Fixing the First-Turn Bottleneck: A Better Pre-SRS Laddering System

Why SRS has a First-Turn Bottleneck

LinGoat redesigned pre-SRS vocabulary onboarding by eliminating multiple-choice tests and inserting a 5 to 10 second interleaved distractor between first exposure and the first blind recall test. That fixes the First-Turn Bottleneck: the gap between seeing a brand-new word and having a memory trace strong enough for spaced repetition to schedule honestly. Every correct and incorrect attempt in that buffer is logged and passed to FSRS when the word graduates, so day-one scheduling reflects real difficulty, not a lucky guess from short-term memory.

Spaced repetition systems (SRS) are excellent at maintaining memory once a trace exists. Algorithms like FSRS can predict when you are about to forget a word with remarkable accuracy. But they share a blind spot: the very first turn. If you drop a completely unseen word straight into a review queue, users often pass by reading a sensory echo still in working memory, or they fail repeatedly in one session and corrupt the card's initial difficulty rating. Acquisition (first contact) and maintenance (SRS) require different cognitive mechanisms. LinGoat's new laddering system treats them separately while keeping the data connected.

The Multiple-Choice Trap: The Illusion of Competence

The industry standard for introducing a new word is to show it, translate it, and immediately test with multiple choice. It feels frictionless. It also measures the wrong skill.

Multiple-choice formats test passive recognition, not active recall. When the correct answer is on screen, the brain recognizes it, gets a small reward signal, and can mistake recognition for learning. Worse, incorrect options can harm future recall. Roediger and Marsh (2005) showed that exposure to wrong multiple-choice lures can encode false associations: if you guess the wrong definition, your brain may wire that error to the target word, producing confident failures later.¹

To build a durable memory trace, the learner must generate the word (or its meaning) from minimal cues, not pick among labeled buttons.

The Working Memory Problem: Why 5 to 10 Seconds Matters

Active recall beats multiple choice. The next question is when to test.

Show a word, hide it, and ask the user to type it 1 to 2 seconds later, and many will succeed every time. That success is misleading. The brain is not pulling from long-term storage; it is re-reading the sensory echo still in the phonological or visual working memory buffer.

Classic Brown-Peterson style tasks show that unattended verbal items decay within roughly 10 to 18 seconds when rehearsal is blocked.² In flashcard research, Karpicke and Roediger (2007) found that essentially immediate ("0-second") tests contribute little to long-term retention compared with tests after a short delay.³ Robert Bjork's desirable difficulty principle says learning strengthens when retrieval is genuinely effortful. For a brand-new word, the sweet spot is often when that echo has faded: about 5 to 10 seconds after exposure, roughly the time to complete one short intervening task.

The Solution: The Interleaved Micro-Buffer

You cannot fix this by showing a blank screen for 5 to 10 seconds. Idle time invites maintenance rehearsal ("gato, gato, gato"), which keeps the word in working memory without deep semantic encoding.⁴

LinGoat uses an interleaved micro-buffer: a distractor exercise between initial exposure and the first blind test. The learner's attention shifts to unrelated work. When the original word returns, the short-term cache is empty and retrieval must do real work. Interleaving unrelated items during study also supports vocabulary learning in second-language research.⁵

Here is the pre-SRS onboarding queue before any word enters the main FSRS schedule:

1. Priming (Word A: working memory loaded). The user sees Word A and its translation, then types the word while it is still visible to anchor form, sound, and meaning.
2. Interrupter exercise (working memory evicted). Before Word A is tested blind, the user completes an unrelated task for about 5 to 10 seconds. That may be an SRS review card, a grammar exercise, or another routine item. The goal is to flush Word A from phonological and visual loops.
3. Active recall (Word A: true encoding). The user sees only the translation (or cue) and must type Word A from scratch. With working memory cleared, success reflects retrieval, not echo.

Read-recite-review style strategies show that production steps after exposure improve retention compared with passive re-reading alone.⁶ LinGoat's laddering applies the same logic in a tight, automated loop tuned for vocabulary onboarding.

The Handoff: Seeding the FSRS Algorithm

Scheduling during laddering is intentionally separate from main SRS timing. FSRS should not project month-long intervals for a word that has not yet been encoded. The micro-buffer controls the seconds-and-minutes rhythm of first contact.

But the data is shared. LinGoat records every correct and incorrect attempt inside the buffer. When a word passes active recall and graduates, that full telemetry log initializes FSRS:

• Clean first recall: FSRS can start with a standard "Good" baseline.
• Multiple failures before success: FSRS receives those misses, registers higher intrinsic difficulty, and schedules a tighter first review interval.

Result: fewer frustration loops during onboarding, and honestly calibrated forgetting curves from day one of spaced repetition. Acquisition and maintenance stay in separate lanes; the memory model still gets the full story.

LinGoat builds this laddering into full-sentence practice: you type real Spanish (or your target language), get word- and grammar-level feedback, and each graded piece enters FSRS when it is ready, not when a multiple-choice screen says you are. See how LinGoat works on the homepage or open the app to try the new onboarding flow.

References

1. Roediger, H. L., & Marsh, E. J. (2005). The positive and negative consequences of multiple-choice testing. Journal of Experimental Psychology: Learning, Memory, and Cognition, 31(5), 1155–1159. https://doi.org/10.1037/0278-7393.31.5.1155
2. Peterson, L. R., & Peterson, M. J. (1959). Short-term retention of individual verbal items. Journal of Experimental Psychology, 58(3), 193–198. https://doi.org/10.1037/h0049234
3. Karpicke, J. D., & Roediger, H. L. (2007). Expanding retrieval practice promotes short-term retention, but equally spaced retrieval enhances long-term retention. Journal of Experimental Psychology: Learning, Memory, and Cognition, 33(4), 704–719. https://doi.org/10.1037/0278-7393.33.4.704
4. Bartlett, J. C., & Tulving, E. (1974). Effects of temporal and semantic encoding in immediate recall upon subsequent retrieval. Journal of Verbal Learning and Verbal Behavior, 13(3), 297–309. https://doi.org/10.1016/S0022-5371(74)80066-6
5. Nakata, T., & Suzuki, Y. (2019). Effects of interleaving on second language vocabulary acquisition. Modern Language Journal, 103(1), 1–19. https://doi.org/10.1111/modl.12581
6. McDaniel, M. A., Howard, D. C., & Einstein, G. O. (2009). The read-recite-review study strategy: Effective and portable. Psychological Science, 20(4), 516–522. https://doi.org/10.1111/j.1467-9280.2009.02325.x