Memorizing Vocabulary

The core problem with vocabulary learning

Vocabulary is an arbitrary mapping problem. There is no logical reason that the French word grenouille means frog, or that the Japanese word taberu means to eat. You need to store thousands of these form-to-meaning connections, retrieve them in both directions (hearing the word → retrieving the meaning; wanting to express a concept → retrieving the word), and keep them accessible over years of study.

Rote repetition — writing a word and its translation 20 times, or reading it out of a list until it "sticks" — works in the short term but produces poor retention. The word is accessible that day because it's fresh in working memory; a week later, without reinforcement, the forgetting curve takes it.

The failure mode is: study 50 words on Monday, feel like you know them, and encounter the same words as strangers again on Thursday.

The keyword method

Richard Atkinson (Stanford University, later UC President) developed the keyword method in the mid-1970s. His 1975 paper with Michael Raugh, "An Application of the Mnemonic Keyword Method to the Acquisition of a Russian Vocabulary," published in the Journal of Experimental Psychology, introduced the two-stage procedure:

Stage 1: The acoustic link. Find an English word (the "keyword") that sounds like some part of the foreign word. For caballo (Spanish for horse), the keyword might be "cab" or "cable." For the Russian linkor (battleship), the keyword might be "Lincoln."

Stage 2: The imagery link. Form a vivid mental image of the keyword interacting with the word's meaning. For linkor/battleship: imagine Abraham Lincoln standing on the deck of a battleship, his tall hat blowing in the wind.

When you later encounter linkor, the sound triggers "Lincoln," and the image of Lincoln-on-a-battleship retrieves "battleship."

Atkinson and Raugh's study tested 52 Stanford graduate students learning 120 Russian vocabulary words. The keyword method group significantly outperformed the control group on an immediate recall test (72% vs. 46%). A subsequent experiment found the advantage persisted over days.

In a 1993 synthesis of keyword method research, Loretta Kasper (Foreign Language Annals) reviewed decades of studies across Spanish, French, German, Japanese, and other languages. The keyword method consistently produced better initial recall and better delayed recall than rote and control methods, though the advantage sometimes attenuated at very long delays (months) without intervening review.

Combining keyword encoding with spaced repetition

The keyword method solves the encoding problem. Spaced repetition solves the retention problem. Combined, they address the full vocabulary-learning challenge:

1. Learn the word using the keyword method — build the acoustic and imagery links.
2. Add the word to a spaced repetition deck with the keyword and image noted in the card's extra field.
3. Let the scheduler return the card at progressively longer intervals, targeting 90% recall probability at each review.

Without spaced repetition, even keyword-encoded words follow the forgetting curve. Without keyword encoding, even well-scheduled review is dry rote retrieval. Together, you get the strong initial encoding from imagery plus the long-term durability from timed retrieval.

Creating effective vocabulary cards

For foreign-language vocabulary, card format matters:

Both directions. Study word → meaning and meaning → word. You need to recognize the word when you hear it (reading/listening) and produce it when you want to express something (speaking/writing). These are different skills that reinforce each other.

Include pronunciation. In languages with non-obvious pronunciation (French vowel length, Japanese pitch accent, Mandarin tones), include the pronunciation on the card. Neurako's audio card feature can attach a pronunciation recording so you hear the word at each review — an additional encoding channel.

Add the word in a sentence context. A card showing a word in isolation ("taberu → to eat") is weaker than a card showing it in a sentence ("Maiasa taberu: I eat every morning. [taberu = to eat]"). The sentence provides context that aids both encoding and real-world recognition.

Note grammatical features. For languages with grammatical gender (French, Spanish, German), include the article or gender marker as part of the card. "Le chat (m) → cat" encodes gender alongside meaning.

Use audio. The acoustic-link stage of the keyword method works better if you've actually heard the word in your target language, not just read a phonetic transcription. Record yourself, use text-to-speech, or find a native speaker recording.

Beyond flashcards: input-rich immersion

Vocabulary flashcards produce word-level knowledge. Reading, listening, and speaking produce the collocational and contextual knowledge that makes words actually usable in conversation.

A word you know from flashcards alone might let you understand a written sentence but not produce the word naturally when speaking. You know grenouille means frog, but you don't know that French speakers say avoir la grenouille qui picote les amygdales (to have a frog tickle your tonsils) for having a sore throat.

The most efficient vocabulary learners combine targeted flashcard study with input-rich practice:

• Graded readers — books or articles written at a learner's current level, where roughly 95–98% of words are already known. Encountering known words in rich context reinforces them; the remaining 2–5% are learnable from context.
• Parallel texts — Reading a book in your target language alongside its translation. New vocabulary appears in natural sentences, not lists.
• Podcast and audio with transcripts — Encounter vocabulary through sound and text simultaneously.
• Shadowing — Repeating audio while it plays, which forces processing of both sound and meaning simultaneously (a dual-coding approach to pronunciation).

Flashcards and input are complementary. Flashcards systematically cover the vocabulary you need; input provides the contextual depth that makes the words real.

The comprehensible input debate

Stephen Krashen's influential "input hypothesis" (1977–1982) argued that language acquisition happens primarily through comprehensible input — exposure slightly above your current level — and that explicit vocabulary study has limited value. Many researchers and thousands of successful language learners have disagreed in practice.

The current consensus, informed by decades of second-language acquisition research (Nation, 2001; Schmitt, 2010), is closer to a middle path:

• A core vocabulary of high-frequency words (the 2,000–3,000 most common words in a language) benefits from explicit study — these words are important enough to know reliably, and comprehensible input alone can take years to cover them.
• Beyond the core vocabulary, extensive reading and listening are the most efficient acquisition routes, because they provide frequency proportional to natural use.

Neurako is most valuable for the explicit-study phase of vocabulary learning — particularly for those common, high-frequency words that form the foundation of comprehension.

Practical keyword method examples

Japanese: taberu (to eat)

• Acoustic link: "taberu" sounds like "table" + "roo."
• Imagery link: Picture a kangaroo (roo) sitting at a dining table (table), eating a huge meal.
• Review: Hear "taberu" → "table" → kangaroo eating → "to eat."

Spanish: murciélago (bat, the animal)

• Acoustic link: "murciélago" sounds like "Mercy-lago."
• Imagery link: A bat flying over Lake Como (Lago di Como) holding a "Mercy" sign.
• Review: murciélago → "Mercy-lago" → bat over the lake → bat.

German: Handschuh (glove; literally "hand shoe")

• German compound words often self-explain: Hand + Schuh (shoe). Image: a shoe worn on your hand.
• This is elaborative encoding rather than keyword method — the word's etymology carries the image.

Note: The keyword method is most useful when no transparent etymology or obvious connection exists. When a word does have a transparent connection to something you know, use that connection directly — it's already an imagery link.

Related journal guide

For a more opinionated product comparison or narrative walkthrough, read Language learning with spaced repetition on the Neurako Journal.

Related reading

Spaced Repetition Explained

How spaced repetition works, why timing matters, and how adaptive schedulers make reviews more efficient.

Elaborative Encoding

Craik and Lockhart's levels-of-processing framework: why connecting new information to what you already know produces deeper, more durable memory than surface processing.

Language Learning

How to use spaced repetition for vocabulary, grammar patterns, script learning, and audio in language study.