A short guide to the research behind this game: what "receptive bilingualism" is, why it is not a failure (yours or theirs), and the surprisingly specific things that turn understanding into speaking.
A child who understands a home language well but doesn't speak it is called a receptive bilingual. Among families raising kids in one language while the world outside speaks another, this is one of the most common outcomes there is — documented across Japanese-, Spanish-, German- and Korean-speaking families all over the world.
Here is the important part: your child is not refusing, and hasn't "failed to learn." The comprehension is real language knowledge — years of it. What's missing is only the production habit. Those are different skills, and research on heritage speakers (Silvina Montrul and others) consistently finds comprehension running far ahead of speaking. The gap is normal. It is also very fixable.
Researchers who study families like ours (Elizabeth Lanza, Susanne Döpke, Janice Nakamura) found that receptive bilingualism is usually created by a completely loving, completely natural habit they call the "move-on" pattern:
Nothing broke. Dinner got served. And that's exactly the problem: the conversation succeeds without Kannada ever leaving the child's mouth, so from the child's side, speaking Kannada has no job to do. Repeat this a few thousand times and you get a child with excellent ears and a quiet mouth.
The flip side of the same research: families whose children do actively speak the home language respond differently in that exact moment — gently, but in ways that make Kannada momentarily necessary. That moment, times a thousand, is the whole game. Section 5 gives you the exact moves.
UCLA ran a famous series of studies on "overhearers" — adults who had grown up hearing a language at home without speaking it. When they finally started producing it, their pronunciation was measurably near-native, unlike classroom learners who started from zero. All those years of listening quietly banked the sound system of the language.
Those early "wow, I sound like I know this" moments matter enormously, because the same literature is clear that what usually keeps heritage kids silent isn't laziness — it's self-consciousness about speaking "badly" in front of fluent adults. A child who discovers their accent is genuinely good loses the main reason to stay quiet.
A natural instinct is "they just need to hear more Kannada." The evidence says otherwise. Canada ran the experiment for us: children in French immersion received years of comprehensible French input, understood everything — and still spoke poorly. Merrill Swain's conclusion, now a cornerstone of the field, is that listening builds understanding, but only speaking builds speaking. The mouth needs its own reps.
Case studies of receptive bilinguals who suddenly became speakers (Nakamura's work) all share one trigger — not an app, not flashcards: the language became genuinely necessary with a real person. One child began speaking Bulgarian after ten days visiting family in Bulgaria. A monolingual grandparent, a cousin, a trip — necessity plus a real audience converts receptive kids remarkably fast.
Memory research adds a nice bonus: the "production effect" — words a child says out loud are remembered dramatically better than words merely heard or tapped on a screen, and the effect is strongest in children.
These are the documented "discourse strategies" from the family-bilingualism research, ordered from gentlest to strongest. They work in the exact moment your child answers you in English. Pick one and use it for a week — consistency beats intensity.
Respond as if you didn't quite catch the English. Warm face, real curiosity, then wait. The pause is the tool — let them reach for the Kannada word.
ಏನು? ಗೊತ್ತಾಗ್ಲಿಲ್ಲ… (“What? I didn't get that…”)Guess what they meant, in Kannada, as a yes/no question. This hands them the exact words they were missing, and all they have to do is confirm — the lowest-effort Kannada sentence there is.
ನಿಂಗೆ ನೀರು ಬೇಕಾ? (“You want water?”) — now “ಹೌದು, ನೀರು ಬೇಕು” is easy.Say their English sentence back in Kannada, naturally, then leave a beat of silence before continuing. No "repeat after me" — just model it and give the silence a chance to pull an echo out of them.
Ask questions that contain their answer inside them. Instead of an open question, offer two choices in Kannada — the child can answer with a real Kannada word they just heard, without having to retrieve it cold. Perfect for younger kids.
ದೋಸೆ ಬೇಕಾ, ಇಡ್ಲಿ ಬೇಕಾ? (“Dosa or idli?”)Give the frame and let them finish it. Sentence-starters remove the scariest part — beginning.
ನನಗೆ ಬೇಕು… ? (“I want… ?” — they add the last word.)And the guardrails — the same research is equally clear about what backfires:
Honestly? The game is the practice court, not the match. ಹೇಳು-ಕೇಳು is built directly on the strongest finding above: it makes Kannada momentarily necessary between the two brothers. The speaker only scores if his Kannada actually carries the meaning across to his brother — real communication, tiny stakes, and the points always go to the team, never against each other.
That's also why every round ends in the mouth, not the finger: the speaker must say the word out loud before anything else happens, and the native audio replays afterwards so the sound settles in. When they play, resist the urge to coach mid-round — let the brothers repair misunderstandings themselves in whatever Kannada they can find. That repair is the exercise.
Beyond the two-player game, the home screen now has daily missions (one real-world Kannada speech act a day, you give the star), the parrot echo (record-and-compare shadowing), ajji-call prep (rehearse five phrases before the weekly call — a grandparent is the most "necessary" Kannada audience there is), and a story with listen-and-tap questions.