Expanding retrieval schedules have not shown consistent advantages over equally spaced or contracting schedules matched on total spacing¶

Claim · All claims
Evidence · 3 studies (5 entries) · 1 causal r?, 1 quant-synthesis r?, 1 review r3 · q2–q3
Currency · newest study 2022 · 2 of 5 entries checked against their sources (automated) · not reviewed by a person

Subclaims¶

q2 i? The chapter reports that Karpicke and Bauernschmidt (2011) found no discernable difference among expanding, equally spaced, and contracting relative spacing conditions. → Karpicke 2017 q2 i? The chapter reports, citing Kang et al. (2014), a 3% advantage of an expanding over an equally spaced schedule across days that did not reach statistical significance. → Karpicke 2017 (2) q2 i? The chapter reports, citing Kupper-Tetzel et al. (2014), no evidence for superiority of an expanding schedule, with contracting best at 1-day and 7-day intervals. → Karpicke 2017 (3)

q3 i? A meta-analysis of 48 second-language spacing experiments found equal and expanding spacing statistically equivalent. → Kim and Webb 2022 q3 i? In one experiment with 128 Japanese college students learning 20 English–Japanese word pairs, expanding spacing had a limited but statistically significant advantage over equal spacing, and the author suggests introducing spacing may have a larger effect. → Nakata 2015

Evidence¶

Karpicke 2017¶

Karpicke, J. D. (2017). Retrieval-Based Learning: A Decade of Progress. Learning and Memory: A Comprehensive Reference, 2nd edition, Volume 2. https://doi.org/10.1016/B978-0-12-809324-5.21055-9

q2 · i? · review · r3

The chapter reports Karpicke and Bauernschmidt (2011), which minimized differences in initial retrieval success: "there was no discernable difference between the three relative spacing conditions". No test statistic is printed; equivalence was not tested.

"However, there was no discernable difference between the three relative spacing conditions, expanding, equally spaced, or contracting."

Karpicke 2017 (2)¶

Karpicke, J. D. (2017). Retrieval-Based Learning: A Decade of Progress. Learning and Memory: A Comprehensive Reference, 2nd edition, Volume 2. https://doi.org/10.1016/B978-0-12-809324-5.21055-9

q2 · i? · review · r3

The chapter reports Kang et al. (2014), with sessions spaced 2-6-19 or 9-9-9 days and a test 56 days later: "Kang et al. found a 3% advantage of the expanding schedule relative to the equally spaced schedule", not significant.

"Kang et al. found a 3% advantage of the expanding schedule relative to the equally spaced schedule, a difference that did not reach statistical signi ficance."

Karpicke 2017 (3)¶

Karpicke, J. D. (2017). Retrieval-Based Learning: A Decade of Progress. Learning and Memory: A Comprehensive Reference, 2nd edition, Volume 2. https://doi.org/10.1016/B978-0-12-809324-5.21055-9

q2 · i? · review · r3

The chapter reports Kupper-Tetzel et al. (2014), with relearning sessions expanding, equal, or contracting: "they found no evidence for the superiority of an expanding retrieval schedule"; expanding and equal were best at 35 days.

"Overall, they found no evidence for the superiority of an expanding retrieval schedule. In fact, the contracting schedule produced the best performance at the 1-day and 7-day retention intervals"

Kim and Webb 2022¶

Kim, S. K., & Webb, S. (2022). The effects of spaced practice on second language learning: A meta-analysis. Language Learning, 72(1), 269–319. doi:10.1111/lang.12479

q3 · meta-analysis (abstract only) · i? · no effect size printed in the abstract; spacing effect described as medium-to-large · n=48 experiments (98 effect sizes, N=3,411) · quant-synthesis · r?

This meta-analysis pooled 98 effect sizes from 48 second-language experiments (3,411 participants) on spaced practice, comparing spaced with massed practice, longer with shorter spacing, and equal with expanding spacing on immediate and delayed posttests. The authors report that spacing had a medium-to-large effect on second language learning; that shorter spacing was as effective as longer spacing on immediate posttests but less effective on delayed posttests; and that equal and expanding spacing were statistically equivalent. Differences in the spacing effect across studies were explained by the learning target, number of sessions, type of practice, activity type, feedback timing and retention interval. Read from the abstract only: it prints no pooled effect sizes, confidence intervals or spacing intervals, so the size of each comparison, the gaps the studies used and the learners' ages and proficiency are not established here. The abstract does not say whether the expanding and equal conditions it compared were matched on total spacing, which this claim requires.

"Results showed that (a) spacing had a medium‐to‐large effect on second language learning; (b) shorter spacing was as effective as longer spacing in immediate posttests but was less effective in delayed posttests than longer spacing; (c) equal and expanding spacing were statistically equivalent"

Nakata 2015¶

Nakata, T. (2015). Effects of expanding and equal spacing on second language vocabulary learning: Does gradually increasing spacing increase vocabulary learning? Studies in Second Language Acquisition, 37(4), 677–711. doi:10.1017/s0272263114000825

q3 · peer-reviewed experiment (abstract only) · i? · no effect size printed in the abstract · n=128 · causal · r?

128 Japanese college students studied 20 English–Japanese word pairs; the type of spacing (expanding or equal) and the amount of spacing (massed, short, medium, long) were both manipulated. Expanding spacing had a limited, yet statistically significant, advantage over equal spacing, which the author calls the first L2 study to find expanding superior; the main effect of the amount of spacing was also significant, with large effect sizes, and the author suggests introducing spacing may have a larger effect than its shape. This qualifies the claim: here expanding did beat equal spacing, if modestly. Read from the abstract only: it prints no effect sizes or retention intervals, does not say whether the advantage held at every amount of spacing, and does not say how learners were assigned.

"Results demonstrated a limited, yet statistically significant, advantage of expanding spacing."

Discussion¶

Second-language evidence is split, but small either way. Kim and Webb's (2022) meta-analysis of 48 L2 experiments found equal and expanding spacing statistically equivalent, which supports this claim, while Nakata (2015), a single L2 vocabulary experiment, found a limited but significant advantage for expanding spacing and large effects of the amount of spacing. Both read together suggest the schedule's shape matters less than whether practice is spaced and by how much; both entries rest on abstracts.

Related Claims¶

• Total retrieval spacing improved one-week vocabulary recall while relative schedules were not distinguished — the general claim this finding bears on, with the studies behind it
• Spaced Repetition Improves Retention
• Spaced Retrieval Improves Retention — related
• Expanding and uniform spacing schedules of retrieval practice do not differ significantly in final retention (g = 0.034) — related
• Spaced retrieval practice (successive relearning) produced multi-year retention of foreign language vocabulary in adults, as reported by the review — related
• The effectiveness of dropout schedules depends critically on the length of the interval between presentations of an item — related
• The authors' findings add to evidence that expanding-interval schedules are not more effective than fixed-interval schedules of practice — possibly the same claim (merge candidate)
• Number of retrieval exposures moderates the expanding-versus-uniform schedule benefit: more tests favour the expanding schedule — related
• Expanding and uniform spaced practice better for long retention intervals; contracting better for short intervals — reports the opposite
• Attempted recall during vocabulary study showed no demonstrable superiority over mere reading and re-reading in adult learners — related