23-017 · EVIDENCE A · LOW-RISK READING
Mixing different types of exercises rather than doing twenty of the same kind
几种题型混着练，别一种连做二十道
No cost at all. Simply rearrange the order of exercises in your workbook so they’re not grouped by type. The downside is that mixing them often lowers your immediate accuracy, which can feel like a step backward.
Practicing one type of problem over and over makes it feel easy at the time, but later you might struggle to recall which method to apply. Mixing several types together yields better long‑term results — though the improvement is moderate. This technique works best with visual material such as drawings or categorization tasks; it has a smaller effect on math problems, and for vocabulary memorization it’s actually better to practice in blocks.
A
Move “Mixing different types of exercises rather than doing twenty of the same kind” to Future Me's calendar
The excuse
Future Me has a clearer calendar, more patience, and apparently the same responsibilities.
What feels good now
Today I keep a few minutes, a bit of persistence and avoid interrupting my default routine.
What the source says
The source summarizes the return this way: “Practicing one type of problem over and over makes it feel easy at the time, but later you might struggle to recall which method to apply. Mixing several types together yields better long‑term results — though the improvement is moderate. This technique works best with visual material such as drawings or categorization tasks; it has a smaller effect on math problems, and for vocabulary memorization it’s actually better to practice in blocks.” Evidence grade A. The original states the population, caveats, and sources; this is not a personal forecast.
Open the original entry, check who the evidence applies to, then pick one small step that fits.
THE ENGLISH EDITION
Original recommendation and evidence
17. Mixing different types of exercises rather than doing twenty of the same kind
Cost: No cost at all. Simply rearrange the order of exercises in your workbook so they’re not grouped by type. The downside is that mixing them often lowers your immediate accuracy, which can feel like a step backward.
In plain terms: Practicing one type of problem over and over makes it feel easy at the time, but later you might struggle to recall which method to apply. Mixing several types together yields better long‑term results — though the improvement is moderate. This technique works best with visual material such as drawings or categorization tasks; it has a smaller effect on math problems, and for vocabulary memorization it’s actually better to practice in blocks.
Benefit: This finding comes from a comprehensive meta‑analysis that included 59 studies, 238 effect sizes, and 158 samples. Overall, interleaved practice produces a moderate benefit compared with blocked practice (Hedges’ g = 0.42). When broken down by material, visual tasks benefit the most (g = 0.67), while math problems show a smaller effect (g = 0.34). No significant advantage was found for expository texts or taste‑related tasks; for vocabulary, blocked practice actually performed better (g = −0.39). Regression analysis indicates that the benefit grows when categories are more similar, when items within a category are less alike, and when the material itself is more complex.
Evidence grade: A
Sources: Brunmair M, Richter T (2019). Similarity matters: A meta-analysis of interleaved learning and its moderators. Psychological Bulletin, 145(11), 1029–1052. <https://doi.org/10.1037/bul0000209>
Notes: The review cited in item 16 — which warns against relying solely on highlighting — rated interleaved practice as having a moderate effect, noting that evidence at the time was limited. Taken together, these studies suggest that interleaving works well for material that requires you to recognize a category and pick the right method, such as different math problem types, medical cases, or legal statutes. In contrast, content meant for rote memorization — like vocabulary or definitions — is better learned through blocked practice.
Read the complete Chinese source and notes
17. 几种题型混着练，别一种连做二十道
成本: 不花钱。把练习册按题型顺序做改成打乱着做。难在混着练当场正确率更低，感觉像退步。
说人话: 同一类题连做一大串，当场做得很顺，过后不一定分得清该用哪个方法。把几类题打乱着练，长期效果更好，属于中等幅度。但这招挑材料：练图像、练分类最有效，数学题小一些，背单词反而不如分块练。
收益: 一份纳入 59 项研究、238 个效应量、158 个样本的多层荟萃分析。交错练习相对于分块练习的总体效应属于中等（Hedges' g = 0.42）。按材料分：绘画类最好（g = 0.67），其他视觉材料也较好；数学任务的效应较小（g = 0.34）；说明性文本和味觉材料的总体效应不显著；词汇类反而是分块练习更好（g = −0.39）。回归分析显示：类别之间越相似、同一类别内部越不相似、材料越复杂，交错的好处越大。
证据等级: A
来源: Brunmair M, Richter T (2019). Similarity matters: A meta-analysis of interleaved learning and its moderators. Psychological Bulletin, 145(11), 1029–1052. <https://doi.org/10.1037/bul0000209>
备注: 第 16 条（别把划重点当主要学法）引的那份综述把交错练习评为中等一档，理由是当时研究还不够多。两份合起来看，用法是：需要「认出这是哪一类、该用哪个方法」的内容混着练，比如数学题型、病例、法条适用；需要死记的内容分块练，比如单词、名词解释。
YOUR PRIVATE LIST