation. This concept can be illustrated by exploring simplified physician-patient hierarchical interactions at the extremes of each spectrum. Low-acuity situations, such as a patient with viral upper respiratory tract symptoms (common cold symptoms), can be addressed in 1 of 2 ways. An autocratically minded physician, using interactions toward the full deprivation-no domination side of the spectrum, does not invite the patient to participate in defining the problem or producing solutions and after examining the patient concludes, “You have a cold.” A democratically minded physician, on the other hand, using interactions toward the full engagement-no transparency side of the spectrum, collaborates with their patient to truly understand the underlying problem: Perhaps the patient needs a note for work absence, or the patient wonders if something can help the cough because it keeps him/her up at night, or their patient’s grandmother is getting chemotherapy and the patient wonders if s/he can visit. Then the democratically minded physician makes resources available to collaboratively and effectively solve the problem with the patient.

Conversely, a high-acuity situation, such as a heart attack, can also be addressed in 1 of 2 ways. An autocratically minded physician, using interactions toward the no deprivation-full domination side of the spectrum, looks at the data and proceeds with the algorithm to treat a heart attack. A democratically minded physician, on the other hand, using interactions toward the no engagement-full transparency side of the spectrum, shares his/her understanding of the problem with the patient as well as how s/he plans to address it (no engagement) and then asks his/her patient if the physician is missing anything (full transparency). This “ask” allows the democratically minded physician to check his/her work, check his/her understanding of the problem and solution, and make sure that s/he is not missing a vital patient factor that may affect the outcome of the treatment.

Interestingly, as I began to sort the patterns of supervisor-subordinate interactions, I also started to see how previous efforts to piece together these patterns in the Social Styles Model, for example, did not go far enough. The researchers who created the Social Styles framework also recognized the patterns of interactions between supervisors and subordinates, which they call ask, people, task, tell, but they saw these as fixed behaviors attributable to a personality type.17 What they did not recognize, however, is that the behaviors of people-ask actually represent the engagement-transparency spectrum of interactions that a democratically minded supervisor drives with his/her subordinate, whereas the behaviors of task-tell actually represent the deprivation-domination spectrum of interactions that an autocratically minded supervisor drives with his/her subordinate. When I shared this insight with one of my physician leaders, she agreed that the social styles concept of fixed behaviors does not ring true. She shared her insight that as an emerging leader she initially functioned on the task-tell spectrum, and as she became a more experienced and successful leader, she began to function on the people-ask spectrum. More specifically, she was able to figure out how to toggle from an autocratic belief-behavior system into a democratic belief-behavior system more often.

A supervisor once asked me, “What if I just know that I am right?” The answer is twofold. First, the supervisor must truly value the insights, knowledge, and experience of his/her subordinate. And second, the supervisor can then just ask his/her subordinate whether the supervisor is “right.” As a subordinate, if my supervisor truly values my insights, knowledge, and experience, then I can tell my supervisor if s/he is right or if s/he is missing something. However, as a subordinate, it is painfully obvious when my supervisor is in his/her autocratic mindset, not valuing my insights, knowledge, and experience; in this case, it is safer and smarter to avoid highlighting a knowledge gap because I then avoid the autocratic feelings-thoughts-behaviors cascade.

This autocratic feelings-thoughts-behaviors cascade can be further illustrated by the following theoretical example. Let us say that I want to share my ideas about the hierarchical drivers of burnout with my supervisor because I believe that this knowledge and insight will improve our organizational performance; however, if my supervisor has toggled into the autocratic belief-behavior system, s/he might respond, albeit kindly, with, “Please stop sharing your ideas about the causes of burnout because we already have a group working to understand the causes of burnout, and I feel like you are undermining my authority when you share your ideas.” This autocratic belief-behavior pattern in which the supervisor perceives his/her subordinate as insurgent when the subordinate is merely going out of the way to try to help the organization defines poetic injustice.18 The literary definition, provided by Thaddeus Metz, Professor of Philosophy at the University of Johannesburg, is that poetic injustice is “… characteristically a matter of a person having gone beyond the call of moral duty to help others and then received harm, perhaps of the same sort s/he was trying to alleviate, and (worst of all?) from those s/he was trying to help.”18 The instant a supervisor unconsciously believes s/he has nothing to learn from his/her subordinate is the moment the supervisor has closed him/herself off to discovery.


Understanding the pathways and rules for organizational learning is imperative today because supervisors who are content to function in their autocratic belief-behavior system create substantial liability for their organizations in 2 formidable ways. First, autocratic supervisors, not valuing the knowledge and experience of their subordinates, behave like judges who listen to only 1 side of a case before issuing a ruling, making uninformed, unjust, and generally wrong decisions. Second, autocratic supervisors create suffocating cultures, suppressing and/or driving away talent from the organization because the path of least resistance for subordinates who are repeatedly oppressed is naturally to disengage or even completely extricate themselves from the hierarchy for self-preservation, leaving autocratic supervisors to toil with their own knowledge gaps and risky decisions, rather than undertake the monumental act of heroism or self-sacrifice that is required to elucidate to such a supervisor that they are functioning in their autocratic belief-behavior system and demonstrating knowledge gaps and imperfect solutions.

And unlike the Toyoda family, who instinctually selected and promoted democratically minded supervisors throughout their company from its inception to create a culture of innovation and growth,8 organizations today generally cannot afford to replace all their autocratic supervisors for those who are already intuitively democratically driven. Thus, established organizations must train their supervisors to become more democratically minded more often.

The most obvious place to start such supervisory training is to help supervisors learn to recognize the clues and evidence of our own feelings, thoughts, and behaviors that result from our unconscious beliefs. If, as supervisors, we are feeling excited and curious and we are thinking of protecting and serving our subordinates and solving a problem together, then we are in our democratic belief-behavior system. If, however, as supervisors, we are feeling vulnerable or threatened and if we are thinking of attacking or blaming our subordinates and trying to get our subordinates to do what we want them to do, then we are functioning in our autocratic belief-behavior system. Furthermore, if our subordinates are feeling valued and engaged and they are demonstrating loyalty and trust, then we are functioning in our democratic belief-behavior system. If, however, our subordinates are feeling oppressed or replaceable and they are demonstrating resistance or disengagement, then we are functioning in our autocratic belief-behavior system. When we recognize the clues and evidence that we are functioning in our autocratic belief-behavior system, we can appreciate that we do not have all the information or knowledge needed to solve the problem, and we should try to toggle into our democratic belief-behavior system by genuinely and respectfully seeking out more information, knowledge, and insights from our subordinates.

Less obvious but equally necessary and effective in changing a conformity process into a learning process is to give subordinates access to the resources of time and/or money to comprehend problems and discover solutions. Many technology companies as well as Toyota have recognized that subordinates must have access to the resources for learning if they are to discover, and such companies generated strategies, such as Google’s 20% time policy concept (Google encouraged employees to spend 20% of their time working on projects that the employees themselves believed would benefit Google),19 to give subordinates access to the resources for learning. In medicine, for example, a supervisor could easily apply this principle by allocating organizational time to subordinate physicians for collaboration, problem comprehension, and solution development to improve the health of their populations. As a preceptor, for example, I allow my medical students to own our time together, and thus they choose how to use that time to promote their individual practice improvement as well as our shared learning. Because these rules and pathways apply to all hierarchies, we can also apply them to our supervisory parenting role. As parents, for example, we could give our child time to troubleshoot a problem or provide the resources that the child requests to better understand and solve a problem.

And finally, the least obvious but also necessary and effective way to convert a conformity process into a learning process is to use the tools that are meant to guide learning in a just, rather than unjust, manner. This is, however, where many organizations and hierarchies scramble. In medicine, for example, supervisors routinely and erroneously use outcome metrics (eg, a measure of disease management or cancer screening in a population) as a proxy for indolence, an end point, often penalizing subordinate physicians for not achieving metric goals rather than as a tool to guide a learning process to better understand the barriers that patients experience when accessing the medical system or managing their disease. Recognizing that the learning process is where discovery happens, and a missed outcome or gap in performance just means that more learning is needed, a democratic supervisor rewards subordinates for participating in the learning process rather than penalizes them for missing an outcome or for experiencing a gap in performance. Conversely, when the tools that are meant to guide learning are used as an end point, meaning the autocratically minded supervisor has transferred the responsibility for addressing the tool to the subordinates, and the resources for learning are sequestered from the subordinates, another example of poetic injustice, the organization ultimately perpetuates its unresponsiveness to problems and thus cannot generate effective or innovative solutions. It is not the tools themselves but rather how a supervisor uses those tools that determines if the supervisor is running a conformity process or a learning process.

Another example of how to use learning tools to drive a learning process can be illustrated by the use of feedback evaluations. Supervisors in many organizations routinely and erroneously use feedback evaluations (a summary of an individual’s practice improvement efforts) as a proxy for how the subordinate makes the supervisor feel and for the fragment of learning that the supervisor witnessed or remembers, again, an end point, rather than as a tool to guide a learning process to improve individual skill development and knowledge acquisition. To change such a status quo, as supervisors, we must take responsibility not only for recognizing and resolving our own knowledge gaps but also for recognizing and resolving any deficiencies in our process that result in performance gaps for our subordinates. I do this, for example, by holding myself accountable for the learning of my medical students; if they are not meeting their practice improvement goals while we work together, then I understand that I am not facilitating our learning process ideally, and I must figure out how to be a more effective facilitator of our learning. The tool is not used as an end point to judge the subordinate but rather as a guide for the supervisor to facilitate more learning when more learning is needed. Second, I reward my students for participating in the learning process by asking my medical students to document their performance improvement as well as our learning process on their evaluation. At the end of the rotation, their documentation is what is recorded as their “grade,” and they are justly rewarded for the time and effort they invested in learning. Because my medical students control the documentation about their practice improvement and our learning, I avoid unjustly completing an evaluation with knowledge gaps and subjective feelings. Again, it’s not the tools themselves but rather how a supervisor uses those tools that determine if the supervisor is running a conformity process or a learning process.

By using this operationalized framework, supervisors can convert conformity processes into learning processes, generating organizational innovation and growth as well as employee engagement, loyalty, and trust. We don’t have to teach people how to discover or solve problems; we merely need to create and maintain the environments that support such discovery. I hope that you enjoy trying out these concepts in the workplace and perhaps in your other hierarchies.

Disclosure Statement

The author(s) have no conflicts of interest to disclose.


I would like to thank my colleagues, collaborators, students, reviewers, and editors for their comments and assistance.

How to Cite this Article

Emberton M. Learning in humans versus hierarchies. Perm J 2020;20:19.083. DOI: https://doi.org/10.7812/TPP/19.083

Author Affiliations

1 Adult and Family Medicine, San Francisco Medical Center, CA

Corresponding Author

Mihal Emberton, MD, MPH, MS (mihal.x.emberton@kp.org)

1.    Groysberg B, Slind M. Leadership is a conversation: How to improve employee engagement and alignment in today’s flatter, more networked organizations. Harv Bus Rev 2012; 90:78-84.
2.    Beer M. How to develop an organization capable of sustained high performance: Embrace the drive for results-capability development paradox. Organ Dyn 2001;29(4):233-47. DOI: https://doi.org/10.1016/S0090-2616(01)00030-4
3.    Igel L. Six core principles for creating strong physician leaders. Physician Exec 2012 May-Jun;38(3):42-6.
4.    Sinsky CA, Willard-Grace R, Schutzbank AM, Sinsky TA, Margolius D, Bodenheimer T. In search of joy in practice: A report of 23 high-functioning primary care practices. Ann Fam Med 2013 May-Jun;11(3):272-8. DOI: https://doi.org/10.1370/afm.1531
5.    Ogunlayi F, Britton P. Achieving a ‘top-down’ change agenda by driving and supporting a collaborative ‘bottom-up’ process: Case study of a large-scale enhanced recovery programme. BMJ Open Qual 2017 Oct 26;6(2):e000008. DOI: https://doi.org/10.1136/bmjoq-2017-000008
6.    Chen Z, Zhu J, Zhou M. How does a servant leader fuel the service fire? A multilevel model of servant leadership, individual self identity, group competition climate, and customer service performance. J Appl Psychol 2015 Mar;100(2):511-21. DOI: https://doi.org​/10.1037/a0038036
7.    Brown PC, Roediger HL 3rd, McDaniel MA. Make it stick: The science of successful learning. Cambridge, MA: The Belknap Press of Harvard University Press; 2014. DOI: https://doi.org/10.4159​/9780674419377
8.    Liker J, Convis GL. The Toyota Way to lean leadership, achieving and sustaining excellence through leadership development. New York, NY: McGraw-Hill Education; 2012.
9.    Raps S. Scrum of scrums: Scaling up agile to create efficiencies, reduce redundancies. Defense AT&L 2017; 46:34-7.
10.    Butcher D, Atkinson S. The bottom-up principle. Manage Rev 2000; 89:48-53.
11.    Kluger AN, Van Dijk D. Feedback, the various tasks of the doctor, and the feedforward alternative. Med Educ 2010 Dec;44(12):1166-74. DOI: https://doi.org​/10.1111/j.1365-2923.2010.03849.x
12.    Rigby DK, Sutherland J, Takeuchi H. The secret history of agile innovation [Internet]. Harvard Business Rev 2016 Apr 20 [cited 2019 Oct 16]. Available from: https://hbr.org/2016/04/the-secret-history-of-agile-innovation
13.    Cable D. How humble leadership really works [Internet]. Harvard Business Review 2018 Apr 23 [cited 2019 Oct 16]. Available from: https://hbr.org/20​18/04/how-humble-leadership-really-works
14.    Prottas DJ, Nummelin MR. Theory X/Y in the health care setting: Employee perceptions, attitudes, and behaviors. Health Care Manag (Frederick) 2018 Apr/Jun;37(2):109-17. DOI: https://doi.org/10.1097/HCM​.0000000000000210
15.    Morse JJ, Lorsch JW. Beyond Theory Y [Internet]. Harvard Business Rev May 1970 [cited 2019 Oct 16]. Available from: https://hbr.org/1970/05/beyond-theory-y
16.    Burn-out an “occupational phenomenon”: International Classification of Diseases [Internet]. Geneva, Switzerland: World Health Organization; 2019 May 28 [cited 2019 Aug 9]. Available from: www.who.int​/mental_health/evidence/burn-out/en
17.    Merrill DW, Reid RH. Personal Styles & Effective Performance. Boca Raton, FL: CRC Press; 1981. DOI: https://doi.org/10.4324/9781466593435
18.    Metz T. Why we welcome poetic justice and despair at poetic injustice [Internet]. Boston, MA: The Conversation 2017 Jun 27 [cited 2019 Aug 9]. Available from: http://theconversation.com/why-we​-welcome-poetic-justice-and-despair-at-poetic-injustice-79771
19.    D’Onfro, J. The truth about Google’s famous ‘20% time’ policy [Internet]. Business Insider 2015 Apr 17 [cited 2019 Oct 24]. Available from: https://www​.businessinsider.com/google-20-percent-time-policy-2015-4

Keywords: active lifelong learning, burnout, democracy, effective parenting, employee engagement, feedback evaluation, feedforward evaluation, learning process, organizational culture change, overcoming unconscious bias, patient engagement, social justice, successful leadership, teaching collaboration


Click here to join the eTOC list or text ETOC to 22828. You will receive an email notice with the Table of Contents of The Permanente Journal.


2 million page views of TPJ articles in PubMed from a broad international readership.


Indexed in MEDLINE, PubMed Central, EMBASE, EBSCO Academic Search Complete, and CrossRef.




ISSN 1552-5775 Copyright © 2021 thepermanentejournal.org

All Rights Reserved