apply-interleaving

name	apply-interleaving
description	Use when designing practice schedules that mix different problem types or topics to improve discrimination and long-term retention
source	Rohrer & Taylor "The Shuffling of Mathematics Problems Improves Learning" Instructional Science (2007); Kornell & Bjork "Learning Concepts and Categories" Psychological Science (2008)
tags	["interleaving","learning-science","practice","curriculum","problem-solving"]
related	["apply-productive-failure-technique"]
verified	true

Apply Interleaving

Design practice and study schedules that alternate between different problem types, topics, or skills rather than practicing one type to mastery before moving to the next, producing superior discrimination and long-term retention.

Why This Is Best Practice

Adopted by: Sports science training methodology, musical instrument instruction (practice regimens mixing scales, repertoire, sight-reading), effective math tutoring programs Impact: Rohrer & Taylor (2007) showed interleaved mathematics practice produced test scores 43% higher than blocked practice (studying one problem type until mastered, then the next); Kornell & Bjork (2008) showed interleaving improved concept generalization by 78% for visual art style identification Why best: Blocked practice feels efficient (good performance during practice) but produces shallow, context-bound learning; interleaving feels harder (more errors during practice) but builds the critical skill of discrimination — identifying which approach to use when.

Sources: Rohrer & Taylor Instructional Science (2007); Kornell & Bjork Psychological Science (2008); Rohrer "Interleaving Helps Students Distinguish Among Similar Concepts" Educational Psychology Review (2012)

Steps

1. Identify candidate skills/topics — select 3–6 related skills or problem types that learners need to discriminate between; interleaving is most powerful when skills require selecting the correct approach (which is itself a skill).
2. Confirm prerequisite encoding — learners must have initial exposure to all skills before interleaving begins; interleaving before any encoding produces confusion, not learning.
3. Design the interleaved practice set — create a practice set where problems from different skill areas are randomly or systematically mixed; each problem requires the learner to first identify which skill applies, then apply it.
4. Reduce blocked practice to introduction only — use blocked practice (same skill type repeated) only for initial encoding of a new concept; switch to interleaved practice after 1–2 initial sessions.
5. Mix at the problem level, not just the session level — interleaving 20 math problems across 3 types in one session is effective; completing 10 of one type in Monday's session and 10 of another type in Tuesday's session is blocked, not interleaved.
6. Accept performance dip during practice — interleaved practice produces more errors during the practice session than blocked practice; this is expected and indicates the learning difficulty that produces better retention (desirable difficulties, Bjork).
7. Build discrimination prompts — for learners new to interleaving, include a step 0 question: "What type of problem is this? How do you know?" before solving; this explicitly trains the selection skill.
8. Sequence: easy blocked introduction → mixed interleaved practice → transfer assessment — use this 3-phase structure: introduce with blocked examples, practice with interleaving, test transfer to new contexts.
9. Apply to concept learning, not just problems — interleaving works for concept learning: mix examples of Concept A, Concept B, and Concept C rather than teaching all of A, then all of B; learners form discriminating features through comparison.
10. Manage learner frustration — explain the interleaving effect explicitly to learners: "This feels harder and produces more errors but produces much better learning than repeating the same type"; buy-in improves persistence.

Rules

• Interleaving requires prior exposure to all interleaved items — do not interleave concepts that have not been encoded; teach first, interleave in practice.
• The difficulty of interleaving is the mechanism — do not remove the discrimination challenge by flagging which concept each problem belongs to; learners must identify this themselves.
• Blocked practice for complex novel concepts is appropriate for the first 1–2 exposures — interleaving is a practice strategy, not an introduction strategy.
• Interleaving must be explicit — mixed homework sets that do not require discrimination (e.g., chapter review sections mixing trivially labeled question types) are not interleaving.
• Do not interleave unrelated skills — interleaving works because learners learn to discriminate similar, easily confused concepts; mixing completely unrelated topics produces no advantage over blocked.

Common Mistakes

• Interleaving before any encoding — interleaving a concept a learner has never seen produces confusion and frustration without the learning benefit.
• Telling learners which concept each problem uses — labeled interleaving eliminates the discrimination requirement; all benefit comes from requiring learners to identify the correct approach.
• Abandoning interleaving because it "feels harder" — the difficulty and error rate of interleaved practice is the mechanism of benefit, not a sign that it is not working.
• Only interleaving at session level — practicing Problem Type A on Monday and B on Tuesday is not interleaved practice; mixing within the session is required.
• Using interleaving for skills that do not require discrimination — interleaving purely procedural skills with no conceptual choice (e.g., practicing penmanship) produces no advantage.

When NOT to Use

• Initial introduction of any new concept (use blocked examples to establish the concept before mixing)
• Skills requiring physical muscle memory where discrimination of approach is not relevant (e.g., basic physical therapy exercises)
• Learning situations where cognitive overload is already high and adding discrimination demand would exceed working memory capacity