
Sign up to save your podcasts
Or


This final episode does not introduce new research. It synthesizes what the series has built and returns it to the listener as something they can carry forward. The sources below are therefore a curated retrospective — the most important research threads from across the series, gathered in one place for anyone who wants to go deeper into the ideas that have run through all ten episodes.
Deci, E. L., & Ryan, R. M. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55(1), 68–78.
Albert Bandura — self-efficacy research
The Nervous System — Stress, Capacity, and Co-Regulation
The neurobiological framework for understanding safety, engagement, and shutdown — and why the regulated presence of an adult is not a soft nicety but a literal precondition for learning.
Bruce McEwen — allostatic load
Burnout — The Loop Between Teachers and Students
Maslach, C., & Leiter, M. P. — The Truth About Burnout (1997)
The systemic argument: individual teacher quality matters less than the conditions that allow that quality to develop and be sustained. You cannot build engaged classrooms inside demoralized schools.
Childhood, Play, and the Cost of Overprotection
The antifragility argument: some systems don’t just survive stress — they require it. The psychological immune system, like the biological one, needs exposure to manageable challenge to develop properly.
Technology — Amplifier, Not Villain
Gabor Maté — In the Realm of Hungry Ghosts (2008)
Institutional Design — Control, Trust, and the Compliance Trap
The investment argument: sustained professional development — not mandated curricula — is the highest-leverage intervention available to a school system.
Project-Based Learning and Instructional Design
Buck Institute for Education / PBLWorks
Adolescent Development — The Brain, Identity, and Autonomy
Grace — The Unstated Research
A Final Note
This episode moves from diagnosis to practice. It argues that agency is not a personality trait but a skill — one that must be taught, scaffolded, and developed through genuine experience. The sources below span motivation research, instructional design, project-based learning, and the psychology of choice. Together they make the case that restoring agency does not require abandoning structure — it requires designing structure that serves agency rather than replacing it.
On Agency as a Skill
On Project-Based Learning
On Self-Determination Theory in Instructional Design
On the Psychology of Choice
On Scaffolding and Gradual Release
On Rubrics, Standards, and the Structure of Genuine Choice
On Belief and Teacher Expectation
On Small Agency and Incremental Practice
A Note on Implementation
The Agency Problem is produced independently. No sponsors. No agenda beyond the work.
This episode brings together the research threads from the entire series and applies them to a single, specific dynamic: the feedback loop between teacher burnout and student disengagement. The sources below span burnout research, polyvagal theory, organizational psychology, and the neuroscience of co-regulation. Together they make the case that teacher wellbeing is not a peripheral concern — it is a neurological precondition for student engagement.
Christina Maslach — Maslach Burnout Inventory and burnout research
• Maslach, C., & Leiter, M. P. — The Truth About Burnout (1997)
Stephen Porges — Polyvagal Theory
• Stephen Porges — The Pocket Guide to the Polyvagal Theory (2017)
Allan Schore — research on affect regulation and co-regulation
As noted in Episode 5 program notes, research by Roth, Assor, Kanat-Maymon, and Kaplan has established that teachers who feel autonomy-supported by their administrators are significantly more likely to be autonomy-supportive with their students. The chain runs directly from institutional conditions to teacher experience to student outcomes.
Elaine Hatfield, John Cacioppo & Richard Rapson — Emotional Contagion (1994)
Hatfield and colleagues established the research basis for emotional contagion — the process by which emotions and moods spread between people through unconscious mimicry and feedback. Their work provides the social-psychological parallel to Porges’ neurobiological account of co-regulation. Both point to the same conclusion: the emotional and nervous system state of the adults in a classroom is not background noise. It is active input.
Julian Rotter — locus of control research
This episode makes an argument that individual choices matter while acknowledging they are not sufficient. The research supports both halves of that claim.
Research on teacher resilience — including work by Christopher Day and colleagues — consistently shows that teachers who maintain a sense of professional identity and purpose, who cultivate genuine relationships with students, and who actively manage their own wellbeing are more likely to sustain effective practice over time. Individual agency within the system matters.
Andy Hargreaves & Michael Fullan — Professional Capital (2012)
As referenced in Episode 5 notes, Hargreaves and Fullan’s framework makes the systemic argument explicitly: individual teacher quality matters less than the conditions that allow that quality to develop and be sustained. Schools that invest in collaborative professional culture, teacher autonomy, and genuine professional development produce better outcomes than schools that rely on individual heroism.
The venting cycle described in this episode — shared frustration as a form of connection that simultaneously makes things worse — is consistent with research on what organizational psychologists call negative emotional spirals in workplace groups.
The relationship described in this episode — built across five years, from eighth grade to junior year — is consistent with research on the importance of sustained teacher-student relationships for student engagement and outcomes.
This episode is more autobiographical than the others in this series. The staffroom dynamic, the video game kid, the girl across five years, the recruitment — these are real experiences from a real career.
The Agency Problem is produced independently. No sponsors. No agenda beyond the work.
This episode argues that anxiety in young people is not a sign of weakness or fragility — it is a signal. Specifically, a signal that the gap between what life demands and what a person has been allowed to practice has grown too wide. The sources below span developmental psychology, resilience research, and the emerging science of antifragility. Together they make the case that protection, when it becomes the primary response to difficulty, prevents the very experiences that build capacity.
Haidt and Lukianoff argue that three “great untruths” — that fragility is normal, that emotional reasoning is reliable, and that the world is a battle between good and evil — have taken root in how young people are being raised and educated, with measurable consequences for mental health and resilience. Their concept of “safetyism” — the prioritization of emotional and physical safety above almost all other values — is directly relevant to what this episode describes.
The antifragility framework they draw on comes from:
Taleb’s central argument — that some systems don’t just survive stress but require it to develop properly — is the philosophical foundation for what this episode argues about difficulty as curriculum. The immune system analogy is drawn from his work.
On Free Play, Risk, and Child Development
On Anxiety as a Signal
On Scaffolding and the Zone of Proximal Development
On Helicopter Parenting and Its Consequences
On the Immune System Analogy
On Schools and Institutional Overprotection
A Note on Balance
The Agency Problem is produced independently. No sponsors. No agenda beyond the work.
As always, I would like to thank Aaron Paul for his song “Whispers in the Dark”.
This episode makes a specific argument about technology: that smartphones and social media are amplifiers of existing conditions rather than primary causes of the youth mental health crisis. The sources below represent both sides of a genuinely contested research debate. This episode does not pretend the debate is settled — it tries to follow the evidence honestly to a more nuanced position than either “ban the phones” or “phones are fine.”
Haidt is the most prominent voice arguing that smartphones and social media are the primary drivers of rising anxiety, depression, and mental health challenges in young people. His argument centers on the displacement of unstructured play and in-person social connection by device-based interaction — particularly for girls and the social comparison dynamics of Instagram and similar platforms.
Jean Twenge—iGen: Why Today’s Super-Connected Kids Are Growing Up Less Rebellious, More Tolerant, Less Happy — and Completely Unprepared for Adulthood (2017)
Twenge’s generational research documents the timing correlation between smartphone adoption and shifts in teen mental health, social behavior, and wellbeing. Like Haidt, she takes the relationship between technology and adolescent mental health seriously. Her longitudinal data is worth understanding even if her causal claims are contested.
The Contested Research — A Note on Intellectual Honesty
Key paper:
Candice Odgers —research on adolescent technology use
On the Amplifier Argument
On COVID, Isolation, and the Return to School
The APA documented the significant mental health impacts of pandemic-era isolation on young people — increases in anxiety, depression, and loneliness, particularly among adolescents. These reports provide the research context for what this episode describes observationally: kids who came back from COVID carrying more than they had left with.
On Polyvagal Theory and Safety as a Precondition for Learning Stephen Porges — Polyvagal Theory
On Gaming, Escape, and What Device Use Is Managing
Maté’s work on addiction — particularly his argument that addiction is always an attempt to solve a problem, to manage pain or fill a void — is useful context for understanding escape gaming. The game is rarely the problem. It is the solution to a problem. Understanding what problem it is solving is the beginning of a useful response.
On Phone Policies in Schools
The research on school phone policies is still developing. A few honest observations:
A Note on the Classroom Data
The Agency Problem is produced independently. No sponsors. No agenda beyond the work.
A detour between episodes 5 and 6. If lost agency explains why a kid checks out in third period, does it explain why a whole country checks out too? I took the classroom framework; autonomy, competence, relatedness, to Norway, took some real pushback online, and I’m bringing it back to the room it started in.
I worked with AI tools in developing this series; as thought partners and editors, not as authors. Everything in here came from two decades in classrooms. The AI helped me find the words.
Episode 6 drops July 28th. Reach out: [email protected]
This episode examines the institutional reflex toward control — and the double bind it creates for teachers and students alike. The sources below span organizational psychology, educational policy research, philosophy, and motivation science. Together they make the case that control-based systems are not just ineffective — they are actively counterproductive to the goals they claim to serve.
Michel Foucault - Discipline and Punish: The Birth of the Prison (1975)
Foucault’s analysis of how institutions shape behavior through surveillance and normalization remains one of the most penetrating frameworks for understanding what happens inside schools, workplaces, and any system where visibility is used as a control mechanism. His concept of the panopticon — a structure in which people behave as if they are always being watched, regardless of whether they actually are — describes something that plays out in classrooms and staffrooms every day.
On Autonomy, Trust, and Organizational Psychology
As established in Episodes 2 and 4, SDT is foundational here. The research consistently shows that controlling environments — characterized by mandates, surveillance, and external pressure — undermine intrinsic motivation and produce compliance at the expense of genuine engagement. This applies to teachers as much as to students.
Daniel Pink — Drive: The Surprising Truth About What Motivates Us (2009)
On Teacher Autonomy and Professional Development
• Linda Darling-Hammond — The Flat World and Education (2010)
Hargreaves and Fullan argue that the solution to educational underperformance is not more control but more professional capital — the combination of human capital (individual teacher skill), social capital (collaborative professional relationships), and decisional capital (the professional judgment built through experience). Their framework is a direct rebuttal to the logic of canned curricula and scripted lessons.
On Standardization, Equity, and the Guaranteed Curriculum
W. James Popham — The Truth About Testing (2001)
On the Loss of Play, Recess, and Unstructured Time
On Burnout and the Conditions That Produce It
Maslach’s identification of the six organizational risk factors for burnout — workload, control, reward, community, fairness, and values — maps almost perfectly onto the conditions this episode describes. The combination of high demand, low control, and minimal support is not just demoralizing. It is a documented pathway to burnout.
• Maslach, C., & Leiter, M. P. — The Truth About Burnout (1997)
On the Foucault Reference — A Note
The performance that results is not engagement. It is the appearance of engagement. Which is exactly the compliance problem Episode 4 named — and which control-based systems consistently reproduce rather than solve.
The Parallel — A Note for Educators
Music by Aaron Paul - Whispers in the Dark
The Agency Problem is produced independently. No sponsors. No agenda beyond the work.
On Intrinsic Motivation and the Limits of External Measurement Edward Deci & Richard Ryan — Self-Determination Theory As established in Episode 2, Deci and Ryan’s research is foundational here. Their distinction between controlled motivation (doing something because of external pressure) and autonomous motivation (doing something because it aligns with genuine interest or values) maps directly onto the compliance vs. engagement distinction this episode makes. The critical finding: systems that reward compliance actively undermine the intrinsic motivation required for genuine engagement. You cannot mandate engagement. You can only create conditions that make it possible. • Deci, E. L., & Ryan, R. M. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55(1), 68–78. • Edward Deci — Why We Do What We Do (1995) On Rewards, Grades, and the Undermining of Learning Alfie Kohn — Punished by Rewards (1993) and The Schools Our Children Deserve (1999) Kohn’s work is the most direct popular treatment of how grading and reward systems produce compliance at the expense of genuine learning. His argument — that grades shift students’ focus from learning to grade management — is directly relevant to the student described in this episode’s cold open. Key insight: when the metric becomes the goal, the metric stops measuring what it was designed to measure. This is sometimes called Goodhart’s Law, and it applies to education as much as it applies to economics. Alfie Kohn — Punished by Rewards remains the most accessible entry point. For a more education-specific treatment: The Schools Our Children Deserve (1999). On Assessment and What Gets Measured Dylan Wiliam — research on formative assessment Wiliam is one of the most rigorous researchers working on educational assessment. His work distinguishes between assessment that informs learning (formative) and assessment that merely records compliance (summative, when misapplied). His research consistently shows that what gets assessed shapes what students actually do — and that most assessment systems are measuring the wrong things. • Dylan Wiliam — Embedded Formative Assessment (2011) — practical and research-grounded Goodhart’s Law Originally an economic principle — “when a measure becomes a target, it ceases to be a good measure” — Goodhart’s Law applies with uncomfortable precision to educational metrics. Grade inflation, assignment completion rates, and standardized test scores all illustrate the same dynamic: once the measurement is the goal, people optimize for the measurement rather than the thing the measurement was supposed to represent. On Engagement as a Distinct Construct Mihaly Csikszentmihalyi — Flow: The Psychology of Optimal Experience (1990) Flow — the state of complete absorption in a genuinely challenging task — is the far end of the engagement spectrum. Csikszentmihalyi’s research defines the conditions under which genuine engagement becomes possible: clear goals, immediate feedback, challenge matched to skill. None of these conditions are produced by compliance-based systems. Engagement research in education: Researchers including Jennifer Fredricks, Phyllis Blumenfeld, and Alison Paris have developed frameworks distinguishing behavioral engagement (compliance), cognitive engagement (genuine mental effort), and emotional engagement (investment and interest). Their work makes the case that behavioral engagement — the kind schools typically measure — is the least predictive of actual learning outcomes. • Fredricks, J. A., Blumenfeld, P. C., & Paris, A. H. (2004). School engagement: Potential of the concept, state of the evidence. Review of Educational Research, 74(1), 59–109. On Discussion-Based Learning and Visible Thinking Project Zero — Harvard Graduate School of Education Project Zero has spent decades researching what genuine thinking looks like in classrooms and how to make it visible. Their “Thinking Routines” framework is built on exactly the distinction this episode makes: the difference between producing an answer and actually thinking. pz.harvard.edu Paulo Freire — Pedagogy of the Oppressed (1968) Freire’s critique of what he called the “banking model” of education — in which students are treated as empty vessels to be filled with information — is one of the most enduring arguments against compliance-based schooling. Students are not passive recipients. When treated as such, they become passive performers. On Teacher Burnout and the Performance Dynamic Christina Maslach — burnout research The relational cost of compliance described in this episode — the loneliness of performing teaching to an audience performing learning — connects directly to Maslach’s dimension of depersonalization in burnout. When genuine human contact is replaced by performance on both sides, something essential drains out of the work. • Maslach, C., & Leiter, M. P. — The Truth About Burnout (1997) On the Long-Term Cost of Compliance Training The argument that compliance training teaches students a “lesson with a very long shelf life” — that appearance of competence is as good as competence — connects to research on academic dishonesty, credential inflation, and the gap between educational attainment and actual skill development. Bryan Caplan — The Case Against Education (2018) Caplan’s controversial but data-rich argument is that much of what education produces is signaling — the appearance of competence — rather than actual human capital development. Whether you agree with his conclusions or not, his documentation of the gap between credentials and skills is directly relevant to what this episode argues. A Note on the Noise Question This episode deliberately complicates the argument by acknowledging that noise is not synonymous with engagement, and quiet is not synonymous with compliance. The research on this is consistent: what predicts learning is not the surface environment but the internal cognitive and emotional state of the learner. For a practical framework on designing for genuine engagement rather than compliance, the Self-Determination Theory applications in education literature — particularly the work of Johnmarshall Reeve referenced in Episode 2 notes — remains the most research-grounded starting point. The Agency Problem is produced independently. No sponsors. No agenda beyond the work. I worked with AI tools in developing this series — as thought partners and editors, not as authors. Everything in here came from two decades in classrooms. The AI helped me find the words. Questions, thoughts, or responses to this episode: [email protected]
music is The Other Side by Aaron Paul
From the publisher's feed