Interleaving: Why Mixing Up Spelling Words Beats Drilling One Pattern

Here is a familiar story from many classrooms. The week's list is all -tion words: station, motion, action, fraction, invention. By Friday nearly everyone spells them perfectly. Three weeks later, a child writing a story about a science fair produces "invension", and a week after that "confution". Nothing about the Friday test was fake. The children really could spell those words, on that day, in that setting. What they could not yet do was the thing spelling actually demands in real writing: decide, with no list to cue them, which pattern a word belongs to.

That gap between performance during practice and performance later is the subject of a surprising finding in learning research. It has a name. Interleaving means mixing different kinds of problems within a practice session, instead of practicing one kind at a time. It consistently feels worse to learners. And in study after study, it produces better learning.

This article walks through the key experiments, from college geometry to landscape painting, and explains why mixed practice works even though it feels as if it doesn't. It is honest about what we do and do not know for spelling specifically. And it shows how to reshape a spelling week so that old words, new words and competing patterns are shuffled together.

Blocked and interleaved, in plain terms

Suppose you have three spelling patterns to learn: -tion, -sion and -cian, the three common ways English writes the "shun" sound at the end of a word. The approach called blocked practice works through them one at a time: ten -tion words, then ten -sion words, then ten -cian words. The approach called interleaved practice uses exactly the same thirty words, shuffled, so that "musician" might follow "station" and precede "confusion".

Nothing else changes. Same words, same number of attempts, same time spent.

The difference is what each attempt asks of the learner. In a blocked set, once you have spelled two -tion words correctly, you know the ending of the third before you hear it. You are executing a pattern someone has already chosen for you. In an interleaved set, every word begins with a question: which of these is it? That question is exactly the one a writer faces on a blank page, and blocked practice never lets you rehearse it.

What the experiments found

The strongest evidence for interleaving comes from mathematics, where it has been tested repeatedly and in real schools.

In a 2007 study, Doug Rohrer and Kelli Taylor taught college students to calculate the volumes of four unusual solids. One group practiced in blocks, all the problems for one shape and then all for the next. The other group worked the same problems in a shuffled order. During practice, the blocked group did better, which is what anyone would predict: they always knew which formula was coming. On a test one week later, the result reversed dramatically. The mixed group scored about 63 percent and the blocked group about 20 percent, roughly a threefold difference.

Would it hold with children? In 2010, Taylor and Rohrer ran a similar study with fourth graders learning four kinds of problems about prisms. On a test the next day, children who had practiced mixed problems scored about 77 percent; those who had practiced in blocks scored about 38.

Then the research moved into ordinary classrooms. In a 2014 study, Rohrer and colleagues gave seventh graders either mixed or blocked homework over nine weeks, and on a surprise test two weeks later the mixed group scored about 72 percent against 38 for the blocked group. A larger trial, published in 2020, ran across 54 seventh-grade math classes. A month after the practice ended, students who had done interleaved assignments scored about 61 percent and those who had done blocked assignments about 38.

These are unusually large effects for education research, where a change of a few percentage points is often considered a success. They also replicated, which matters more than the size of any single result.

The paintings experiment

Mathematics has explicit rules and procedures, so it is fair to ask whether interleaving only helps when there is a formula to choose. In 2008, Nate Kornell and Robert Bjork tested something with no formula at all: artistic style.

Participants studied paintings by twelve landscape artists, most of them little known, six paintings per artist. Some artists' paintings were shown in blocks, six in a row. Others were shown interleaved, mixed in among the other painters. Afterwards, participants saw new paintings they had not studied and had to name the artist.

They were more accurate for the artists whose works they had seen interleaved. Seeing one painter's work beside another's seemed to teach the eye what made each painter distinctive, in a way that six paintings by the same artist in a row did not.

The most striking result came next. When asked which method had helped them learn, most participants said blocking, or that the two were equal, even though their own scores showed the opposite. They had learned more from the method they believed was worse.

Why mixing works, and why it feels wrong

Researchers have proposed two main reasons for the interleaving advantage, and both apply comfortably to spelling.

The first is discrimination. When examples of different categories sit side by side, the learner is forced to notice what distinguishes them. A -sion word next to a -tion word invites the question of why they differ, and the answer (words from verbs ending in d or de, like "decide" and "decision", often take -sion) becomes part of what is learned. Blocked practice hides the contrast, because every item in the block looks the same.

The second is choosing the approach. Real tasks rarely announce which method they require. A math test does not label its problems by chapter, and a sentence you are writing does not label its words by spelling pattern. Interleaved practice rehearses that choice on every item. Blocked practice lets you skip it.

A third benefit comes along almost for free. Mixing naturally puts gaps between examples of the same kind, which introduces a small dose of spacing, the well-established advantage of spreading practice out over time.

So why does it feel worse? Robert Bjork coined the term desirable difficulties in the 1990s to describe conditions that slow down performance during practice but improve long-term learning. Interleaving is a classic example. During blocked practice, each correct answer comes easily, and that ease is intoxicating; it feels like mastery. Researchers call this a fluency illusion. Mixed practice produces more errors and more hesitation, which feel like failure, even though the effort of choosing is precisely what builds durable, flexible knowledge.

The practical lesson is uncomfortable. Your sense of how well practice is going is a poor guide to how much you are learning. If a session feels easy, it is worth asking whether it is easy because you have learned the material or because the format is doing the thinking for you.

What this means for spelling, honestly

It would be neat to cite a large classroom trial showing that interleaved spelling lists beat blocked ones. The truth is that the strongest interleaving evidence comes from mathematics and category learning, and the research on spelling specifically is much thinner. Anyone who tells you the spelling result is settled is getting ahead of the evidence.

What we can say is that spelling is full of exactly the kind of decisions where interleaving helps most. Should it be -tion or -sion? -able or -ible? Double the consonant or not? -ance or -ence? Each is a choice between confusable options, and the research points to confusability as the key condition. Paulo Carvalho and Robert Goldstone showed in 2014 that interleaving helped most when the categories being learned were similar to one another, while for very different categories, blocking did about as well or better.

That result suggests a sensible middle path. When a pattern is brand new, a short blocked introduction lets the learner see it clearly. Once it is understood, mixing it with its competitors trains the decision that real writing requires. If you want to go deeper into one of these contrasts, our guides to choosing between -tion, -sion and -cian and between -able and -ible are natural places to practice mixing.

How to shuffle a spelling week

Most school spelling programs are organized around one pattern per week, which is a reasonable way to introduce patterns. The change interleaving suggests is small: keep the pattern of the week for teaching, but mix the practice.

The Friday change is the one that makes the biggest difference. A test that includes words from three weeks ago tells you, and your student, what has actually been retained rather than what was crammed on Thursday night. Teachers who try it often find it sobering at first, and very informative.

A few principles hold whatever the setup. First, practice from memory rather than by copying. Say or play the word and have the learner write it without seeing it, since retrieval is where most of the learning happens, as our piece on why trying to spell a word beats looking at it explains. Second, shuffle the order each time, so that no word is cued by the one before it. Third, when an error appears, name the choice the learner faced ("you picked -tion; this one comes from 'confuse', so it's -sion") rather than simply showing the correct spelling.

For adults working on their own words, the same idea applies. Resist the temptation to drill all your double-letter words in one sitting. Mix them with your -ance/-ence words and your homophones, so each word arrives without a clue. Spelling.School mixes review this way by design: words from different lists and patterns come back in an interleaved queue, each scheduled for the moment it is about to fade.

Does it work for young children?

The evidence with children is encouraging but limited, and it comes mainly from mathematics. Taylor and Rohrer's 2010 study involved fourth graders, and the classroom trials involved seventh graders, all with clear benefits.

For the youngest learners, a sensible approach is to lean more heavily on blocked introduction. A five-year-old meeting the idea that "sh" is written with two letters needs to see it clearly and often before being asked to choose between "sh" and "ch". Interleaving works best once each option is at least partly familiar, which is the same principle that applies to adults, just with a longer introduction.

From around second or third grade, when children are juggling several vowel patterns and ending rules at once, mixing old and new words is both practical and well aligned with the research. Keep sessions short, keep the tone light, and expect more errors during mixed practice than during blocked practice. Those errors are not a sign that the method is failing. They are the method working.

What to do next

Take one change and try it for a month: add three or four words from earlier weeks to every practice session and every test, shuffled in with the new ones. If you teach, keep the pattern of the week for Monday's introduction, then mix from Tuesday on, and include a couple of genuinely competing patterns side by side.

Watch the Friday scores and, more importantly, watch the spelling in free writing a month later, which is where the real difference shows up. If you would like to see interleaved review in action on your own words, you can try five words free on Spelling.School.

And when mixed practice feels harder and messier than drilling one pattern, remember the painting study. The learners who felt least sure were the ones who learned most.

Questions people ask

What is interleaving in learning?

Interleaving means mixing different types of problems or items within a single practice session, instead of practicing one type at a time. In spelling, that might mean shuffling -tion, -sion and -cian words together rather than practicing each ending as a separate block. Research, mainly in mathematics and category learning, shows that interleaved practice usually produces better long-term performance, because learners must decide which approach each item needs.

Why does mixed practice feel harder than blocked practice?

In blocked practice, the correct approach is obvious from the item before, so answers come quickly and feel like mastery. Mixed practice forces you to work out which pattern applies each time, which produces more hesitation and more errors. Psychologist Robert Bjork calls this kind of effort a desirable difficulty: it slows performance during practice but strengthens learning. Studies have found that learners often rate blocked practice as more effective even when their own test scores show the opposite.

Should spelling lists be grouped by pattern?

Grouping by pattern is useful for introducing a new pattern, because it lets learners see the rule clearly. Once the pattern is understood, mixing it with competing patterns and with older words is likely to produce better long-term spelling, since real writing never tells you which pattern a word follows. A practical approach is to introduce the week's pattern in a group, then practice and test it mixed with other words.

How do I mix old and new spelling words?

Add three or four words from previous weeks to each practice session, shuffled in with the new words so that no word is predictable from the one before. Include a competing pattern where possible, such as -sion words alongside -tion words. Practice from memory rather than by copying, and include a few older words on the weekly test. This combines interleaving with spaced review, which together help words last beyond Friday.

Does interleaving work for young children?

Studies have found clear benefits of interleaved practice for fourth graders and seventh graders, mostly in mathematics, while research with younger children and with spelling specifically is more limited. For very young learners, it makes sense to introduce each pattern clearly on its own first, then begin mixing once the options are familiar. From around second or third grade, mixing old and new spelling words fits both the research and the way children use spelling in real writing.

Is interleaving the same as spaced repetition?

They are related but different. Spaced repetition means spreading practice of the same item over time, with gaps between reviews, while interleaving means mixing different kinds of items within a session. Interleaving naturally introduces some spacing, because examples of the same kind are separated by other items. Using both together, mixed sessions with words returning over days and weeks, is a strong combination for spelling.

Fix the words you actually get wrong

A few minutes a day, spaced repetition, built only for spelling. Try five words without signing up.