You do not need a classroom, a curriculum or an expensive app to help a child pick up a second language. A few minutes of playful practice folded into routines you already have, getting dressed, cooking dinner, bath time, does more for a young child's ear than an hour of formal lessons once a week.
That is not a feel-good claim. It comes out of some of the most-cited research in language acquisition, and the findings are unusually practical: they tell you when to start, how often to practise, and, crucially, which kinds of exposure do almost nothing.
When should we start? Does age really matter?
Earlier is easier, but the window is wider than most parents fear. The largest study ever run on the question analysed 669,498 native and non-native English speakers (Hartshorne, Tenenbaum and Pinker, Cognition, 2018). It found that grammar-learning ability holds up remarkably well through childhood and only begins to decline around age 17.4.
The catch is the runway. Because it takes years of exposure to reach full proficiency, children who start after about age 10 usually run out of window before their learning ability drops, which is why starting by 10 is associated with native-like grammar, and starting at 15 generally is not.
Practical translation: if your child is 3 to 10, you are comfortably inside the window. There is no need to panic, and no need to buy anything. We go deeper into the timing evidence in our article on the best age to start a second language.
Does play actually help kids learn a language?
Yes, and one specific ingredient does the heavy lifting: live social interaction. In a landmark experiment published in PNAS, Patricia Kuhl's team exposed 9-month-old American infants to Mandarin over 12 short sessions. Infants who heard Mandarin from a live person in the room learned to distinguish its phonemes. A second group heard exactly the same material via professionally produced video or audio, and learned nothing measurable at all.
Kuhl called this "social gating": young brains appear to treat responsive interaction as the signal that language is worth encoding. Passive playback, however well produced, does not clear the bar.
The one-line takeaway: what matters is not how much language your child hears, but how much they are responded to. Back-and-forth beats playback, every time.
Harvard's Center on the Developing Child calls the same mechanism "serve and return", the responsive exchanges that build brain architecture:
What activities actually work at home?
The best activities are the ones that fit a routine you already have, rather than extra items on a to-do list. Ten worth trying, with rough time costs:
| Activity | Best for | Time |
|---|---|---|
| Narrate routines: dressing, meals, bath | Everyday vocabulary | 0 min extra |
| Play "Simon Says" in the target language | Listening, verbs | 5 min |
| Label objects around the house with sticky notes | Nouns, reading | 20 min once |
| Sing songs and nursery rhymes | Pronunciation, rhythm | 5 min |
| Read the same picture book in both languages | Comprehension | 10 min |
| Cook together, naming ingredients and steps | Sequencing, verbs | 0 min extra |
| Short spaced flashcard sessions | Retention | 5 min |
| Watch a children's show in the target language | Exposure, culture | 10-15 min |
| Let a talking companion tell stories in the language | Conversation practice | 10 min |
| Video call a relative who speaks the language | Real motivation | 15 min weekly |
Notice that the two screen-based entries sit low on the list. Given the Kuhl findings, treat them as supporting cast, not the main event.
How often should we practise?
Little and often beats long and rare. Research summarised by The Conversation found that distributed practice, meaning short sessions spread across the week, helps children retain vocabulary far better than cramming. Ten to fifteen minutes a day, most days, outperforms one long weekly lesson. We looked at the dose question in detail in Raising a Bilingual Child.
How much exposure do children actually get?
Less than most parents assume, and it varies enormously by country. Only about 20% of US K-12 students study a foreign language, against roughly 92% of pupils in Europe, according to the American Academy of Arts & Sciences. Meanwhile the Migration Policy Institute counts about 10.8 million US children under 9, some 32% of that age group, growing up with a parent who speaks a language other than English at home.
There is a related finding worth knowing: among US-born children of two foreign-born parents, roughly 70% shift to English-only preferences even though 87% grew up hearing another language at home. Heritage languages are lost at home far more often than they are lost in the classroom, which is an argument for small daily habits rather than waiting for school to handle it.
Does it matter if I am not fluent myself?
Less than parents assume. Exposure and consistency matter more than a parent's own fluency, and Connections Academy's guidance recommends leaning on native-speaker audio, shows and books to fill the gaps while keeping practice part of daily life.
The honest difficulty is the Kuhl finding: the resources that fill the gap most easily, video and audio, are exactly the ones that produced no measurable phonetic learning. What distinguishes an effective substitute is whether it responds. A recording talks at a child; a conversation partner waits, answers the actual question, and adjusts.
That gap is the reason Ted was built around real back-and-forth rather than pre-recorded phrases: a child asks something, Ted answers that specific thing, in any of 60+ languages set by a parent in the Ted&Co app. It is not a substitute for a bilingual parent, a grandparent or a teacher, because those relationships carry motivation that no product can manufacture, and if you have access to one, use it first. Where it earns its place is the Tuesday evening when nobody bilingual is available and the alternative is a screen.
Frequently asked questions
What age should I start teaching my child a second language?
The research points to starting by around age 10 for the best shot at native-like grammar, though learning ability itself holds up until about 17.4. For a 3- to 6-year-old, you are well within the window.
Is 15 minutes a day really enough?
Yes, if it is consistent. Short, frequent sessions beat occasional long ones for vocabulary retention, and daily routines you already have can carry most of that time at no extra cost.
Will learning two languages delay my child's speech?
No. Bilingual children may mix languages as toddlers, but this is normal code-switching rather than confusion, and total vocabulary across both languages tracks monolingual peers. See what the research actually shows.
Do language apps and cartoons work for young children?
They help with exposure and motivation but are weak on their own. Kuhl's PNAS study found infants learned foreign phonemes from a live person and not at all from matched video or audio, so interaction appears to be the active ingredient.
How many US children grow up bilingual?
About 10.8 million US children under 9, or 32% of that age group, have at least one parent who speaks a language other than English at home, according to the Migration Policy Institute.