Spiral Review Is a Few Different Things: Spacing, Interleaving, and Why Skills Fade
You taught fractions in October. It went well. In February you put a fraction problem on the board and roughly half the class looks at it like they've never seen one.
That isn't a teaching failure and it isn't a class of students who weren't paying attention. Forgetting is what memory does when material isn't revisited, and the fix is a scheduling problem more than an instructional one.
The usual answer is spiral review. The trouble is that "spiral review" names two different techniques that work for different reasons, and most classrooms end up doing a weak version of one and none of the other.
The two things hiding inside "spiral review"
Spacing is about when you come back to something. The gap between practice sessions is the part that does the work.
Interleaving is about what order you mix problems in. Shuffling similar problem types together so a student has to figure out which method applies, instead of doing twenty of the same kind in a row.
They're often treated as one idea because both involve not-teaching-things-in-one-solid-block. But they're separate, they're separately supported, and a Friday worksheet reviewing Monday through Thursday is weak spacing with no interleaving at all.
Spacing: the delay is the active ingredient
The federal practice guide on organizing instruction puts "space learning over time" among its recommendations, at Moderate Evidence, and it's unusually specific about the interval: "arrange to review key elements of course content after a delay of several weeks to several months after initial presentation" (IES/WWC).
The mechanism is counterintuitive, but spacing works because of the forgetting. As the Learning Scientists put it, to get the benefit "you need to forget a little bit" (Weinstein & Smith); the effortful reconstruction of something half-lost is what makes the memory more durable next time. Practice that happens before any forgetting has occurred is just repetition with extra steps.
Practically, that means the useful review is the unit you finished five weeks ago.
Interleaving: the mixing is the active ingredient
Interleaving is "the simple strategy of mixing up concepts to be learned" (Agarwal & Agostinelli, American Educator), and in math it means a practice set where problem types are shuffled rather than blocked.
The size of the effect is the surprising part. Reporting Doug Rohrer's classroom studies, that article documents 4th graders working on prisms scoring 77% after interleaved practice versus 38% after blocked practice on a test a day later; 7th graders on graphs and slope at 74% versus 42% a full month afterward; and a study of 800 7th graders in Florida finishing at 61% versus 38% on a final test.
Why such a gap? Because blocked practice quietly removes the hardest part of the work. When every problem on the page is a fraction-division problem, a student never has to decide that it's a fraction-division problem. They execute a procedure they were handed at the top of the page. As the article puts it, "blocked practice becomes repetitive about procedure." Mix the types and the student has to classify before they can solve, which is exactly what a test, and life, will ask them to do.
Two cautions keep this from going wrong. First, mix similar things: "the key to interleaving is mixing up similar ideas," because the skill being built is telling near-neighbors apart, not lurching between unrelated topics. Second, don't overdo the switching. "Don't switch too often, or spend too little time on any one idea; you need to make sure you've covered some ground before tackling a new idea" (Weinstein & Smith). Students need enough blocked practice to learn a method before mixing makes sense.
One precision note, since these get conflated: the WWC guide also recommends interleaving, at Moderate Evidence, but its recommendation is to "interleave worked example solutions with problem-solving exercises," which means alternating examples with practice. That's a different technique from mixing problem types. Both are useful; they aren't the same recommendation.
Both of them work through retrieval, not review
Here's the thread connecting them. Spacing and interleaving both work by making a student pull something out of memory rather than read it again.
That distinction is the single most durable finding in this literature. In the large review by Dunlosky and colleagues, practice testing and distributed practice were rated high utility, "because they benefited students of many different ages and ability levels and enhanced performance in many different areas," while rereading, underlining and mnemonic devices "were found to be of surprisingly low utility" (APS). The federal guide reaches the same place from another direction, rating the use of quizzing to re-expose students to key content at Strong Evidence, its highest level (IES/WWC).
So a spiral review where students copy a worked example is barely spiral review. The version that works asks them to produce something from memory; and it's fine, in fact necessary, that this is harder.
Where this actually goes in a math block
None of this needs a new program. It needs four small scheduling decisions in slots that already exist.
- The warm-up. Two problems from a unit five weeks back beats two problems from yesterday. This is what makes spiral review worth its slot in your opening five minutes.
- Homework. If something goes home, pull it from three weeks ago rather than tonight's lesson. Old content is the only content a student can practice unsupervised without rehearsing a fresh misconception.
- Centers and independent practice. This is the easiest place to interleave, because a mixed set is no harder to prepare than a blocked one. Mix three related problem types rather than running twenty of one.
- Quick checks. A check on something you taught last month tells you more than a check on what you taught this morning, which mostly measures short-term recall.
A 1st-grade version looks like putting a few "make 10" problems into a February warm-up rather than assuming September settled it. A 5th-grade version looks like a practice set where dividing fractions sits next to multiplying them, so students have to notice which one they're looking at.
Where a practice tool helps
The scheduling is the hard part, and it's the part software can genuinely lighten. In Boddle, adaptive practice pulls across standards, including what a student has already worked on, rather than drilling a single skill. This produces mixed sets without you building them, and the learning-gaps report shows which skills have quietly faded since you taught them. The penalty-free format matters more here than it looks: spaced and interleaved practice should produce more errors in the moment, because that difficulty is the mechanism, and a format that docks students for those errors punishes exactly the practice that's working. It's standards-aligned across math, ELA and science for K-6, and free for teachers and students, so the review schedule doesn't depend on which families upgraded.
What it can't do is know your scope and sequence. No tool knows that your fractions unit ended in October and that February is the right moment to bring it back. The calendar is still yours. You can, however, choose exactly what standards to assign at any time.
Frequently asked questions
What is distributed practice? Distributed practice means spreading study or practice of the same material across multiple sessions separated by time, instead of concentrating it into one block. In a classroom it looks like revisiting a skill several times over weeks or months rather than teaching it, practicing it hard for a week and moving on. It's one of the two techniques with the strongest support in the learning-strategies literature, rated high utility alongside practice testing (APS). The gap between sessions isn't wasted time; it's the part that produces the durability.
How far apart should spiral review be? Further apart than most schedules assume. The federal guidance is to review key content "after a delay of several weeks to several months" from when it was first taught (IES/WWC), which puts October's unit in a December or February warm-up rather than next Friday's worksheet. A reasonable rule of thumb: if students would clearly remember it, you're too early. Some forgetting before review is a feature, not a problem to schedule around.
What's the difference between spacing and interleaving? Spacing is about when you practice: the same skill revisited after a delay. Interleaving is about what order: different but related problem types mixed within one practice set, so students must identify which method applies before solving. You can space without interleaving (the same skill, revisited monthly) and interleave without spacing (a mixed set, all in one week). They're separately supported and they're strongest together, since a mixed set of older material does both jobs at once.
Close
The two moves are simple to state and genuinely uncomfortable to run: come back to things later than feels responsible, and mix problems more than feels tidy. Both make practice look worse in the moment and better in April.
A tool can carry the mixing and flag what's fading, free for every student in the room; deciding when February's warm-up should quietly return to fractions is the part that's yours.
