This isn't an attack on your practice. It's five short essays that name things experienced teachers have already observed — and explain the mechanism behind each one. No jargon. No citations in the text. Just the clearest description I can give of how learning actually works and why conventional instruction often doesn't.
Written by Gregory Stuart Lacefield — who built a pedagogical framework independently from seven years of direct classroom observation, without access to the academic literature, and later found that the research had been saying the same things for decades.
When a student fails at algebra, the natural assumption is that they have an algebra problem. So you reteach the algebra. You try a different explanation, a different approach, more practice problems. Sometimes it works. Often it doesn't — and you can't explain why, because the student seems to understand when you walk them through it but falls apart the moment they try it alone.
Here is what is usually actually happening: the student doesn't have an algebra problem. They have a problem at a concept that algebra depends on — something foundational that was never built correctly, or was built incorrectly, years earlier. The algebra failure is the symptom. The cause is somewhere upstream, invisible to a teacher focused on the current curriculum.
In seven years of teaching GED mathematics to adults who had failed in traditional classrooms, I saw this pattern more times than I can count. A student who couldn't solve a linear equation wasn't confused about algebra. They were confused about what an equals sign means. Not as an abstract philosophical question — concretely, operationally, in the moment of solving: they read the equals sign as "the answer goes here" rather than as a statement that two expressions represent the same quantity. Every algebraic operation they attempted was built on that broken foundation. No amount of algebra instruction was going to fix an equals-sign problem.
The failure shows up at the level you're teaching. The cause lives at the level below it — sometimes several levels below.
This is what I came to call the schema floor — the highest level at which a student's understanding is genuinely sound enough to build on. Everything above it is unstable. It looks like understanding, because the student has learned to approximate the right behaviors. But it collapses under novel conditions, under pressure, under any problem that doesn't look exactly like the examples they practiced.
The practical implication for teaching is uncomfortable: the student in front of you who is failing the current material needs instruction that starts below the current material. Not remediation in the pejorative sense — systematic diagnosis of where the understanding is actually sound and where it isn't, followed by building from there. The current curriculum is irrelevant until the floor is found.
Most teachers don't have time for this. The curriculum moves and students move with it whether or not they're ready. That structural constraint is real. But knowing it's happening — knowing that the failure you're observing has a specific upstream cause you haven't identified — changes how you look at a struggling student. They're not failing because they can't. They're failing at a point that isn't the point they appear to be failing at.
Every experienced teacher has had this student: intelligent, capable in other areas, completely shut down in mathematics. They say they're not a math person. They say it with conviction, as if it's a biological fact — like height, or eye color. Something fixed. Something they were born without.
I spent seven years watching what happened when those students were put in a room where that story wasn't accepted. Not argued with — just not accepted. Where the assumption was that the capability was there and the question was only what had gotten in the way of it.
What I found, consistently, was that the "I can't do math" students weren't cognitively limited. They were students who had, at some identifiable point, stopped engaging in ways that would produce improvement. Usually the stopping happened because something early went badly — a concept never explained clearly, a test failed at a vulnerable moment, a teacher who moved too fast or marked correct reasoning wrong. The student drew a conclusion from that experience: I am not capable. And once that conclusion was drawn, every subsequent struggle confirmed it rather than challenging it.
Students don't fail because they lack intelligence. They fail because they stop engaging before momentum develops — and the stopping looks, from the outside, exactly like inability.
The distinction matters enormously for instruction. A student who genuinely cannot do something needs different content. A student who stopped engaging needs a different experience — specifically, they need to accumulate evidence that contradicts the story they've been telling themselves. That means starting below where they are, at a level where they can succeed, and building up slowly enough that each step is achievable. Not to protect them from difficulty. To give them a foundation of real evidence that their effort produces results before the difficulty increases.
The research on this is extensive. Benjamin Bloom's 2-sigma finding — that one-on-one tutoring produces results two standard deviations better than conventional classroom instruction — is largely explained by this mechanism. The tutor knows when the student has stopped engaging and can pull them back. The classroom teacher managing thirty students cannot.
What this means practically: when a student tells you they're not a math person, the appropriate response is not argument and it's not consolation. It's to find the level where they can actually succeed and start there. The story changes when the evidence changes. Not before.
There is a common scenario in mathematics classrooms: a student produces an answer that is wrong, or presented in an unusual way, and the teacher marks it wrong without examining the reasoning behind it. The student had the right idea but expressed it imprecisely. Or arrived at the right answer through an unconventional route. Or solved a related problem correctly and made a small notational error at the end.
These are not the same as getting the wrong answer because you don't understand the concept. But they look the same on paper. And if they're treated the same — marked wrong, explained, moved on — something specific and damaging happens.
The student who understood but expressed imprecisely doesn't receive a correction. They receive a contradiction. They were doing the right thing. Their reasoning was sound. And then they were told they were wrong. The cognitive effect of this is not the same as being told you're wrong when you are wrong. It is more destabilizing — it creates doubt not about the specific answer but about the reliability of their own reasoning. If following the logic correctly produced a wrong answer, why follow the logic?
A wrong answer can be fixed. A student who has stopped trusting their own reasoning is a much harder problem — and incorrect correction is one of the fastest ways to produce it.
I developed a practice over years of teaching that addresses this directly: before responding to any wrong answer, I ask the student to explain their thinking. This takes thirty seconds. It immediately distinguishes a student who guessed from a student who reasoned incorrectly from a student who reasoned correctly but expressed imprecisely. Those three students need completely different responses. Treating them the same wastes time at best and causes real damage at worst.
The student who reasoned correctly and expressed imprecisely needs to hear that their reasoning was right, followed by instruction on standard notation. Not "that's wrong." The reasoning was not wrong. The notation was. That distinction is not pedantic — it is the difference between a student who leaves the interaction more confident in their thinking and a student who leaves doubting it.
Precision of evaluation matters as much as precision of execution. A teacher who is imprecise about what a student actually understands will produce students who are imprecise about what they know. In a subject built on logical necessity, that imprecision compounds.
When a student lacks confidence in mathematics, the standard response is encouragement. You can do it. Keep trying. I believe in you. This is well-intentioned. It is also almost entirely useless.
Not because encouragement is bad. Because it is trying to produce a cognitive state through a mechanism that doesn't produce that state. Confidence is not a feeling that can be talked into existence. It is produced by a specific kind of experience: genuine difficulty, followed by genuine success. The success has to be real — not easy, not guaranteed, but earned through effort at an appropriate level of challenge.
A student who has that kind of accumulated experience is confident. Not because someone told them they could do it. Because they have evidence that they can. The evidence came from doing it, repeatedly, at levels hard enough that success wasn't guaranteed but achievable enough that success actually occurred.
Confidence is not a personality trait. It is produced by specific conditions. Those conditions can be designed for deliberately. Encouragement is not one of them.
The practical implication is that confidence is an engineering problem, not a motivational one. The teacher's job is not to convince the student they can succeed. It is to construct the conditions under which they actually will succeed — repeatedly, at increasing levels of difficulty, with enough support that the difficulty remains productive rather than destructive.
This means calibration. If a student is getting everything right, the work is too easy and they're not building the kind of confidence that holds up under real challenge. If a student is getting everything wrong, the work is too hard and they're accumulating evidence that effort doesn't produce results — which is the opposite of confidence. The target is somewhere around 80% — hard enough that success feels real, achievable enough that success actually happens most of the time.
The 20% of problems that students get wrong in that framework are not failures. They are the difficulty that makes the 80% meaningful. A student who succeeds at everything learns nothing about their capacity under pressure. A student who struggles with 20% and succeeds with 80% is building something real.
Encouragement is what you say when a student succeeds. It is not what produces the success. The conditions produce the success. Get the conditions right and encouragement becomes unnecessary — the student has their own evidence and doesn't need yours.
There is a distinction that most educational systems don't make explicit, and the failure to make it is responsible for more damage than any particular content gap or resource shortage. The distinction is between studying to perform on a specific test and studying to actually understand something.
These are not the same thing. They require different kinds of practice. They produce different results. And they feel similar from the outside — a student who has memorized procedures can appear indistinguishable from a student who understands them, right up until the moment a novel problem appears. Then the difference becomes immediately, sometimes painfully, obvious.
Performance training works for tests. It should. A student preparing for a specific standardized exam should know the format, recognize the common question types, practice under timed conditions, and eliminate avoidable errors. That is a real skill and it requires real practice. The problem is not that performance training exists. The problem is treating it as if it produces understanding, or treating understanding as if it produces performance without additional practice.
A student can perform temporarily without understanding. A student can understand without yet being able to perform under pressure. Neither state alone is the goal. Both are required, and they require different kinds of work to develop.
When I say "covering material," I mean the educational version of performance without understanding: a teacher moves through the curriculum, students learn to recognize problem types and apply procedures, and assessment measures whether they can reproduce those procedures on familiar problem formats. The material has been covered. The students have not necessarily learned it in any durable sense.
The evidence for this is what happens over time. Students who learned procedures without understanding forget them quickly. Students who understood the underlying concepts retain them. The procedures can be reconstructed from understanding; they cannot be reconstructed from memory alone.
What produces understanding is slower and less efficient than what produces performance: tracing concepts back to their definitions, asking why the procedure works rather than just how to execute it, connecting new concepts to what is already understood rather than introducing them in isolation. It is harder to schedule, harder to assess, and harder to demonstrate to administrators watching a lesson. But it is what produces the kind of knowledge that transfers to new problems and survives the passage of time.
The practical implication for teachers is not that all performance training is wrong — it isn't. It is that the distinction needs to be made deliberately and explicitly. When you are drilling for a test, say so. When you are building understanding, the work looks different and should look different. Conflating them produces students who can neither perform reliably nor understand durably — which is where much of the system currently sits.
Not teachers who agree with every word — teachers who've watched these dynamics in their own classrooms and want to work within a framework that addresses them systematically. The first implementation of the Lacefield Adaptive Learning System runs through human tutors. Early participants co-generate the research data and shape how the platform develops.