By Dana Dawson, Ph.D.
In his 2018 book How Humans Learn: The Science and Stories Behind Effective College Teaching, Joshua Eyler (our 2026 Annual Faculty Conference keynote speaker) observes a striking paradox at the heart of academia: while scholars understand that failure is a primary engine of discovery in research, our educational systems go to great lengths to stigmatize it for students. Eyler writes: “As academics, we are trained to learn from our failures, to use each as an opportunity to refine our hypotheses and to advance our understanding, until we reach a point where the collective gains we have made from these failures come to be labeled “success.” The same is true of our work in the classroom. We (ideally) try out a new teaching strategy or a new assignment, determine whether it has helped our students learn, and then we continue to refine these approaches to achieve better results. Why is it, then, that our educational systems stigmatize failure so profoundly for students themselves?” (172-173).
We routinely design courses and assessments in which small errors can have outsized consequences on a student’s overall grade, ensuring mistakes are seen as something to be avoided rather than as possible learning experiences. We avoid sharing our own messy processes, failures and doubts with students, who see only our published manuscripts and completed lectures.
If we want our students to engage deeply with course material and develop the resilience needed for a lifetime of learning, we need a paradigm shift toward what Eyler calls a “pedagogy of failure” (174). In this post, I share recommendations from Eyler and the scholars who have inspired his work for creating learning experiences that leverage failure to achieve lasting results.
1. Distinguish Between Performance and Learning
One of the most persistent hurdles in teaching is the “incommensurability” between performance and learning. As Richard Schmidt and Robert Bjork argued in their seminal 1992 article, a learner performing well in the moment might be benefiting from conditions that actually undermine durable learning, while a learner who appears to struggle might be engaging in exactly the processing that builds lasting skill. Schmidt and Bjork’s research shows that what looks like failure in the moment often signals deeper cognitive processing happening beneath the surface. This challenges a common instinct in teaching: rather than smoothing the path and ensuring students succeed at every step, we might better serve their learning by building in productive struggle and treating apparent setbacks as part of the process, not problems to eliminate.
As instructors, we can help students by explaining this distinction. When a task feels too easy or “frictionless,” students might be experiencing what Bjork calls “fluent” learning, which often leads to rapid forgetting. We should encourage them to value the struggle, reminding them that a “performance dip” or a high error rate during practice is not a sign of failure in the traditional sense; it is often the very signal that deep, durable learning is taking place.
2. Aim for “Productive Failure”
This insight is expanded upon by Manu Kapur who finds that students who struggle first and receive instruction second outperform those who receive instruction first on measures of conceptual understanding and transfer. What Kapur calls “productive failure” involves engaging students in solving complex problems involving concepts they haven’t learned yet. While they will almost certainly fail to find the correct solution, the resultant struggle will help students confront the limits of their current knowledge and discern critical features of the problem.
Allowing students to struggle first prepares them for the “direct instruction” that follows because they have already seen why the concept is necessary to solve the problem. This is preferable to “unproductive success,” where students achieve a correct answer without the cognitive “friction” that makes the knowledge stick.
3. Cultivate an “Error-Positive” Climate
If we want students to take the risks necessary for productive failure, Eyler suggests we must actively manage the “error climate” of our courses to ensure students feel psychologically safe.
To build this climate, we can:
- Normalize struggle: Tell students that feeling like an “imposter” or feeling confused is a normal part of learning and share occasions where you struggled in your academic journey to understand a concept, complete a manuscript, make sense of findings, and so forth.
- Provide frequent low-stakes opportunities: Design assignments where students can fail and then subsequently build conceptual understanding as a result of that process, without it devastating their grade.
- Decouple feedback from grades: Look for opportunities to give feedback independent of grades so that efforts to help students learn from failure can be more fruitful.
4. Leverage the Power of the Pre-test
One practical way to implement failure-based pedagogy is through the pre-test. While we often think of pre-tests as a way for us to gauge student knowledge, they are also a powerful learning tool for the students themselves. Even if students get every question wrong, the act of attempting to answer them facilitates later learning of that material.
A pre-test forces students to “confront the limits of their knowledge”. It serves as a roadmap, signaling to the brain what information is important to pay attention to in the coming weeks. To design a useful pre-test, Eyler recommends identifying 10–20 concepts that students must know to succeed and building short multiple-choice or short-answer questions around them.
In closing…
The need for failure-friendly pedagogy is particularly important now, as generative AI becomes ever more ubiquitous. Marc Watkins’ points out on his Substack that we are currently being flooded with “frictionless” interfaces like ChatGPT, Grammarly and Claude that allow students to bypass the “productive failure” phase entirely. If learning is, at its core, “friction,” then a technology that instantly provides an answer is a technology that may inadvertently stymie deep learning. As Watkins suggests, we need to “advocate for slowness” and integrate moments of “built-in inquiry” into the learning process.
The capacity to err is also, according to Kathryn Schulz, inextricable from qualities like imagination and courage. By making our classrooms friendlier to failure, we aren’t just helping students get better grades in the long run. We are teaching them how to be resilient and curious, and how to navigate a world that is increasingly defined by rapid change and “frictionless” answers. So take a moment to let your students know that you value their curiosity and the messy, imperfect process of learning more than a perfect score every time.
—
Works Referenced
- Bjork, E. L., & Bjork, R. A. (2014). Making things hard on yourself, but in a good way: Creating desirable difficulties to enhance learning. In M. A. Gernsbacher & J. Pomerantz (Eds.), Psychology and the real world: Essays illustrating fundamental contributions to society (2nd ed., pp. 59-68). Worth Publishers.
- Eyler, J. R. (2018). How humans learn: The science and stories behind effective college teaching. West Virginia University Press.
- Kapur, M. (2016). Examining Productive Failure, Productive Success, Unproductive Failure, and Unproductive Success in Learning. Educational Psychologist, 51(2), 289-299.
- Schmidt, R. A., & Bjork, R. A. (1992). New Conceptualizations of Practice: Common Principles in Three Paradigms Suggest New Concepts for Training. Psychological Science, 3(4), 207-217.
- Schulz, Kathryn. (2010). Being Wrong: Adventures in the Margin of Error. Ecco.
- Watkins, M. (2024, April 28). We Need to Reclaim Slowness. Rhetorica.
Written with input by Gemini, Google, 19 Jan. 2026, gemini.google.com.
Dana Dawson serves as Associate Director of Teaching and Learning at Temple University’s Center for the Advancement of Teaching.