An AI tutor,
available right now.

A textbook explains a concept once, in one register, for an average reader who does not exist. If that explanation does not land, re-reading it produces the same result more slowly.

What actually helps is the fourth explanation - the analogy, the worked example, the version that starts from what you already know. That is what a tutor provides, and it is the part that costs $60 an hour and gets rationed accordingly.

Two findings are unusually well established in learning research. Spaced practice beats massed practice, and retrieval - actually testing yourself - beats review by a wide margin.

Both are unpopular because both feel worse. Re-reading feels productive; failing a self-test feels like failing. The one that feels worse is the one that works, which is inconvenient and true.

Using this to understand something is learning. Using it to produce work you submit as your own is a conduct violation at every institution that has a policy, and most now have one.

The distinction is not subtle: ask it to explain, to check, to generate practice. Do not ask it to write the thing being graded. Your institution's policy governs, and it is worth reading rather than assuming.

Learning research is unusually clear on two points, and almost nobody acts on either.

The first is retrieval practice. Testing yourself on material produces substantially better retention than reviewing it, and the effect is large and replicated across decades of studies. Reading a chapter three times feels productive and produces recognition; closing the book and writing down what you remember feels like failure and produces recall.

The second is spacing. The same total study time distributed across several sessions substantially outperforms the same time massed into one, and the advantage grows the longer the retention interval. Cramming works for tomorrow and reliably fails for the exam in three weeks.

Both findings share an inconvenient property: the effective method feels worse while you are doing it. Retrieval is uncomfortable because you discover what you do not know. Spacing is uncomfortable because you forget between sessions and have to work to recover it. That difficulty is not a side effect - it appears to be the mechanism.

Which means the confident feeling of having studied well is a poor guide, and it is why highlighting and re-reading remain the most popular techniques despite performing worst in the research.

The practical version is small: after reading something, close it and write what you remember. Then check. Then do it again in three days.

A concept that has not landed after three readings will not land on the fourth reading of the same text. What changes understanding is a different route to it, and textbooks provide exactly one.

Good explanation tends to work by connecting new material to something already understood. That is why analogies are effective and why the right analogy is intensely personal - it depends on what the learner already knows, which a textbook cannot know and a tutor discovers in conversation.

It also works by changing the level of abstraction. A student stuck on a formula often unsticks when shown a worked example; a student stuck on examples often unsticks when shown why the formula is shaped that way. Moving in the right direction is a diagnosis, and it takes a few questions to make.

The third route is asking the learner to explain it back. This is uncomfortable and diagnostic in equal measure - the point at which an explanation breaks down is precisely the point at which the understanding does, and it is invisible until somebody has to produce it out loud.

That is what a tutor provides and what costs $60 an hour. It is also why unlimited patience matters more than expertise for most stuck points: the material is rarely beyond a student, and the fourth angle is rarely available.

Knowing a subject and being able to demonstrate it under timed conditions are related and not identical, and the second is what gets marked.

Past papers are the most underused resource in exam preparation. They reveal the format, the recurring question types, the mark allocation and the level of detail expected, all of which shape what is worth studying. A topic that has appeared in some form every year is a different priority from one that has appeared twice in a decade.

Mark schemes matter more than most students realise. They show what the examiner is actually rewarding, which is frequently narrower and more mechanical than a well-prepared student assumes. Marks are lost routinely for not showing working, for answering a different question than the one asked, and for writing at length where a mark scheme wanted two specific points.

Timing is a separate skill and needs separate practice. A student who knows the material and has never worked to the clock will run out of time, and the marks lost that way are indistinguishable on the certificate from marks lost to ignorance.

One full paper, timed and closed-book, a week before the exam is worth more than another week of revision. It converts unknown weaknesses into a specific list, and a week is enough to act on it.

The standard explanation for not studying is discipline, and it is mostly wrong. Study behaviour responds far more reliably to structure than to resolve.

The most consistent finding is that specificity beats intention. Deciding to study more produces very little; deciding to study a named topic at a named time in a named place produces substantially more. The mechanism appears to be that the decision has already been made, so the moment of starting does not require a fresh one.

The second is that starting is the hard part rather than continuing. Sessions that begin almost always run longer than planned, and the resistance sits entirely at the threshold. That argues for making the first step trivially small - open the book, do one problem - because the small commitment is the one that gets made.

The third is that progress has to be visible. Studying for an exam six weeks away offers no feedback at all, which is why it loses to anything with an immediate response. Recording what was covered, or tracking a score on practice questions, supplies the signal the task otherwise lacks.

Environment does more than most people credit. A phone in another room is a substantially different working condition from a phone face down on the desk, and the difference shows up in how much gets done.

None of this requires becoming a different person. It requires deciding in advance, starting small, and making progress visible - three structural changes that reliably outperform trying harder.