Everyone picks up algorithms differently. Some people need to see a case hundreds of times before it sticks; others understand the structure once and never forget it. Some want to drill one thing to perfection; others get bored and learn better when everything is mixed together.
This guide describes seven practice types. None of them is "right" — they describe how practice feels to you and how your memory behaves, and the trainer uses your choice to decide what to show you, in what order, and how often.
A note on the science: the popular idea of fixed "learning styles" (visual vs. auditory vs. hands-on) is not well supported by research. What is well supported is that different practice schedules and presentation formats affect how durable a skill becomes. The types below are grounded in that research. Think of them as practice preferences rather than personality labels — you can change them at any time.
Pick one scheduling type and one presentation type. Each chapter below describes what the type is, how to tell whether it fits you, and what the trainer will do differently if you choose it.
What it is
Spaced repetition is the practice of reviewing a skill at increasing intervals: after a day, then a few days, then a week, then a month. It is built on the forgetting curve, the observation that memories decay rapidly after learning unless they are refreshed. Each successful review at the right moment flattens the curve and makes the memory last longer.
The important insight for cubing is that acing an algorithm today does not mean you own it. A skill that has been learned once is still fragile. Continuing to practice past the point where you first get it right — called overlearning — is what turns a fragile memory into a durable one.
This is probably you if…
- You have learned an algorithm perfectly, stopped practicing it for a week, and found it gone.
- You need many more repetitions than other people seem to, and that is simply how your memory works.
- Long single sessions on one case do not stick as well as short sessions spread across days.
- You often recognize a case but cannot start the algorithm until you have seen it once more.
- You have ADHD or another condition that affects working memory and retention. Many people in this situation find spacing is the only schedule that actually holds.
How the trainer adapts
- Every algorithm gets its own review date. A case you solved correctly moves further into the future; a case you missed comes back soon.
- The gap grows with each success (for example 1 day → 3 days → 1 week → 2 weeks → 1 month) and shrinks after a miss.
- Algorithms you "already know" never fully leave the queue. They come back at long intervals so they stay sharp.
- Sessions are short and frequent by design. A daily 10-minute review beats one long weekly session for this type.
Tips
- Do not skip reviews of cases you feel confident about. The review is cheap, and it is exactly what keeps the skill from fading.
- If a case keeps coming back too often, that is useful information: it is one you have not truly consolidated yet.
- Consistency matters more than intensity. A missed week undoes more than a short session gains.
What it is
Blocked drilling means working on a single algorithm, over and over, until it becomes automatic — and only then moving to the next one. Every repetition is the same case, so you can focus entirely on execution: finger tricks, speed, and smoothness.
This is the way most people instinctively practice, and it is very effective for building fast initial fluency. Its weakness is that it trains execution much more than recognition, and skills learned this way can decay faster than skills learned with spacing or interleaving. Many cubers use blocked drilling for new algorithms and switch to one of the other two schedules to keep them.
This is probably you if…
- You like the feeling of a long streak on one case, and switching cases mid-session breaks your focus.
- You want to get an algorithm fully into your hands before you worry about recognizing it in a solve.
- Mixed practice feels frustrating or chaotic rather than helpful.
- You prefer fewer, longer sessions over many short ones.
How the trainer adapts
- The trainer picks one case and stays on it until you hit a streak threshold (for example 10 correct in a row under a target time).
- Only after the threshold is met does it introduce the next case.
- Previously learned cases are revisited in their own blocks rather than mixed into the current one.
- Progress is shown per case, so you can see exactly how close you are to "done" with the current block.
Tips
- Because blocked skills can fade, consider adding a short mixed warm-up of old cases at the start of each session.
- Set a realistic streak threshold. Too low and the skill is not consolidated; too high and sessions become tedious.
- Once a full algorithm set is learned, switching to Spaced Repetition or Interleaved Practice will help it stick long-term.
What it is
Interleaving means mixing different cases within a single session so that you never know which one is coming next. Each repetition forces you to first recognize the case, then retrieve the right algorithm, then execute it — the exact sequence you perform in a real solve.
Interleaving feels harder than blocked drilling and progress feels slower in the moment, but it tends to produce better long-term retention and much better recognition. The difficulty is the point: the extra effort of retrieval is what strengthens the memory.
This is probably you if…
- You can execute an algorithm perfectly when you know it is coming, but freeze when it appears unexpectedly in a solve.
- Repeating the same case many times in a row bores you and you start executing on autopilot.
- You learn well under a bit of variety and challenge.
- Your goal is faster recognition in real solves, not just faster execution.
How the trainer adapts
- Cases from several algorithm sets are shuffled together in each session.
- The trainer avoids showing the same case twice in a row.
- You see the case first and must identify it before the algorithm is revealed, so recognition is always trained.
- Newly learned cases are mixed in with well-known ones rather than drilled in isolation.
Tips
- Expect a lower hit rate than you would get with blocked drilling. That is normal and not a sign that it is not working.
- If a case is brand new, a few blocked repetitions first can help, after which it should go straight into the mix.
- Interleaving pairs well with Spaced Repetition: spacing decides when a case comes back; interleaving decides what it is mixed with.
What it is
Chunking is the process of grouping individual moves into larger meaningful units. Instead of memorizing R U R' U' R' F R2 U' R' U' R U R' F' as fourteen separate moves, a pattern-based learner sees it as a few familiar building blocks: a sexy move, a setup, a known trigger, and its undo.
Many algorithms share these blocks. Commutators, conjugates and common triggers (such as R U R' U' or F R U R' U' F') appear everywhere. Once you know the blocks, new algorithms become recombinations of things you already know, and related cases become variations of one idea rather than separate things to memorize.
This is probably you if…
- You find a long string of notation meaningless until someone explains why it works.
- You remember algorithms better when you can see how they relate to each other.
- You naturally notice when two algorithms share a sub-sequence.
- You would rather learn 5 ideas than 20 facts.
How the trainer adapts
- Algorithms are displayed broken into named chunks, with each trigger highlighted.
- Related cases are grouped and introduced together, so you see them as a family.
- Where an algorithm is a commutator or conjugate, the trainer labels the parts (setup, commutator, undo).
- A reference of common triggers is one tap away during practice.
Tips
- When you learn a new algorithm, try to identify its chunks yourself before looking at the trainer's breakdown.
- Learning mirror and inverse versions of an algorithm together reinforces the structure.
- This type is about how algorithms are shown. Combine it with whichever scheduling type matches your memory.
What it is
Recognition-first learners build a direct link between the picture of a case — the pattern of colors on the top face and sides — and the solution. The notation is a means to an end; what you are really storing is "this image → these hand movements."
In a real solve, recognition is the first step and often the slowest, so training it directly pays off. Learners of this type often find that once the image is firmly associated with the algorithm, they stop thinking in notation at all.
This is probably you if…
- You remember what a case looks like far more easily than what its algorithm is written as.
- Reading notation feels like a translation step that slows you down.
- You recognize cases in solves before you can consciously name them.
- Diagrams and color patterns stick in your memory; text does not.
How the trainer adapts
- Every practice item leads with the cube diagram. The notation is hidden until you ask for it.
- Cases are shown from the angle you would actually see them during a solve, and with varied color schemes so you learn the pattern rather than specific colors.
- Recognition speed is tracked separately from execution speed.
- Similar-looking cases are deliberately shown near each other so you learn to tell them apart.
Tips
- Train with a fixed color scheme first, then switch to random colors to make sure you are recognizing the shape, not the colors.
- Practice identifying the case from a single glance before picking up the cube.
- If two cases keep getting confused, ask the trainer to show them side by side.
What it is
Kinesthetic learners store algorithms in their hands rather than in words or images. The algorithm is a physical gesture — a particular rhythm of finger tricks and regrips — and the notation is simply a way to describe that gesture to someone else.
For this type, the quality of the first few repetitions matters enormously. An algorithm learned with awkward fingering becomes an awkward habit that is hard to fix. Learning the correct finger tricks from the start is the key to both speed and retention.
This is probably you if…
- You cannot recall an algorithm without a cube in your hands, but your hands know it instantly.
- Watching someone execute an algorithm teaches you more than reading it.
- You remember algorithms as a rhythm or a feeling rather than as a sequence of letters.
- You tend to learn by doing and get little from reading.
How the trainer adapts
- Each algorithm comes with a finger-trick video or animation shown before the notation.
- Regrips and finger-trick notes (for example "push with the ring finger") are shown alongside each move.
- Sessions emphasize physical repetitions and execution timing rather than reading.
- Alternative fingerings for the same algorithm are available where common variants exist.
Tips
- Execute a new algorithm slowly and correctly several times before trying it at speed. Speed comes from smoothness, not effort.
- If an algorithm feels awkward, try an alternative version with different fingering rather than forcing the uncomfortable one.
- Film your own execution occasionally; it reveals regrips you do not notice while solving.
What it is
Deliberate practice means spending your time deliberately on the things you are worst at, with clear feedback after each attempt. Rather than running through a whole algorithm set evenly, you let your own performance data decide what to work on: the slowest cases, the most-missed cases, the ones you hesitate on.
This is the practice style used by high-level performers in most skill domains. It is efficient but demanding: you are, by design, always working at the edge of what you can do.
This is probably you if…
- You want data on your performance and will act on it.
- Practicing cases you already know well feels like wasted time.
- You are motivated by seeing your weakest numbers improve.
- You are comfortable with practice being difficult and sometimes discouraging.
How the trainer adapts
- The trainer tracks accuracy, recognition time and execution time for every case.
- The practice queue is weighted toward your lowest-performing cases.
- After each attempt you get immediate feedback: correct or not, time, and comparison with your average.
- You can review a ranked list of your weakest cases and choose to focus a session on just those.
Tips
- Pair this type with occasional full-set reviews so strong cases do not quietly decay while you focus on weak ones.
- Define what "weak" means for you: slow recognition and slow execution need different fixes.
- Take breaks. Deliberate practice is mentally expensive, and quality drops quickly when you are tired.