{
  "schema_version": 1,
  "id": "vocabulary-pretesting-guess-feedback-boundary-2026",
  "canonical_url": "https://brali-lifeos.github.io/evidence/vocabulary-pretesting-guess-feedback-boundary-2026/",
  "json_url": "https://brali-lifeos.github.io/evidence/vocabulary-pretesting-guess-feedback-boundary-2026/index.json",
  "decision": "propose-protocol",
  "decision_label": "Protocol candidate",
  "reviewed_at": "2026-09-02",
  "reviewed_by": "Brali Evidence Reviewer",
  "source": {
    "title": "Duolingo-inspired pretesting with words and pictures improves vocabulary learning",
    "url": "https://link.springer.com/article/10.1186/s41235-026-00708-y",
    "type": "primary-study",
    "doi": "10.1186/s41235-026-00708-y",
    "citation_text": "Chua TJE, Pan SC. Cognitive Research: Principles and Implications. 2026;11:20.",
    "study_design": "Four preregistered controlled online experiments comparing within-person pretesting versus reading for Spanish vocabulary. Experiments 1 and 3 used intermixed criterial tests; Experiments 2 and 4 additionally randomized participants to blocked versus intermixed test arrangements. Items and condition assignments were randomized and counterbalanced.",
    "population": "Adult Prolific participants aged 21–45, fluent in English and resident in English-speaking countries. Final samples were 58 in Experiment 1, 123 in Experiment 2, 47 in Experiment 3 and 113 in Experiment 4 after exclusions for prior Spanish knowledge, comprehension failures, non-completion or off-task behavior.",
    "intervention_or_exposure": "For word-image learning, learners saw a Spanish word and chose its meaning from four images; for image-word learning, they saw an image and chose the Spanish word. They had up to 8 seconds to guess and then 5 seconds of correct-answer feedback. Reading trials showed the correct pair directly for 5 seconds. Memory was tested after five one-minute distractor tasks using cued recall and multiple choice.",
    "outcomes": [
      "Short-delay cued recall of word meanings or Spanish words",
      "Short-delay multiple-choice recognition",
      "Learner preference and metacognitive postdictions"
    ]
  },
  "supported_claim": "For adults learning new, concrete second-language vocabulary paired with images, adding a forced multiple-choice guess before immediately revealing the correct pairing can produce a modest short-delay memory advantage over simply reading the pair. A bounded Brali protocol may therefore use a cue → guess → immediate correct feedback → later recall sequence for vocabulary practice, while keeping the claim limited to this learning format and time horizon.",
  "unsupported_or_overstated_claims": [
    "Making errors before learning is generally better than studying correct information.",
    "Pretesting improves grammar, reading comprehension, conversation skill, pronunciation or complex skill learning.",
    "The observed advantage has been shown to persist over days, weeks or months.",
    "The exact 8-second guess window, 5-second feedback window, four-option format or image format is optimal.",
    "Multiple-choice pretesting always improves later multiple-choice performance; Experiment 3 did not show a significant multiple-choice advantage.",
    "The effect proves that error correction, associative strengthening or any single proposed mechanism caused the benefit.",
    "The findings generalize to children, advanced language learners, non-concrete vocabulary or learners outside the sampled online populations."
  ],
  "limitations": [
    "All criterial tests followed a short same-session distractor period rather than a delayed retention interval, so long-term retention was not established.",
    "The studies used concrete Spanish nouns paired with images; vocabulary type and language-learning context were narrow.",
    "Participants were adult Prolific users from English-speaking countries and exclusions were substantial in several experiments.",
    "Pretesting trials included up to 8 seconds of cue-only guessing before the same 5 seconds of correct pair exposure used in reading, so total cue exposure was longer in the pretesting condition.",
    "Initial guess accuracy was about 35–38%, raising the possibility of some prior familiarity despite exclusions; the authors note that benefits also appeared for incorrectly guessed items.",
    "Learning condition was primarily within-subjects, and the authors identify between-subject replication as a future research need.",
    "Multiple-choice benefits were not significant in Experiment 3, so recognition effects were less uniform than cued-recall effects."
  ],
  "target_hack_ids": [
    "guess-before-reveal-vocabulary"
  ],
  "target_protocol_ids": [
    "brali:guess-before-reveal-vocabulary"
  ],
  "risk_flags": [],
  "notes": "Protocol-design direction: use this only when introducing unfamiliar concrete L2 vocabulary. Show a cue, require a real guess before the answer is visible, reveal the correct answer immediately, then test unaided recall later in the session. Do not reward confident guessing as correctness, and do not replace spaced later review with this one-pass encoding technique."
}