Blocked Versus Random Practice¶

Why It Matters¶

Intuitive practice is blocked - repeat the same thing until you "get it," then move on. This feels effective (rapid improvement within session) but produces poor long-term retention. Random practice feels inefficient (slower within-session improvement) but creates more durable learning. This tool corrects the systematic error of optimizing for immediate performance rather than long-term learning.

The Key Move¶

For skill acquisition, use blocked practice early (establishing basic patterns and building confidence), then shift to random practice for consolidation and durability. In random practice, deliberately interleave different skills or variations within a session. The interleaving creates retrieval practice - each repetition requires reconstructing the skill from memory rather than maintaining it in working memory. This reconstruction strengthens retention.

Classic Application¶

Baseball batting practice. Blocked practice: 20 fastballs, then 20 curveballs, then 20 changeups. The batter knows what's coming and optimizes timing for each pitch type. Random practice: fastball, curveball, changeup, fastball, fastball, curveball - unpredictable sequence. Random practice is harder (more swings and misses) but produces better game performance where pitch type is unknown. The random condition forces the batter to identify pitch type and adjust, not just time a known pitch.

Surprising Application¶

Learning programming concepts and problem-solving. Blocked practice: solve 10 recursion problems in a row, then 10 dynamic programming problems. You get good at recognizing "this is another DP problem." Random practice: recursion, DP, graphs, recursion, strings, DP - mixed problem types. Random practice is harder (you must first identify problem type, not just apply a known technique) but produces better transfer to new problems where type isn't labeled. The tool makes explicit why mixed problem sets create better learning than topical problem sets.

Failure Modes¶

Random practice too early, before basic skill patterns are established, creates confusion and error reinforcement. The tool requires appropriate sequencing: blocked first, then random. It fails when skills don't transfer well (very distinct skills with no common elements benefit less from interleaving). Also fails when feedback is delayed or unclear - random practice requires knowing when you're wrong. The worst failure is abandoning random practice because it feels inefficient, optimizing for session performance rather than learning outcomes.

Go Deeper¶

Shea, J.B. & Morgan, R.L. (1979). "Contextual Interference Effects on the Acquisition, Retention, and Transfer of a Motor Skill." Journal of Experimental Psychology: Human Learning and Memory, 5(2), 179-187; Rohrer, D. & Taylor, K. (2007). "The Shuffling of Mathematics Problems Improves Learning." Instructional Science, 35(6), 481-498.