Exam Prep
GMAT Focus EditionLiveExecutive AssessmentIn DevelopmentGRE General TestIn DevelopmentLSATIn DevelopmentMCATIn DevelopmentSATIn DevelopmentACTIn DevelopmentAll Exam Guides
Lists
MBA RankingsLiveUndergraduateComing SoonLaw SchoolsComing SoonMedical SchoolsComing Soon
Resources
The Study Room (Blog)Exam GuidesCompare PlansFAQTerms of UsePrivacy Policy
Community
All ForumsLiveGMAT PrepStudy LogsAdmissionsQuestion of the DaySite Feedback
Account
Sign InCreate Account

Interleaving: Why Mixed Practice Beats Topic Blocks

Aisha Thompson · September 6, 2026 · 6 min read

Interleaving means mixing different problem types within one practice session, in a pattern like ABC ABC ABC instead of AAA BBB CCC. Mixed practice beats topic blocks because a mixed set forces you to recognize each problem and choose a method before applying it, and that choice is what the exam actually tests. The GMAT Focus Edition asks 64 questions, 21 Quant, 23 Verbal, and 20 Data Insights (mba.com, 2026), and none of them is labeled by chapter.

What separates interleaved practice from blocked practice?

Blocked practice means solving many problems of the same type in a row: twenty rate problems, then twenty exponent problems, then twenty Critical Reasoning questions. It is the default almost everyone inherits, because textbooks are organized by chapter and chapter exercises block themselves. Interleaved practice shuffles those same problems together, so consecutive questions rarely share a method.

The research contrast between the two is well established and runs in a direction most students find uncomfortable. In studies of mathematics practice by Doug Rohrer and colleagues, students doing blocked work looked stronger during the practice itself, then lost that edge on delayed tests, where the interleaved groups came out ahead. Nate Kornell and Robert Bjork found the same shape outside of math: people shown painters' styles in mixed order got better at identifying new paintings by those artists than people shown one artist at a time, and most participants still believed blocking had served them better. That last detail matters. Interleaving does not merely beat blocking; it beats blocking while convincing you it is losing.

Why does mixed practice feel harder and store better?

Blocking lets you load a method once and reuse it. By the fifth rate problem in a row, you are no longer deciding how to solve anything. You are executing a procedure that is already sitting in working memory, and each repetition gets smoother. That smoothness is the fluency illusion at work: ease of execution in the moment gets mistaken for durable learning.

Interleaving takes that shortcut away. Every switch forces you to set one method down and retrieve the next from long-term memory, and effortful retrieval is precisely what strengthens a memory. Cognitive scientists in the tradition of Robert Bjork call this a desirable difficulty: a condition that makes practice feel worse while making retention better. Mixing also smuggles in a second advantage. When rate problems are spread through a week instead of packed into Monday, each return to the topic arrives after a delay, which is the same mechanism behind spaced repetition. An interleaved plan spaces itself.

What is discrimination learning, and why does it decide GMAT questions?

Discrimination learning means learning to tell similar things apart by the features that actually distinguish them. It is the quiet half of every GMAT skill. Knowing how to solve a weighted average problem is worth little if you cannot tell, in the first ten seconds, that a problem is a weighted average problem and not a simple ratio wearing a disguise.

Blocked practice removes that decision entirely. When the page says Rates, the hardest step of every problem on it has been answered for you before you start. Mixed practice restores the decision, and it does something subtler too: it puts confusable neighbors side by side. Solving a rate problem immediately after a work problem forces you to notice exactly where the two setups differ. The same logic runs through the verbal side. Strengthen and evaluate questions read alike until you have met them back to back. Data Sufficiency turns almost entirely on one discrimination, whether the question needs a specific value or only a relationship, and that judgment sharpens fastest when the two kinds alternate. Choosing the method is part of the skill, so practice that hides the choice trains a smaller skill than the exam measures.

What does an interleaved week look like for the same hours?

Here is one week of prep, eight hours in both columns. The hours and problem counts are illustrative, not prescriptions. The difference to notice is structural: the left column meets each topic once, in one large dose, while the right column meets every topic several times, in small doses, with a decision before every problem.

DayBlocked week (8 hours)Interleaved week (8 hours)
Monday, 1 hourRate and work problems, one chapter start to finishMixed set: rates, exponents, two Critical Reasoning, one Data Sufficiency
Tuesday, 1 hourExponents and roots drillMixed set: work, number properties, one Reading Comprehension passage, one table question
Wednesday, 1 hourCritical Reasoning onlyMixed set: exponents, rates, Critical Reasoning, Data Sufficiency
Thursday, 1 hourData Sufficiency onlyMixed set weighted toward the week's misses so far
Friday, 1 hourReading Comprehension onlyShort timed set drawn from all three sections
Saturday, 2 hoursNumber properties, long single-topic sessionFull mixed timed set, then error log entries for every miss
Sunday, 1 hourReread the week's chaptersRe-solve the week's misses cold, oldest first

Both weeks cover the same material in the same time. The blocked week feels better every single day. The interleaved week performs better when it counts, because by Sunday every topic has been retrieved, dropped, and retrieved again several times, and every problem began with the question the exam always asks first: what kind of problem is this?

How does Start From Nowhere interleave by design?

Start From Nowhere treats mixing as the default rather than a setting to configure. The trainer's rounds draw on 180 original practice items spanning Quant, Verbal, and Data Insights, so a round moves between skills instead of marching through a chapter. The selection is adaptive: every skill carries its own rating, keyed to the skill labels on the official GMAT Focus score report, and the mix leans toward the skills whose ratings lag. Misses return on a spaced schedule, and because a returning miss lands inside a later mixed round, your review arrives interleaved as well, surrounded by problems that do not resemble it. Timing is tracked per question, which matters here because hesitation at the start of a problem is often the method-selection step running slow. The first round is free and needs no account. The trainer supplies the mixing; the retrieval effort on each question is still yours, which is why it pairs naturally with active recall.

Is blocked practice ever the right choice?

Yes, briefly, at the start. When a method is genuinely new to you, a short block is how you learn its mechanics, because you need a few consecutive repetitions to build the procedure at all. There is nothing to discriminate before there is something to execute. The mistake is staying blocked after that point, practicing execution long past the moment when selection became the real bottleneck. A workable rule: block until you can solve the new problem type slowly but correctly without notes, then release it into the mix and never drill it in isolation again. From that day on, the mix does the teaching. It will feel rougher than the chapter drills you are used to. That roughness is not a malfunction. It is the sensation of practicing the entire skill, including the part the answer key cannot see: knowing what to do before you do it.

Put this into practice
Run a free adaptive round. No account needed; the trainer finds your weak skills in one session.
Start a free round

Frequently asked questions

What is interleaved practice?
Interleaved practice means mixing different problem types within one study session, in a pattern like ABC ABC ABC, instead of finishing all of one type before starting the next. The mix forces you to identify each problem and select a method before solving, which is a skill blocked practice never exercises.
Why does interleaving feel less productive than blocked practice?
Blocked practice gets smoother with every repetition, and that smoothness reads as progress. Interleaving keeps every problem cold, so accuracy during the session looks worse. The session feeling measures fluency right now, not what you will retain. Judge a method by what survives to next week.
Does interleaving work for GMAT Verbal and Data Insights, or only for Quant?
It works across all three sections. Verbal rewards telling apart similar question tasks, such as strengthen versus evaluate, and Data Insights rewards deciding on sight what a Data Sufficiency statement must pin down. Those are discrimination skills, and mixing is how discrimination is trained.
Should I interleave a topic I am learning for the first time?
Not immediately. A short focused block is the right way to learn a new method, because you need a few consecutive repetitions to get the mechanics down. Once you can execute the method slowly but correctly, move it into the mix and let interleaving take over.

Keep reading

Feedback loops
practice with a purpose
Study Science

Deliberate Practice: Train Like a Coach Is Watching

September 4, 2026 · 6 min
Spacing effect
forgetting is the tool
Study Science

Spaced Repetition for GMAT Prep: Why Daily Rounds Win

August 17, 2026 · 6 min