You do not "study" English.
You acquire it like a first language.

Human language began as sound; writing was invented only about 5,000 years ago. The brain is biologically tuned for speech. Babies go listening → babbling → words → sentences, and learn to read years later as a separate skill. That is why people who cannot read still speak their mother tongue perfectly.

→ Huston: Teaching reading, writing and grammar first runs against how a first language forms. Huston puts listening and speaking first, reading and writing later.

Eight-month-old babies find word boundaries on their own, using nothing but the transition probabilities between syllables. Nobody explains grammar to them. Given enough speech, the brain extracts regularities automatically.

→ Huston: Instead of memorising rules, we expose the same pattern about 40 times inside meaningful context (TPRS) so the brain extracts the rule itself.

Between 6 and 12 months, the ability to tell apart contrasts absent from the mother tongue (such as R/L) fades — "perceptual narrowing". That is the root reason Korean adults cannot hear English sounds like R/L, F/V and Θ.

→ Huston: We retrain that perception with large amounts of accurate native audio and meaningful repetition. Retuning the ear comes before correcting the mouth.

Children build language out of concrete usage events — real conversations with a caregiver — not abstract rules. Joint attention and shared intention are the key. Television watched alone does not grow language.

→ Huston: The AI tutor genuinely responds to what the learner says and asks back, turning one-way exposure into two-way interaction.

A child hears tens of millions of words before becoming fluent in a first language. A word is acquired only after being met 6–20 times in varied contexts (Nation). Sheer volume of exposure is decisive.

→ Huston: Two or three classes a week can never supply that volume. Ten minutes of AI conversation at home every day multiplies exposure — the one decisive advantage of online.

A baby listens for more than a year before speech arrives. Once enough comprehensible input has accumulated, output appears by itself. Forcing output early only raises anxiety.

→ Huston: Early on we never force output; we concentrate on comprehensible input and invite speech once the learner is ready.