Somewhere tonight, a grandmother is holding a phone at arm's length, saying her grandson's name in a voice she hasn't used since her own children were small. On the other end, a fourteen-month-old is pressing a sticky palm against the screen, then wandering off to chase the cat. She keeps talking to the ceiling. Later she'll tell a friend it went well, and she won't quite believe it. She'll wonder whether a face on glass can mean anything to someone that young. Does he know who I am? Does any of this count?

If you raise a child far from the people who love them, you probably carry a quieter version of that same worry. You've absorbed the warnings about screens before two. Then you prop up a tablet so your mother can sing to the baby, and the guilt and the gratitude arrive together.

The research has something specific to say about this, and most of it is good news. Whether a screen teaches a toddler anything depends less on the screen than on whether the person on it is responding to the child.

The video deficit: why screens usually lose

Researchers have documented a pattern called the video deficit effect for a couple of decades. Show a toddler how to do something in person and they'll copy it readily. Show them the same demonstration on video and they learn much less. The same goes for words. Under roughly two and a half, children have a hard time carrying what they see on a screen over into the three-dimensional world.

There are a few reasons. Young toddlers are still working out that a picture of a thing stands for the thing. Screens are flat, small, and missing the depth and touch cues children use to make sense of objects. But one explanation has held up especially well, and it's the one that matters for video chat. A recording doesn't respond. It can't notice that your child looked away, or pointed, or babbled a question, and it carries on regardless.

For a toddler, language is a back-and-forth. A word lands best at the moment the child is already attending, already curious, already reaching. A prerecorded voice keeps talking whatever the child does. A grandmother on FaceTime can stop and say, "Oh, you've got your truck!"

What happened when researchers put toddlers on Skype

In a study published in Child Development in 2014, Sarah Roseberry, Kathy Hirsh-Pasek, and Roberta Golinkoff taught two-year-olds (24 to 30 months) new verbs in three ways: from a live person in the room, from a live person over video chat, and from a prerecorded video of a person. The recorded video was carefully matched to the video chat, so the children saw the same kind of footage. The only difference was whether the adult could respond to them in real time.

Children learned the new verbs both in person and over video chat. From the recording, which couldn't react to them, they didn't. The authors described the key ingredient as social contingency: the adult's responses were timed to what the child was doing.

A few years later, Lauren Myers, Rachel LeWitt, Renee Gallo, and Nicole Maselli tried something closer to ordinary family life. Toddlers between about 12 and 25 months spent a week either FaceTiming with an adult or watching prerecorded videos of the same kind of interaction. The FaceTime toddlers responded to their partner in sync, as though in a real exchange, and a week later they recognized the person they'd been chatting with. The older toddlers in the group showed some signs of learning from those calls too. The evidence there was more modest, but the pattern pointed the same way.

These are a handful of studies, not a final verdict, so it's worth reading them with care. Still, the direction is clear enough that the American Academy of Pediatrics carved video chatting out of its screen-time cautions for the youngest children. A live, responsive call with someone who loves your child is a different kind of thing from a cartoon.

Why it can still feel like it isn't working

Knowing all this, many calls still go the way the grandmother's did: a toddler wandering away and a grown-up talking to the ceiling. That isn't failure. It shows what contingency needs to work.

The benefit depends on the connection behaving like a real exchange. Lag and frozen frames break the timing that makes it work. A reply that arrives two seconds late no longer matches what the child was looking at. Researchers who study screen learning have pointed out that delays like these can quietly undo the advantage of live video.

There's also a quieter ingredient. In the studies, children often did best when a familiar adult was beside them, helping them connect what was on the screen to the room around them. When you sit next to your toddler and say, "Look, Grandpa has a ball too!", you're doing something the screen can't do on its own. You're the bridge.

And a lot of the time the toddler isn't the one who gives up. The adult on the call is. When a child walks off, it's natural to fill the silence with more talk: questions, songs, "Can you say Nana?" But the research suggests the payoff comes from responding to the child, not from performing for them. A grandparent who follows the toddler's lead, even to the cat, is having a better conversation than one who keeps reciting the script.

What makes a call a conversation

Think about what a responsive partner actually does over video, because it's the same thing that works across a kitchen table. They watch what the child is doing. They name the thing the child is already interested in. They pause long enough for a babble, a point, or a shriek. Then they answer it.

This takes some pressure off. A grandparent doesn't need a lesson plan or a stack of flashcards. They need to notice your child, react, and wait. For a toddler, a two-minute call where someone they love says "Shoes! You found your shoes!" at the moment they hold up a shoe is worth more than twenty minutes of being sung at.

It also changes the guilt. A video call with family isn't a moral lapse to make up for later. Done in this spirit, it's one of the few screen experiences that resembles what toddlers actually learn from: being seen, and being answered.

Your next moves

  • Mail one matching object before the next call. Send the grandparent a duplicate of something your toddler loves, like a toy duck, a board book, or a ball. When both of them hold the same thing on camera, the word has a real object behind it on both sides of the screen.
  • Sit beside your toddler, not behind the phone. For at least the first few minutes, be in the frame with them, point at the screen, and repeat what the grandparent says about the room you're in ("Nana sees your cup!"). You're the bridge between the flat image and the real world.
  • Give the grandparent one rule: name it, then wait. Text them before the call: Whatever she holds up, just say its name and count to five before saying anything else. That single pause turns a performance into a turn-taking exchange.
  • Fix the lag before you fix the content. Move closer to the router, close other apps, or call at a quieter time of day. A smooth two-minute call beats a stuttering fifteen-minute one, because the timing is what does the teaching.
  • Let them walk away, and carry the phone along. When your toddler wanders off, take the grandparent with you and let them narrate what they see: "Oh, you're going to the window!" Following the child's interest is the point.

A small thought about Acorn

The idea underneath all of this is that toddlers learn words from being answered: short, warm, back-and-forth moments built around what they're already paying attention to. We built Acorn around that same small rhythm. It's a daily three-minute first-words session meant to be done with your child beside you, with real words for real things and no ads or upsells pulling at anyone's attention. It's not a replacement for Grandpa's voice on a Sunday call, or for yours. If you'd like a gentle daily ritual that leaves you with words to bring into those calls and the rest of the day, you can find it at acorn.lumenlabs.works.