Why is deep learning so hard?

It is not easy to figure out how to learn deeply. In the early stages of developing your identity as a deep learner, one of the smoothest ways to engage in deep learning is to seek out skilled coaches who understand the principles behind deep learning. If the mentors you find are good coaches and have many years of practice in the art of teaching, they will likely have collected effective techniques to guide you in deep learning and effective repetition. Assuming these teachers are willing and able to act as your mentor, you can leverage their guidance simply by following their directions. In this scenario, all you have to do is give your best effort on pre-designed activities that specifically engage you in deep learning. Sadly, this “simple” solution of finding master coaches who are also content experts is out of reach for most novice learners. This is especially true in the context of the US higher education system. In this post, we explore some substantial barriers to seeking out expert coaching while earning your college degree.


The US higher-education system maintains policies to ensure that content experts who get hired as full-time professors likely know relatively little (compared to their content expertise) about the science of learning and art of teaching.

This first problem highlights the fact that it is possible to become a tenured-track professor at almost any accredited college in the United States with very little knowledge about the science of learning, training in the art of teaching, or understanding of the political forces that shape the nature of their job as a professor.

The minimum job qualifications for a professorship at most US colleges is simply to be a content expert in a specific field. Members of search committees tasked to hire a full-time professor look for recently graduated students who earned multiple degrees in a certain subject area and also amassed impressive accolades while achieving their education. This expertise looks great on paper and does indeed represent a monumental achievement in our society.

However, if we look more closely, the coursework and degree requirements needed to earn most undergraduate, graduate, or professional degrees include almost no formal training on how learning works or in the art of teaching. This leads to a reality in which the vast majority of professors at US Colleges start their careers in the classroom without knowing much about how to teach effectively for diverse student populations. And yet, the moment these young professors start their work as teachers, they assume responsibility for guiding the learning of the next generation of college students.

I like to say that such policies and hiring practices result in a system that is run by the andragogy of introspection. I define this to be practice of designing classroom policies by reflecting internally on your own lived experiences. The central questions asked by professors who uses the andragogy of inspection as a guiding philosophical framework in their teaching include:

  • What teaching policies did my own college professors use in the classes I took?
  • How did those policies work for me?
  • How can I recreate the policies my college teachers used on me as I teach in my own classes?

This type of myopic decision making is completely logical in a system that puts young professors in an impossible work environment. When newly-hired professors are dumped into a classroom and find themselves responsible for inspiring tens or even hundreds of students to learn deeply, what other choice do they have but to rely on introspection to guide their teaching decisions. However, andragogy by introspection is not an effective way to inspire deep learning for a diverse group of students.

Any person who is hired in a full-time position as a college professor was most likely one of the highest-performing students while they earned their own undergraduate, graduate, or technical degrees. Because of inequitable structures in so many aspects of modern-day societies, this high performance is likely tied to socioeconomic, racial, family, and geographic factors that serve to bolster such achievement. In other words, statistically speaking, people who get hired as full-time college professors probably enjoyed an upbringing filled with privileges and support that made deep learning a much easier task. Those factors bolster high performance independent of the learning environment. I should say explicitly, there are professors who come from diverse backgrounds and overcome significant barriers to ascend to the upper echelons of our education system. However, if we look at the ratio of the number of these type of professors divided by the total number of tenured college professors in the US, that ratio is far smaller than it ought to be.

This leads to some large error in thinking about how to meet the needs of diverse students. If the guiding framework for constructing policies in college classes is to reflect on one’s own lived experience, then the decisions that result will center dominant social constructions that have little to do with the lived-experiences and learning needs of diverse students sitting inside the classroom.


College professors at most R1 Doctoral Institutions feel substantial pressure to publish academic work as part of their goal to earn tenure. This pressure is known as the publish or perish phenomena. Such professional pressures disincentivize meaningful focus on the scholarship of teaching and learning.

After reading about Problem 1 above, you might see one possible solution: let college professors train themselves to be better teachers. Anyone who earns a tenure-track position in the academy is very likely an expert learner. Such scholars can no doubt teach themselves anything they have an interest in learning. It stands to reason that if there are huge gaps in their knowledge about teaching and learning, then these professors could decide to do deep research into these topics.  Anyone that is well-qualified to earn a job as a college professor has the ability train themselves to become master educators by engaging in years of intense study.

However, the question of how to get college teachers to take pride in the scholarship of teaching is not about capacity. This is a question about incentives.

Most college professors do their best to build and take pride in their career. By the time they land their tenure-track job, they have probably spent between 18 – 21 years in the education system. During this journey, they likely invested significant intellectual and emotional energy to develop their identity within the context of the existing system. They likely want to earn the type of job security afforded by tenure and they probably also seek validation for the work they do.

It is within these desires that a whole other set of policies practically guarantee that most college professors continually neglect a scholarly study of teaching and learning in favor of content-centered research. Take a look at the following two articles describing how to get tenure at top Universities and also about the purpose of research universities:

How to Get Tenure at a Major Research University by Sean Carroll

The Purpose of Harvard is Not to Educator People by Sean Carroll


In the US, college professors train the vast majority of undergraduate and graduate students in our society. This includes teachers, doctors, lawyers, engineers, scientists, and anyone with higher-education credentials. What results is a system of social hierarchy in which college educators have huge influence over the entire structure of society.

This article is mainly concerned with analyzing major policy barriers that prevent college professors at R1 Doctoral Institutions in the United States from centering the scholarship of teaching and learning as the heart of their professional practice. To identify an institution as an R1 Doctoral Institution, I use the Carnegie Classification of Institutions of Higher Education.

I choose to focus on R1 institutions because these spaces take up so much oxygen in our popular culture. Tenure-year track jobs at these institutions are the most coveted positions in the academy. These spaces enjoy the best research funding. R1 Doctoral Institutions routinely secure much larger revenue streams from governments, industry, and the larger public compared with their smaller counterparts. Moreover, R1 institutions do a great job of marketing their brand to create intense competition for entrance to their undergraduate programs and are integral in the degree-to-career pipeline. In other words, when it comes to influencing the experiences of an average college student and cultural expectations for what it means to be a college professor, the R1 Doctoral Institutions enjoy a monopoly on main stream attention. Much of what I write in this post does not exactly apply to college professors working at non-research-based intuitions. These might include liberal arts colleges, small private colleges, two-year colleges, and state schools. It tends to be the case that college professors at those types of schools take much more pride in their teaching prowess and thus create more student-friendly learning environments in their classrooms. Even while this is true, many of the people that earn full-time jobs at these smaller institutions were trained at R1 institutions. The fabric of their intellectual identities was cut with the cultures of these schools in mind and they bring parts of the culture anywhere they land.

I end this post with the statement of some other problems that effect your ability to learn deeply. I do not expand on these in this post, leaving that task for future discussions.


Starting in the late 1960’s and continuing until present day, state and national governments have been captured by a neo-liberal philosophy to government. This approach centers on turning to private industry to solve public problems. In academia, this approach to government has lead to a massive divestment from education which has caused rising tuition costs and placed substantial pressure on colleges to provide much more support to students with substantially less funding.

In the United State, we have a long history of political systems that center and privilege specific socially constructed identities. These include colonialism, racism, classism, and sexism. These forms of oppression tangle together to create a society in which many policies are designed to benefit rich white men at the expense of the rest of society. While this problem exists in all corners of society, it is particularly pernicious in shaping the experiences, hopes, dreams, and fears of college students trying to navigate the US college education system.

Community Challenge

  1. Click on all the hyperlinks in the article above and read all articles in these links.

  2. Come up with your own descriptions for each of the problems listed above. Write these out in your own words using abuelita language: language that your abuelita (grandmother) can understand.

  3. Think about your own experience in college classes. How have these problems effected your learning?

  4. Where and how do these problems show up in your daily life in college? Do your best to identify specific policies your teacher implements that you feel inhibit your ability to learn deeply. Then, try to map these policies back to the problems listed in this post.

6 thoughts on “Why is deep learning so hard?

  1. If one of few students who benefits from an unfair system is now a professor, he/she will only reflect on their experiences from their point of view and continue the same system as it was successful for them. The outdated system is thus constantly perpetuated. More incentive on studying pedagogy could change that, but that requires major policy changes. ):

    Liked by 1 person

  2. From attending Tiago’s book club zoom event 2/18/21, he said: “Amount of critical / analytical attention is our most scarce resource… that’s the bottleneck to intellectual work” and humanity moving forward. Interesting to think about that live quote as we help our readers make the most out of their education, and what incentives are at play here.

    I just listened to a podcast on how the incentives they give architects around the world are leading to an increase in buildings collapsing… bonuses are only given to bigger yet cheaper buildings, when their base pay is slowly decreasing… yikes. I also recall hearing in passing (hehe): when a prof. signs up to teach 3 math 2A and 3 math 2B sections all in 1 quarter, due to the incentives of registering for a larger teaching load… means that prof’s quality of service per student will inevitably suffer…In other words, it’s extremely difficult to achieve: flow—”the state in which people are so involved in an activity that nothing else seems to matter; the experience itself is so enjoyable that people will do it even at great cost, for the sheer sake of doing it.” – Mihaly Csikszentmihalyi pg 11 of my version, when a prof is truly responsible for over 20 students learning needs in one Q, let alone 150+ lol… Not only does the prof self-sabotage the deep learning possibilities for themselves, but also for their students.

    Incentives are often overlooked in industry and academia… which is so relevant for students and educators when they incentivize deep(er) or shallow(er) learning… Now I see why you study motivation and how that’s a career-long problem to understand, in and of itself haha.

    Incentives also extend beyond the tyranny of shareholder supremacy or merit, because incentives can actively drive a person’s values… It took me some time to realize what you meant by daily habits matter when it came to choosing where to be an educator, but I’m slowly getting it now…

    The longer I work with you (thank you for every mili-second of your presence) and the TLC family, the more I understand the sheer amount of important (deep learning) & urgent (just in time, as well as long term) work there is to be done at the Community College. As much as I want you to empower more students, aka continue supporting you in your full teaching load, now, I want to help you find ways to buy out your time and write the best Linear Algebra / MATlab textbook, curriculum, video course ever replicable & adaptable by other educators, all while teaching, and doing what else you think is essential to this work (I know there’s a ton more you c, that I don’t yet!). I should show you the assigned or recommended readings in these CS courses I’m in right now… an utter catastrophe: from non-effective communication, lack of human dimension, no genuine assessment for learning progress, to non-learner friendly textbooks to begin with…

    I am also better understanding why reducing the teaching load could and would be essential to the course & curriculum development for present & future students. I feel like the 11 students I’ve had last weekend, that’s (easily) already more than enough to genuinely plan-act-reflect on for 1 sem. ! After listening to about 80+ Episodes of The Cult of Pedagogy Podcast over the past 2 years, this burn out / restricting classroom freedom / not student-centered curriculum ideas were articulated in so many different ways by educators, and it’s no wonder the one’s w the “normal” amount of soul and heart end up quitting, and the most who stay get that human dimension drained from them in experience w STEM undergrad courses anyway.

    I love how you designed your Linear Algebra and MATLab course where you literally can “get out of the way of your students” intelligence, after thousands of hours developing the course, activities, and experiments… and focus on igniting (motivating) them to find their sweet spot (productive struggle zone as you call it)… all while giving them individualized (small group) attention live during precious class meeting time, by celebrating student practices, on what and how each 1 of them are doing or not doing, to learn!!! holy macaroni you raised the bar, but you’re definitely leaving me big Reese’s Pieces. #SugarBad

    Can’t wait to see your students learning portfolios in their full beauty. I am aware of how hard it must have been to communicate that to students, especially during these times… but you’re doing it You may invite a few students to do a guest sponsor blog post or interview to document this breakthrough of a Q you’re having Jeff!

    Ps. recently re-listened to your Teaching Code podcast on How you became a teacher… so good. Real talk, consider freestyling through your anchor app on your phone on your walks to capture the multiple breakthroughs that are occurring for you and your learners this Q! I am afraid if you don’t capture now(weekly or biweekly) in some way shape or form, you may lose some of the breakthroughs in their full glory!

    Oh a fun idea that came up is hosting/live streaming (something special about ‘live’ events) a movie night for, 1 or maybe all your students together across 3 classes (even past classes), the movie could be stem related or not! Just a thought to humanize the virtual learning exp. and build a community for you & your learners! Knowing how you been conducting your classes this Q, and you as an educator, I think you will have a wonderful turnout! I also remember u asking me to note down you wanted to assign future Eng 11 classes to watch Hidden Figures the movie, on their own time heh…


  3. Amy J Ko “in my 20 years of peer review experience, I generally don’t see experiments that tackle the root cause of most problems in peer review: the incentives.

    The incentive problem is pretty simple, and has its roots in academic institutional goals and the broader social structure of collective discovery. Researchers are incentivized to make progress and disincentivized to evaluate progress.”

    for more depth on problem 2

    View at Medium.com


    1. Yup! It’s amazing at how robust this system is at maintaining status quo.

      I contend that we have to go deeper than incentives into the underlying beliefs and world view. The incentive problem is at the surface level about how do we create external motivators to manipulate and control the behavior of top actors in the system. The deeper issue is policy design and what beliefs are encoded into the policies we choose. Our current system is designed to replicate the exact outcomes we are creating.

      When we make private the costs of education, we are encoding a belief about wealth, priviledge, and access to power. I have so much more to say about this but I’m in the middle of other work.

      Cool to see Amy Ko thinking about these issues! It makes sense to me that she publishes this not in a journal but in her medium writing. Cheers to this conversation and to the next generation of writers and thinkers who will need to really think deeply about this to make changes.

      Thanks for the pointer Henry!


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s