Privileging the Already Privileged

The Progressive Case for Explicit Instructional Design

In the late 1990s, web designers fell in love with mystery. Navigation menus shed their labels in favour of cryptic icons; buttons became abstract shapes that revealed their purpose only when clicked. The user experience community came to call this “mystery meat navigation,” named after the unidentifiable protein served in school cafeterias: you don’t know what it is until you bite into it, and by then it’s too late.

Credit: Gigilabs

Such interfaces were sharply criticised at the time. Vincent Flanders, writing on Web Pages That Suck, described the phenomenon as “mystery meat navigation”: designs in which users had to click or hover simply to discover basic functionality. The approach was seductive because it looked clean and won design awards, but it transferred cognitive burden from designer to user. Over the following years, usability research vindicated Flanders' critique, and the practice largely faded from professional web design, replaced by explicit labels and descriptive text. What seems elegantly minimal to the expert is often baffling to the novice.

In education, a similar conflation of experts and novices has been far more persistent, and far more consequential. For decades, the dominant view in Western educational thought has been that children learn best through discovery, exploration, and the construction of their own understanding. In many systems, minimal guidance has moved from a pedagogical option to a default stance. This orthodoxy is embedded in curriculum frameworks (the OECD's competency frameworks provided a template for national curricula worldwide) and propagated through teacher training programmes where constructivism is taught as settled science, and culturally enshrined in TED talks that valorise creativity while dismissing "traditional" education as a relic of the industrial age.

The Category Error at the Heart of Discovery Learning

Embedded in this orthodoxy is a category error: the conflation of an outcome with a process. We all want students to be discovering knowledge for themselves as independent learners but in instructional design terms, independent learning is a bad way to become an independent learner. The capacity for autonomous inquiry is not a starting condition; it is an achievement, built through the systematic acquisition of knowledge and skills that make independence possible. To demand independence before providing its prerequisites is to mistake the destination for the route.

Kirschner, Sweller, and Clark put it plainly: problem-solving search is an inefficient way of altering long-term memory because its function is to find a problem solution, not alter long-term memory. Learners can engage in problem-solving activities for extended periods and learn almost nothing. The worked example effect, one of the most robust findings in educational psychology, demonstrates this: students who study worked examples consistently outperform those who solve equivalent problems themselves. Not because problem-solving is unimportant, but because for novices, the cognitive demands of problem-solving actively interfere with learning.

Paul Kirschner also makes an excellent point in this article about the broader paradigm of Constructivism is an epistemological claim about how knowledge is formed and justified, not a pedagogical prescription for how novices should be taught. The error, Kirschner argues, lies in assuming that because knowledge is constructed, it must therefore be learned through unguided construction. This is again, a category mistake. How experts generate knowledge is not how novices acquire it. Constructivism constructs nothing for those who arrive without materials.

---

Discovery Learning and the Inheritance of Advantage

Minimally guided approaches like Discovery learning rests on an often unexamined assumption: that all children come equally equipped to discover. They do not. Discovery learning asks children to find what some of them already have. It’s a mode of instruction that overwhelmingly favours those who already know what to look for.

When we ask students to figure things out for themselves, to construct understanding through exploration, we’re not providing a level playing field. We’re hosting a race where some children start near the finish line while others begin several miles back, blindfolded. Consider what a child needs to successfully “discover” a principle in science, or intuit a pattern in mathematics, or develop sophisticated comprehension strategies in reading: Relevant prior knowledge to connect new information to existing schemas, a vocabulary rich enough to encode and manipulate concepts, Bbackground knowledge that makes problems meaningful rather than arbitrary, metacognitive strategies for monitoring their own understanding and the luxury of cognitive resources not consumed by decoding basics.

Where do children acquire these prerequisites? Largely at home, long before they cross the school threshold. Consider the child of professionals who grows up marinated in language. Dinner table conversations expose her to argumentation, abstraction, and academic register. Her parents scaffold her thinking without realising they’re doing it; asking questions that push her reasoning, providing explanations calibrated to her understanding, modelling curiosity and intellectual engagement.

She’s visited museums, watched documentaries, had books read to her nightly. By the time she encounters photosynthesis in a discovery-based science lesson, she already has mental hooks to hang it on: half-remembered nature programmes, a conversation about why leaves change colour, the time she asked why plants need sunlight and her father actually explained it. For her, discovery learning is a pleasant confirmation of things she half-knows already. The teacher’s carefully designed exploration helps her formalise intuitions she’s been building for years.

Now consider her classmate. His parents work multiple jobs; there’s little time for elaborated explanations of how the world works. English is a second language. Books are scarce. The television plays, but nobody discusses what’s on it. He arrives at school with a fraction of the vocabulary, a fraction of the background knowledge, a fraction of the metacognitive toolkit.

For him, the same discovery lesson is an exercise in bewilderment. He lacks the prior knowledge to generate productive hypotheses. He doesn’t know what he doesn’t know. He can’t discover what he has no foundation to discover. Minimal guidance therefore does not fail equally. It sorts.

In classrooms, this sorting is often misread as engagement or ability. Some students thrive. They contribute readily. They appear curious and capable. Others withdraw, become dependent on peers, or quietly give up. The usual explanation is motivation or confidence. Rarely is the design of curriculum, instruction and assessment itself questioned.

---

The Luxury Belief of Minimally Guided Instruction

As Rebecca Birch argues, exploration and self-discovery are often treated as inherently virtuous. These approaches are no longer just instructional strategies; they are treated as moral goods.

“What is natural is considered good; student failure has no real consequence to the teacher, who is publicly affirmed for their beliefs about the inherent capability of the child to learn by osmosis.”

And when students fail, the failure is rarely read as instructional. It is absorbed into a broader narrative about how “the system” is unjust, the curriculum outdated, or the assessment regime misaligned. The design of instruction itself remains largely untouched. Ironically, all of this is framed in the language of equity and social justice. Minimal guidance is presented as respectful, inclusive, and empowering. Whereas, to intervene too strongly is cast as deficit thinking; to explain too clearly is framed as limiting autonomy.

Yet this moral framing obscures a basic asymmetry. Learners who arrive with background knowledge, language, and cultural capital can navigate exploratory environments with relative ease. Those who do not are left to struggle in the name of fairness. Equality of experience is mistaken for equity of outcome, and the very students these approaches claim to serve are quietly disadvantaged by them.

From an equity perspective, I’ve always found Zig Engelmann’s position far more powerful (and demanding of us as educators); “If the student hasn’t learned, the teacher hasn’t taught—that’s not a slogan, it’s an operating principle.” It’s demanding precisely because it denies education systems the comfort of developmental or moral excuses. Learning failure becomes an instructional design problem, not a child problem. Responsibility is not softened by good intentions or progressive language; it is sharpened by outcomes.

---

What Discovery Learning Really Discovers

There is a peculiar cruelty embedded in much contemporary educational thinking. It masquerades as liberation, as respect for the child, as faith in natural curiosity. But beneath its progressive veneer lies a mechanism that systematically advantages the already advantaged while abandoning those who need explicit guidance the most.

When we withhold clear instruction from novice learners in favour of discovery, exploration, or “authentic” problem solving, we create a learning environment that only works for students who already possess the prerequisite knowledge, the cultural capital, and the cognitive resources to navigate it. For everyone else, minimal guidance is not empowerment; it is abandonment dressed in the language of care.

The persistence of this idea in the face of contradicting evidence is a phenomenon worth pausing over. It suggests that something other than empirical findings is driving the attachment. Douglas Carnine, writing over two decades ago, argued that education remains an immature profession precisely because it privileges ideology over evidence. In medicine, he noted, the Thalidomide disaster led to requirements that drugs be proven effective and safe before prescription. In education, by contrast, we routinely implement approaches based on intuition, ideology, or romantic appeal, testing them only after widespread adoption, if at all.

---

Instructional Design and the Science of Learning as a lever of Equity

The frustrating truth is that we have a good sense of how to do this. Decades of research in cognitive science and instructional design have given us a robust understanding of how novices learn: manage cognitive load, sequence content from simple to complex, provide worked examples before independent practice, build automaticity in foundational skills before demanding their application, ensure high success rates during initial instruction, and space retrieval over time. These are not controversial findings; they are among the most replicated results in educational psychology. The problem is not that we lack the science. The problem is that the science is unfashionable. It doesn’t work well as a TED talk.

This moral framing does not arise accidentally. It is sustained institutionally. Much of the resistance to explicit, evidence-informed instruction does not come from classroom teachers grappling with real constraints, but from education academics and policy actors who shape what counts as legitimate pedagogy. As Professor Pamela Snow observes:

“Much of the opposition to explicit teaching of knowledge comes from education academics (rather than teachers themselves) and is veiled in the language of academic freedom, teacher autonomy and a vague need to ‘re-imagine’ schools and schooling. Unfortunately, this is often a fig-leaf for ‘we don’t like the evidence on the impact of explicit teaching on student academic and wellbeing outcomes’.”

When we design instruction that works only for those who already possess substantial prior knowledge, we systematically disadvantage those who do not. And who lacks prior knowledge? Overwhelmingly, it is students from disadvantaged backgrounds, those whose homes contain fewer books, whose parents lack university degrees, whose early environments provided less of the vocabulary, the concepts, the background knowledge that schools assume rather than teach.

The research is compelling on this point. Clark’s review of aptitude–treatment interaction studies showed that minimally guided instruction systematically disadvantages lower-aptitude and lower-prior-knowledge learners, sometimes producing negative learning effects. Higher-aptitude learners, by contrast, are better able to cope with unstructured environments, masking the instructional failure for those who need guidance most.

The science of learning is not mysterious; it is simply difficult to implement at scale with teachers facing thirty children with vastly different starting points. Into this space comes adaptive environments powered by AI and it’s clear now that LLMS have drastically changed the calculus in terms of how it can augment instructional design, but only if we design these systems according to what cognitive science tells us about how novices learn, how knowledge is structured and the kinds of learning episodes that lead to a durable change in long-term memory, rather than encoding the same constructivist assumptions that have conferred privilege on those who already have it. The risk is that we build digital discovery environments that replicate the inequities of their analogue predecessors, now with adaptive algorithms. The opportunity is that we finally build something better.

Any design of learning that assumes knowledge it doesn’t provide is not progressive. It’s a mechanism for smuggling privilege through the school gates under cover of child-centred rhetoric. Discovery learning asks children to find what some of them already have, and overwhelmingly favours those who already know what to look for.

The advantaged child discovers her way to understanding; an understanding her home has prepared her to reach. The disadvantaged child discovers mainly the limits of what discovery can do for those who lack the prerequisites. The cruel irony is that the kind of education most loudly justified in the name of equity is often the one that most reliably privileges the already privileged.

---

Clark, R. E. (1982). Antagonism between achievement and aptitude treatments. Journal of Educational Psychology, 74(6), 747–754.

Kirschner, P. A. (2025). Epistemology or pedagogy: That is the question. Unpublished manuscript / white paper. (Available via kirschnered.nl)

Kirschner, P. A., Sweller, J., & Clark, R. E. (2006). Why minimal guidance during instruction does not work: An analysis of the failure of constructivist, discovery, problem-based, experiential, and inquiry-based teaching. Educational Psychologist, 41(2), 75–86.

Mayer, R. E. (2004) Should there be a three-strikes rule against pure discovery learning?
American Psychologist, 59(1), 14–19.