Fifths

Science blog ·

Interleaved practice: the counterintuitive way to actually remember what you train

Mixing different exercise types feels harder during practice and produces dramatically better long-term learning. The contextual interference effect, applied to ear training.

interleaving practice contextual-interference music science

There is a finding in motor-learning and educational research that is so reliably misjudged by learners themselves that researchers have given the misjudgment its own name: the illusion of competence. People practicing in a blocked order — drill the same skill repeatedly, then move to the next skill — feel like they are learning faster, perform better in the moment, and rate their own progress higher. Then, on a delayed test, they have learned less than people who practiced the same skills in interleaved (mixed) order, who felt clumsier the entire time [1].

This effect is called contextual interference, and it has been demonstrated across motor skills, mathematics, anatomy, badminton serves, and — directly relevant here — music performance and aural skills.

The music study you should know about

In 2016, Carter and Grahn published a study in Frontiers in Psychology that asked a clean question: does interleaving help with music practice the way it helps with motor learning generally? [2]

They had advanced pianists practice multiple short pieces under either a blocked schedule (work piece A to satisfaction, then piece B, then piece C) or an interleaved schedule (alternate among A, B, and C in short rotations). After the practice phase, an independent panel of judges rated the recordings.

The result: when there was a difference between the conditions, it favored the interleaved practice. The interleaved schedule also produced better self-reported goal-setting, focus, and mistake identification during practice [2:1]. This is in line with a much larger literature on the contextual-interference effect across motor skills [1:1].

Why it works (the part that explains the discomfort)

Blocked practice lets the brain get away with a shortcut: once you’ve solved task A’s retrieval, the second, third, and fourth attempts at A no longer require you to reconstruct the action plan from scratch. You’re effectively running the same loop with the same cached state. Performance feels smooth.

Interleaved practice destroys that cache between attempts. When you return to A after working on B and C, you have to reconstruct the action plan, which is a much more effortful — and more memory-strengthening — operation [3]. The discomfort during practice is the cognitive work of reconstruction, and that work is exactly what produces durable, transferable skill.

The implication is uncomfortable: the version of practice that feels worse in the moment is often the version that produces more learning. Learners who optimize for the feeling of fluency during practice systematically under-train.

Applied to ear training

Most ear-training tools, including Fifths until very recently, are blocked by default. A session is “all major thirds and minor thirds” or “all I-V-vi-IV progressions.” Within the lesson, you may interleave items (major-third, minor-third, major-third) — that is good. But across the larger curriculum, you almost never see intervals and chords and scale degrees mixed in a single session.

The research suggests this is leaving learning on the table. The most likely high-leverage adjustment is a mixed-review session that draws from any topic you’ve previously trained, alternates types unpredictably, and asks you to first identify which kind of question this is before answering it. That second step — type identification — is the additional cognitive work that the contextual-interference effect predicts will produce more durable learning.

A second study, by Pan and colleagues on implicit sequence learning, found that interleaved practice benefited not just retention but transfer to new sequences, with the largest effects appearing when the underlying material shared structure across tasks [4]. In ear-training terms: drilling intervals and triads in mixed order does not just help you remember each — it helps you generalize, because intervals are the building blocks of triads. The mixed practice forces you to see the connection.

How much interleaving is too much?

There is a real boundary. The contextual-interference effect requires that the learner already has some representation of each task to interleave between. If you genuinely do not yet know what a major third sounds like, drilling it interleaved with seventh chords is not productive — it is noise. The ordering matters:

  1. Initial acquisition of a new sound benefits from short blocks (a few minutes of pure focus on the new item).
  2. Consolidation — once you can identify the item correctly more often than not — benefits from interleaving with related, already-learned items.
  3. Long-term mastery and transfer benefit from broad interleaving across topics.

Most ear-training apps stop at stage 1. The contextual-interference literature is the case for building stages 2 and 3 explicitly into the practice schedule.

A practical interleaved session

If you want to apply this without waiting for an app to do it for you:

  1. List 4–6 categories of ear-training material you have at least basic familiarity with — say: intervals, scale degrees in major, scale degrees in minor, triad qualities, simple progressions.
  2. For your session, draw items randomly across all categories. A coin flip, a die, or a shuffled deck of category cards works.
  3. Before answering, identify the category. (“This is a triad — major.”) The category-identification step is part of the practice.
  4. When you miss, do not re-drill that item three times. Note the miss and move on. The next time it comes up — possibly several questions later — let your brain do the reconstruction.

Sessions structured this way feel slower and harder than blocked sessions of equivalent length. They will likely produce lower in-session accuracy. They will, on the timescales that matter — weeks and months — produce more durable, more transferable musical hearing [1:2] [2:2].

The takeaway

Trust the evidence over the feeling. Blocked practice produces the satisfying sensation of fluency; interleaved practice produces the actual capacity to use the skill outside the practice context. For an ear-training app, the most valuable feature beyond the basic lesson grid is a mixed-review mode. For a learner working on their own, the most valuable adjustment to a practice routine is shuffling categories instead of drilling them one at a time.

It will feel worse. The data are clear that you are learning more.

References

  1. Brady, F. (1998). A theoretical and empirical review of the contextual interference effect and the learning of motor skills. Quest, 50(3), 266–293. See also the broader review and accessible summary at https://effectiviology.com/interleaving/. ↩︎ ↩︎ ↩︎

  2. Carter, C. E., & Grahn, J. A. (2016). Optimizing Music Learning: Exploring How Blocked and Interleaved Practice Schedules Affect Advanced Performance. Frontiers in Psychology, 7, 1251. https://doi.org/10.3389/fpsyg.2016.01251. PubMed: https://pubmed.ncbi.nlm.nih.gov/27588014/ ↩︎ ↩︎ ↩︎

  3. Lee, T. D., & Magill, R. A. (1985). Can forgetting facilitate skill acquisition? In D. Goodman, R. B. Wilberg, & I. M. Franks (Eds.), Differing Perspectives in Motor Learning, Memory, and Control (pp. 3–22). Elsevier. ↩︎

  4. Pan, S. C., Tajran, J., Lovelett, J., Osuna, J., & Rickard, T. C. (2021). Interleaved practice benefits implicit sequence learning and transfer. Memory & Cognition. https://doi.org/10.3758/s13421-021-01168-z. PMC: https://pmc.ncbi.nlm.nih.gov/articles/PMC8476370/ ↩︎

← Back to all articles