Why a short gap can work better than a long one
Spacing works by making the learner retrieve a word after some forgetting has begun. If the gap is too short, the word is still so fresh that the repeat is easy and adds little. If the gap is too long, the word may be forgotten entirely, so the next encounter is more like learning it anew. The classroom comparison that directly tested short versus long lags found that shorter lags between learning episodes produced better retention, measured after a delay of about four weeks [1].
That result fits the idea that primary school children need the word to be retrievable at the next encounter. A short gap keeps it just within reach, so the repeat acts as practice at pulling the word back out rather than as a first exposure. The finding is specific to primary school children learning second-language vocabulary in a classroom, so it should not be assumed to hold for older learners or for much larger numbers of study sessions.
What the classroom comparison actually measured
The direct evidence comes from a study of second-language vocabulary learning by primary school children in a classroom setting, comparing short and long spacing intervals and testing retention after a delay of roughly four weeks [1]. That design matters: the advantage of short spacing was observed on a delayed test, not just immediately after study, which is the kind of retention teachers care about.
Other primary-school vocabulary studies examined spacing in a different way. One compared massed practice, where learning is crammed together, with spaced practice, where it is spread out, and found that spreading the words led to better retention than massing them [5]. That tells us spacing in general helps, but it does not by itself tell us whether a short or a long gap is better. The short-versus-long comparison is the narrower question, and the direct classroom test points toward shorter gaps [1].
Why the picture is not simply "shorter is always better"
Several primary-school vocabulary studies looked at the lag between repetitions without framing it as short versus long spacing. One multi-classroom study set out to examine distributed and retrieval practice but noted that its design ended up being more about the optimal study lag than about spaced versus massed practice, and its retention test came only one to three days later [3]. A short retention interval like that can make almost any schedule look effective, so it does not settle the short-versus-long question.
Retrieval practice is a separate ingredient. Studies of primary school vocabulary learning asked whether actively retrieving words, rather than just restudying them, improves long-term retention [2], and the broader work on distributed and retrieval practice treats these as related but distinct strategies [4]. Because retrieval practice and spacing are often combined in real classrooms, a study that changes the gap between sessions may also change how much retrieval happens. That makes it hard to attribute a retention difference to the gap alone unless the design isolates it, which is why the direct short-versus-long comparison [1] carries more weight for this specific question than the broader spacing studies.
What this means for planning vocabulary practice
For a teacher deciding how to spread vocabulary practice, the most directly relevant evidence is that shorter gaps between learning episodes beat longer ones for primary school children's delayed retention [1]. The practical reading is to revisit new words soon enough that children can still recall them, rather than leaving a long stretch before the next encounter. Spacing the words at all is better than cramming them together [5], and adding retrieval practice is a further, separate lever [2][4].
The scope is worth keeping in view. The short-spacing advantage was found in a classroom study of second-language vocabulary with primary school children and a retention test about four weeks out [1]. It does not establish an ideal gap length, and the other primary-school studies either tested much shorter retention intervals [3] or compared spacing with massing rather than short with long gaps [5]. So the honest answer is that short spacing beat long spacing in the comparison that tested it directly, while the exact best gap remains an open question.
Sources used in this answer
Input spacing and the learning of L2 vocabulary in a classroom context
In a classroom study of second-language vocabulary with primary school children, shorter lags between learning episodes led to better retention than longer lags, measured after about four weeks.
The effect of retrieval practice in primary school vocabulary learning
Retrieval practice was examined for its effect on long-term retention of new vocabulary in primary school children, but the study did not compare short versus long spacing intervals.
Distributed practice and retrieval practice in primary school vocabulary learning: A multi‐classroom study
A multi-classroom study of distributed and retrieval practice in primary school vocabulary learning focused on optimal study lag rather than spaced versus massed practice, with a retention test only one to three days later.
Distributed Practice and Retrieval Practice in Primary School Vocabulary Learning
The work examined whether spacing primary school vocabulary learning sessions improves retention and treated distributed practice and retrieval practice as strategies for vocabulary learning.
Spreading the words: A spacing effect in vocabulary learning
Spaced presentation of new words led to better retention than massed presentation for Grade 3 children learning the meanings of 15 new words.