Knowing the Spelling Today, Remembering It Weeks Later: An Early Observation from Vocora
A first-person comparison of review routines and later spelling recall
Early observation · one learner
I am both the learner in this informal comparison and the developer of Vocora. The routines and later spelling results are reported retrospectively. They raise a question about retention without isolating the effect of the review schedule.
I wanted to know what remained after a word list felt finished. Being able to spell every word correctly during practice was satisfying, but I was interested in whether I could still produce those spellings several weeks later.
While using Leitner-style review, I sometimes felt that a word returned when recalling it was becoming difficult. This prompted a personal comparison between a list I reviewed daily until all answers were correct and a list that continued through Vocora’s Leitner progression.
The later spelling results differed substantially in my account. The two routines also differed in duration and stopping criteria, so understanding the contrast requires keeping those differences visible.
Table of contents
How I arranged the comparison
I began with a pool of hundreds of vocabulary words, put them into an array, and randomly divided them into two roughly equal sets. My aim was to avoid knowingly selecting easier words for one routine. This was random allocation of words within one learner’s practice. There were no separate groups of learners, and the allocation did not guarantee equal familiarity or spelling difficulty.
Both sets used the same general task in Vocora: hear a word and enter its spelling. The outcome therefore concerned spelling from audio. I did not test whether I could recall the meaning, use the word in conversation, or recognize it in a different context.
| Feature | Self-managed daily review | Leitner-style review |
|---|---|---|
| Practice routine | Listen to the words and write their spellings each day. | Revisit words according to the application’s scheduling and progression rules. |
| Stopping rule | Stop reviewing the set once I could spell every target word correctly during practice. | Continue until the items reached the system’s mastered state. |
| Approximate training duration | One week. | 45 days. |
The system’s mastered state was a progression label, not an independent assessment of permanent learning. Similarly, my all-correct practice result established what I could do at that point, not what would remain later.
Both routines involved spacing across days. The comparison was between different schedules and stopping rules, not spaced practice versus no spacing. I did not match the total number of retrieval attempts, minutes practiced, or feedback exposures. A longer calendar period alone does not establish more practice time or more retrieval attempts.
The delayed spelling check
I remember the later check as following approximately three weeks without review. The account does not establish whether each set was tested after exactly the same delay from its own final practice session. Because the routines took different lengths of time, that missing alignment matters.
My reported results were:
| Practice routine | Approximate correct spellings | Approximate spelling errors |
|---|---|---|
| Self-managed daily review | 60 per 100 | 40 per 100 |
| Leitner-style review | 94 per 100 | 6 per 100 |
These are the per-100 figures from my retrospective account. They do not establish that exactly 100 words were tested in each set. Exact denominators and an item-level score sheet are not included here, so the table should be read as an approximate report, not a newly computed statistical comparison.
The result surprised me because the list I had completed quickly had felt fully learned. More of the spellings from the longer Leitner routine appeared available when I returned to them. That is the observation I want to preserve; the table alone cannot tell me which difference between the routines explains it.
What the comparison can and cannot distinguish
The largest limitation is that I changed several things together. Leitner progression spread practice over a longer calendar period, used a different stopping rule, and may have involved a different amount of retrieval and feedback. Both routines included retrieval, so this was not a test of retrieval versus restudy either.
The word sets could also have differed in initial familiarity or spelling difficulty. Random allocation reduced deliberate selection but did not remove that possibility. The approximate delay, unreported scoring details, and possible incidental exposure further limit interpretation. Finally, I was the learner, the person arranging the comparison, and the person reporting its outcome.
The figures describe an informal contrast in my later spelling performance; they do not estimate how much Leitner scheduling improves retention.
The feeling that a word returned “just before I forgot it” belongs to the same account. I did not measure an individual forgetting threshold or show that the algorithm identified one. Effortful recall is a possible clue for a future study, not a verified mechanism behind the result.
What published research contributes
Cepeda and colleagues (2006) examined distributed practice in verbal recall. Their synthesis found that the interval between learning episodes and the delay before the final test jointly mattered for retention. This makes the timing of both practice and testing central to interpreting a comparison like mine.
Karpicke and Roediger (2008) studied continued retrieval after initially successful learning of foreign-language vocabulary. Repeated testing supported delayed recall in their experiment. That result is relevant to my decision to stop the daily routine once I had answered everything correctly, although their task and design differed from my spelling comparison.
Kang (2016) reviewed educational implications of spacing and retrieval. For second-language learning specifically, Kim and Webb (2022) synthesized comparisons of spacing schedules. Their findings included differences between immediate and delayed testing and no statistically significant advantage of expanding over equal spacing in the analyzed comparison. They do not justify treating one expanding schedule as universally optimal.
These papers support investigating practice timing, continued retrieval, and delayed assessment. None verified my results or tested the particular Vocora routine described here.
What a stronger comparison would need
The next study should first decide which question it is answering. Comparing complete routines would ask what happens when a learner follows each routine as designed. Isolating scheduling would require closer control of total practice and stopping criteria.
For either question, the essential records would be the exact word sets, initial spelling performance, allocation procedure, and every practice and test timestamp. Retrieval attempts, feedback, and time spent would make differences in exposure visible. The final spelling check should use predefined scoring and comparable conditions, with delays anchored to each item’s last relevant practice.
A study aimed specifically at scheduling would then compare routines with matched retrieval opportunities or study time, while reporting any remaining differences. Additional learners and independently checked scoring would make the results more informative beyond my own experience.
Relationship to the other Vocora notes
This account concerns delayed spelling recall. The dictation-attention note concerns an impression about conscious monitoring while answering and contains no measured attention comparison. These retention figures cannot be used to quantify that impression.
The grammar and reflection notes also describe aspects of my learning with the same system. Practice periods may overlap; the notes should not be counted as independent replications, and their different durations should not be added into one assumed sequence.
For spelling learned from audio, how much does continued scheduled review contribute to later recall beyond total practice and the choice of when to stop? This informal comparison gives me a reason to investigate that question more carefully.
References
- Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., & Rohrer, D. (2006). Distributed practice in verbal recall tasks: A review and quantitative synthesis. Psychological Bulletin, 132(3), 354–380.
- Karpicke, J. D., & Roediger, H. L. (2008). The critical importance of retrieval for learning. Science, 319(5865), 966–968.
- Kang, S. H. K. (2016). Spaced repetition promotes efficient and effective learning: Policy implications for instruction. Policy Insights from the Behavioral and Brain Sciences, 3(1), 12–19.
- Kim, S. K., & Webb, S. (2022). The effects of spaced practice on second language learning: A meta-analysis. Language Learning, 72(1), 269–319.
Acknowledgement of AI Use
ChatGPT (OpenAI) assisted with drafting, organizing, and editing this note. The author supplied the underlying observations and retains responsibility for the account and its interpretation.