Defending the Science of Learning
Carl Hendrick: "Effective Instruction isn't a matter of chance, but of design"
Last week, a curious article set tongues wagging among those of us who argue for evidence-based instruction. The piece in EducationHQ labeled the science of learning movement as "soul destroying," a "dogma," and a “neoliberal monopoly" imposed by governments and amplified by social media. This misguided critique feels like doctors wringing their hands over medical advances that extend patients’ lives—a puzzling rejection of evidence that demonstrably improves outcomes. That’s why I’m excited to republish, with his permission, Carl Hendrick’s incisive defense of the science of learning, originally posted on his indispensable Substack, The Learning Dispatch. Carl is a professor at Academica University of Applied Sciences in Amsterdam and the author of "How Learning Happens" with Paul Kirschner. I’ve connected with him at multiple ResearchEd conferences, where his clarity and dedication to evidence-based teaching shine.
From its inception earlier this year, the goal of “The Next 30 Years” has been to redirect the focus of American education reform efforts, which have over-relied on policy levers. The real catalyst for student success lies in the classroom, where the science of learning can transform practice. A complimentary goal has been to introduce readers to teachers, researchers, and instructional leaders like Carl whose work deserves a broader platform to shape education’s future. His piece dismantles myths and champions cognitive science—a must-read for educators, policymakers, and anyone else invested in improve student outcomes.
– Robert Pondiscio
Effective Instruction isn’t a matter of chance, but of design.
Jun 08, 2025
I don’t particularly like the term science of learning. It implies a single cohesive entity with an agreed-upon, codified set of laws or prescriptions, when in reality it’s a loose and evolving body of interdisciplinary research, drawing from cognitive psychology, neuroscience, education, and design. It offers probabilistic insights, not fixed rules.
However, those different disciplines often converge on similar instructional implications such as the importance of managing cognitive load, using retrieval practice to enhance memory, providing worked examples to reduce unnecessary problem-solving strain, spacing learning over time, and ensuring that novices receive explicit, guided instruction before being asked to solve problems independently.
I was five years into my career as a teacher working in an inner-city London school before I learned any of this, and it profoundly changed my practice. I later discovered that many teachers are in the same boat - they were never taught how learning happens, and like me, want to teach their students better, by using approaches grounded in evidence rather than fads, intuition or well-meaning but unproven methods.
However, despite the substantial empirical support underpinning the core elements of learning, there remains a persistent undercurrent of scepticism among some educators. This scepticism often manifests in critiques that systematically overlook the considerable evidence base supporting these methodologies.
Dr Rachael Jefferson's recent piece in EducationHQ exemplifies this troubling trend. She characterises the growing influence of explicit instruction and cognitive science in schools as "soul destroying," a "dogma," and a “neoliberal monopoly" imposed by governments and amplified by social media. A diversity of views is always welcome, but unfortunately it has to be said that nearly everything in the piece is either a misrepresentation or a confusion of terms, relying on rhetorical tropes and well-worn clichés that mislead more than they clarify.
I’d like to address some of the central claims which are representative of a broader critique of the science of learning.
1. “The science of learning is a dogma.”
This could not be further from the truth. It is a tentative, empirical discipline. It does not prescribe one way to teach, but identifies which methods tend to work better under particular conditions. Direct instruction, for instance, has decades of converging evidence supporting its effectiveness, especially for novice learners and foundational knowledge acquisition. This is not ideology; it’s conditional probability.
It doesn’t claim to have all the answers. It doesn’t prescribe a single method. Rather, it offers probabilistic insights: certain strategies are more likely to be effective under certain conditions, particularly when teaching novices or when long-term retention is a goal.
A dogma cannot be questioned whereas cognitive science is constantly being questioned, revised and debated. Its strength lies not in certainty, but in its openness to disconfirmation, replication, and refinement. That’s not dogma, that’s the scientific method.
2. The Empirical Foundation: More Than "A Handful of Studies"
The idea that the science of learning lacks sufficient evidence is just baffling. Cognitive load theory alone has over thirty years of robust research behind it. The evidence for approaches such as Direct Instruction, retrieval practice, and worked examples is stronger than for nearly any other instructional approach.
This is not a fringe or contested body of work, it is convergent, replicated, and applicable, drawn from decades of rigorous research across cognitive psychology, neuroscience, and instructional design. To dismiss that as a 'handful of studies' is to ignore decades of converging evidence.
I understand that while we have strong evidence about the constituent elements of learning such as memory, encoding, retrieval, and consolidation, we have less about how to apply that evidence precisely in instructional design. Still, the answer is not less evidence but more. Without that evidence, we fall back on intuition, ideology, or tradition, none of which are reliable guides for creating equitable outcomes for all children.
3. “This movement de-skills teachers and destroys autonomy.”
This is a category error. Autonomy is not the same as idiosyncrasy. Professionals in other high-stakes domains such as medicine, aviation, engineering, routinely follow evidence-based protocols, not because they are automatons, but because they are accountable to what works. They still exercise judgement, but that judgement is informed, not improvised.
The oft repeated call for "teacher autonomy" confuses licence with liberty. Autonomy in teaching cannot mean doing whatever one pleases irrespective of outcomes. Professionals are accountable to something larger than themselves. In medicine, law, or aviation, practice is bounded by evidence. Why should education be exempt?
The science of learning does not "mandate" teacher behaviour, it equips teachers with research-informed strategies. Crucially, it leaves space for adaptation. (In fact, adaptive and responsive teaching are core elements of the science of learning movement.) There's no one-size-fits-all approach, and no serious scholar in the field argues otherwise. What it does reject is the idea that teaching should be built on vibes, ideology, or whatever "feels right" to individual practitioners.
If anything, the science of learning raises the bar for professional practice. It demands that we critically examine what we do and why we do it, not in opposition to professionalism, but in service of it.
4. “It’s all memorisation and ignores the body.”
The false dichotomy between memorisation and creativity is a tired canard. No one is arguing for parroted facts over deep thought. But one cannot think critically about what one does not know. The cognitive science literature is clear: knowledge in long-term memory is what makes higher-order thinking possible.
The science of learning does not reduce education to rote recall. It highlights the indispensable role of long-term memory in learning, because, as Dan Willingham puts it, “thinking well requires knowing facts.” You cannot reason about what you don’t remember. You cannot connect the dots if you don’t have any dots.
And while embodied cognition might be a useful lens in certain contexts like physical education or dance, it has not demonstrated a generalisable theory of instruction. Claiming that cognitive science “ignores the body” is like accusing oncology of ignoring yoga. It’s not a rejection; it’s just not the central mechanism under investigation.
5. “make it more student-centred and inclusive”
It just feels like a government agenda, just something is ‘off’ with it for me as a person who celebrates diversity within pedagogy in order to make it more student-centred and inclusive, which is where I always come from.
Her complaint about “student-centred” practices being marginalised betrays a deep irony. The science of learning is student-centred, and just not in the sentimental sense. It attends to the learner’s cognitive architecture, not the teacher’s preferred aesthetic.
It’s also notable that many academics never actually define what they mean by “inclusive” or “diverse” pedagogy. Like many educational buzzwords, these terms generate warmth but resist scrutiny. Inclusivity should mean giving all students access to powerful knowledge, not validating underperformance as a form of empowerment.
6. The ‘context’ Canard
The argument that the science of learning ignores “context” is another red herring. Of course context matters, in fact the best elements of the field is examining just how certain approaches work in different contexts. But general principles, like reducing cognitive load or retrieval as a form of learning, apply across so many contexts, just as principles of nutrition apply whether one is dining in Tokyo or Toulouse. Context shapes application, not validity.
The claim that explicit instruction cannot work in the arts or Physical Education is deeply misguided. Structured modelling, guided practice, and deliberate feedback are foundational in almost all disciplines. Try learning to play the violin without scales, to shoot a basketball without technique drills, or to act without a script and cues. These are all forms of explicit scaffolding, structure, and intentional skill development.
Mastery in the arts and physical education relies not on unstructured discovery, but on clear models, repeated rehearsal, and precise feedback. Creativity is about constraints, not chaos. It flourishes within structure, not in the absence of it.
Explicit or structured teaching is defined as “technocratic” and “heavy-handed,” as though clarity and sequence were instruments of oppression. But one might ask: is it really authoritarian to give students a coherent explanation of photosynthesis? Or is it more authoritarian to insist on mystifying them in the name of vague ‘exploration’?
7. This thing I don’t like is bad because neoliberalism
Like many in education academia, Jefferson invokes the dark spectre of "neoliberalism" (the progressive educator’s all-purpose exorcism) to ward off the dark spirits of explicit instruction. But ideology is not a rebuttal. The question remains: what actually helps children learn? This is largely an empirical question. To answer that, one must turn not to political slogans, but to data, and therein lies the problem.
Framing explicit instruction as "neoliberal dogma" transforms an evidence-based discussion into ideological theatre, precisely the kind of politicisation that undermines rational discourse about effective practice. Imagine invoking Foucault to find out what kind of cancer treatment will save lives? Or rejecting antibiotics because they represent "Western medical imperialism." This is not a serious discussion. Evidence doesn't become invalid simply because it lacks a revolutionary aesthetic.
8. The Equity Imperative
The ultimate tragedy of positions which privilege ideological purity over empirical evidence is that they potentially perpetuate educational inequity. Students from disadvantaged backgrounds, who often lack the cultural capital to navigate discovery-based approaches, benefit most dramatically from explicit, systematic instruction. To reject this evidence in favour of pedagogical romanticism is to abandon those who need us most. As Doug Carnine noted, “The greatest victims of these fads are the very students who are most at risk.”
For example, between 2013 and 2019, Mississippi, once near the bottom of national rankings, saw dramatic gains in reading and math scores, outpacing nearly every other U.S. state despite high poverty rates. These improvements were not the result of vague pedagogical pluralism but a deliberate, statewide adoption of evidence-based reading instruction grounded in cognitive science: systematic phonics, phonemic awareness, and explicit teacher training.
Far from being “oppressive” or a punitive policy, the state's Literacy-Based Promotion Act emphasised coaching, formative assessment, and professional development. The result: students gained nearly a full year’s worth of reading growth, with the strongest gains among disadvantaged children. Crucially, only a small portion of this success was due to retention policies, most gains were instructional.
A Call For Evidence Over Vibes
Jefferson’s complaint that the science of learning feels "off" is perhaps the most honest moment in the piece. Feelings are not arguments. There's nothing wrong with instinct, but when faced with large-scale evidence, one should be prepared to adjust. If the evidence feels "off," it may be because it contradicts one's cherished assumptions.
The science of learning is not a political conspiracy. It is a response to decades of fads, failed reforms, and underwhelming outcomes. It does not offer panaceas, but it does offer a better foundation than vibes, metaphors, and social media polemics.
The science of learning empowers teachers to make better decisions in their classrooms. Decisions grounded in how students actually learn, not in how we might wish they did. It replaces intuition with insight, and replaces well-intentioned guesswork with clearer guidance.
What's truly "soul destroying" is not explicit instruction, it's the idea that we should abandon hard-won insights from research in favour of sentiment, ideology, or the latest educational fad. Education policy cannot be built on vibes alone.
If one wishes to challenge the science of learning, let them meet it on its own terms: with data, with studies, and with better outcomes. That would be a service to teachers, to students, and to the future of our profession.
Excellent article. Thanks.
So glad you are bringing this piece to more eyes. It is so spot-on!