You're sitting on the floor. Your baby picks up a wooden block, turns it, drops it, picks it up again. You watch. You wait. Then — without a word — she looks at you. Not for help. Not because something happened. She looks at you to show you the block.
That glance is one of the most sophisticated cognitive acts your baby performs. Developmental scientists call it joint attention — the ability to share focus on an object with another person, to coordinate your mental spotlight with theirs. By 13 months, your baby has been doing it for months. And it's changing how her brain works in ways you can't see.
Here's the part that might surprise you: the way you respond to that glance — not what you teach, not what you direct, but how you follow — is literally shaping your baby's capacity to concentrate alone.
The Assumption We Get Wrong
Most of us carry an unexamined assumption about attention: that it's something we teach children by directing them. Look at this. Watch me. Focus here. Parenting advice reinforces this — stimulate your baby, point out interesting things, redirect their wandering gaze to something worthwhile.
But a series of studies published in the last year — including two randomized controlled trials — reveals something closer to the opposite.
"Shared attention develops through interactive but asymmetric, infant-led processes."
— Phillips, Goupil & Wass, eLife, 2025
That quote comes from a study that may rewrite how we think about infant attention. Phillips and colleagues put EEG caps on 66 twelve-month-olds and their caregivers during naturalistic play — not a controlled lab task, but actual messy, block-dropping, babbling floor time. They measured moment-by-moment attention in both partners simultaneously.
What they found: the baby's own internal cognitive rhythms — endogenous frontocentral theta oscillations, the brain's self-generated attention pulses — were the primary driver of sustained attention. Not the caregiver's gaze. Not the caregiver's voice. The baby's own neural clock.
And the caregiver? The caregiver predominantly reacted to the baby's attention shifts rather than directing them. Infant attention was not forward-predicted by anything the caregiver did. The baby was leading. The caregiver was following.
Following In
Developmental researchers have a term for this: following in. It means joining the baby's existing focus rather than redirecting it elsewhere. When your baby stares at the spoon, you say "spoon." When they bang the cup, you say "bang bang!" When they gaze at the dog, you gaze at the dog.
It sounds passive. It isn't. It's the most cognitively demanding thing you can do — tracking another mind's focus in real time, matching your response to their moment of maximum interest, then adding just enough language to anchor the experience without disrupting it.
And the evidence that it works is now causal, not just correlational.
The Two RCTs
Salter and colleagues randomly assigned 125 families with 6-month-olds to either a communication-focused intervention — short videos teaching parents to follow their baby's attention and label what the baby was already looking at — or an active control group. By 12 months, trained caregivers gave 40% more semantically contingent responses (following in and labeling), and their infants produced significantly more pre-linguistic communicative acts. The effect was stronger for families without degree-holding caregivers, suggesting this isn't about enrichment — it's about unlocking a natural responsiveness that some social environments suppress.
In a separate trial, Kaletsch and Liszkowski trained 55 caregivers in responsive pointing — following the baby's attention rather than redirecting it. The result was clean: responsive pointing (following the baby) correlated with infant pointing frequency. Initiative pointing (redirecting the baby to new objects) did not. Following in worked. Redirecting didn't.
| Caregiver Strategy | Outcome |
|---|---|
| Following in — labeling what baby already attends to | More communication, more pointing, sustained attention |
| Redirecting — pointing baby toward new objects | No effect on infant attention or communication |
The Bridge to Solo Play
Here's where the story converges on something surprising.
Uzundag and colleagues studied 98 mother-infant pairs (mean age 11.8 months) during two kinds of play: together, and alone. They measured sustained attention in both settings. The finding: joint attention during mother-infant play predicted sustained attention during solo play, even after controlling for age and socioeconomic status.
Read that again. How you play together shapes how your baby plays alone.
This isn't a small effect, and it's not a proxy for general intelligence or temperament. The specific pathway runs through shared attention — those moments when both of you are focused on the same thing, and you're responding to what your baby notices. That responsive loop builds something inside the baby's attention system that persists when you're no longer there.
Theyer and colleagues found the same thing using brain imaging. In a hyperscanning study with 181 infants and caregivers wearing fNIRS caps, longer episodes of joint attention predicted longer continued attention after the joint episode ended. And here's the twist: the caregivers who helped most were those who reduced their own left superior temporal gyrus activation during joint attention. In plain language: the caregivers who dialed back their own cognitive effort — who stopped trying so hard to direct the interaction — produced infants with better attention and visual short-term memory.
The parents who helped most weren't the ones directing more. They were the ones directing less — quieting their own cognitive machinery and letting the baby's attention lead.
What the Baby Is Actually Doing
If the baby is leading, what exactly are they doing?
More than we realized. Salter and Carpenter tracked 25 infants across 124 sessions and found that 44% of babies produce joint attention bids by 6 months, and 92% do so before 9 months. By the time your baby is 13 months old, they've been initiating shared attention for half a year or more. They're not learning to do it. They're refining how they do it.
And the quality of that initiation matters independently. Capelli and colleagues followed 46 infants from 12 to 24 months and found that some babies are more distractible than others — they orient to irrelevant stimuli, losing focus on what they were exploring. Distractibility at 12 months predicted weaker language composition at 24 months. But babies who actively initiated joint attention — who looked at a parent and bid for shared focus — were buffered against that distractibility. Initiation, not just participation, was protective.
Perapoch Amadó's large-scale analysis confirmed the developmental arc: by 15 months, infant-initiated joint attention increases dramatically. Joint attention isn't top-down instruction from parent to child. It's a self-organizing system that emerges from multimodal sensorimotor coupling — two people exploring the same world, with the smaller one increasingly taking the lead.
The Whole Body Follows
Joint attention isn't just about eyes meeting over a toy. It's physical.
Kretch found something unexpected about body position: when babies sat facing away from their parent (parent sitting behind), coordinated attention actually improved — because both partners tracked hand movements instead of gaze. You don't need to be face-to-face to share attention. You need to be tracking the same thing, and hands are a powerful attention signal that works from any angle.
Murillo's longitudinal study of 9-to-12-month-olds showed that caregivers naturally incorporate touch into these moments — adult-initiated touch accompanied speech 85% of the time during object-focused interaction. But here's the elegant part: touch decreased as word comprehension increased. Caregivers naturally titrate their scaffolding. They add more physical grounding when the baby needs it and pull back as the baby's language catches up. No one teaches parents to do this. The dyad self-calibrates.
And the objects matter too. Wijeakumar's study of 181 infants found that multi-component objects — things that stack, sort, or combine — drew more scaffolding from caregivers and predicted better visual working memory. A nesting cup teaches more than a rattle, not because it's more educational, but because its complexity invites shared exploration.
What This Means at 5 AM
If you're reading this in the early morning — and some of you are, because the babies born in February 2025 are 13 months old now and some of them have opinions about sleep — here's what all of this comes down to.
You don't need to teach your baby to pay attention. They already have an internal attention clock, a neural rhythm that drives their focus in pulses. What you can do — what the evidence says actually works — is follow that rhythm.
When they look at the block, look at the block. When they hand you the cup, take the cup. When they gaze at the cat, say "cat." When they move on, move on with them. Don't redirect. Don't improve. Don't enrich. Just follow.
That following builds a loop. The baby notices something. You notice them noticing. You name it. They hear the name at the moment of maximum interest. Their brain binds the word to the object to the shared experience of attending together. And over thousands of these micro-episodes — each one lasting a few seconds, most of them unremembered — something structural changes. The baby's capacity for sustained solo attention grows. Not because you trained it. Because you mirrored it.
What to try
- Follow the eyes. When your baby looks at something, look at it too. Name it if the moment feels right. Don't quiz — just label.
- Resist redirecting. The urge to say "look at THIS" is strong. But following what they already notice works better than pointing them somewhere new.
- Offer complex toys. Stacking cups, nesting blocks, sorting shapes — objects with multiple components invite shared exploration more than single-action toys.
- Try side-by-side. You don't need to be face-to-face. Sitting behind or beside your baby, both of you handling the same objects, can produce beautiful coordinated attention through hand-tracking.
- Let them end it. When they move on to the next thing, follow them there. The baby's attention rhythm — roughly 1-2 second pulses — is faster than yours. Trust it.
When to talk to your pediatrician
If your baby rarely makes eye contact, doesn't follow your pointing, or hasn't started sharing attention with you by looking between objects and your face by 14-15 months, mention it at your next visit. Delayed joint attention can be one early indicator that warrants a closer look — not a diagnosis, but a conversation worth having.
Playing Together, Playing Alone
There's a quiet irony here. We worry about babies who can't play independently. We want them to focus, to entertain themselves, to develop that internal capacity for sustained concentration. And the instinct is to train it — to teach them to sit and focus.
But the research says the path to solo attention runs through shared attention. The path to independence runs through responsiveness. The baby who plays longest alone is the one whose caregiver followed most closely when they were together.
It connects to so much we've explored before. The contingent labeling loop from our language post — now confirmed by a randomized trial. The play bout structure from our cognitive development post — now bridged to joint attention. The predictability principle — now with the specific mechanism: following in.
Your baby's attention is not a problem to be fixed or a skill to be drilled. It's a rhythm to be joined. And the extraordinary thing is that by joining it — by sitting on the floor and watching what they watch and naming what they name — you're building in them the very thing you thought you needed to teach.
Sources: Phillips, Goupil & Wass 2025 (eLife); Salter et al. 2025 (Phil Trans R Soc B); Kaletsch & Liszkowski 2025 (J Cognition & Development); Uzundag et al. 2025 (Infant Behavior & Development); Theyer et al. 2025 (Infancy); Salter & Carpenter 2025 (Infancy); Capelli et al. 2025 (Frontiers in Psychology); Perapoch Amadó et al. 2025 (Child Development); Kretch 2025 (Developmental Psychology); Murillo et al. 2025 (Infancy); Wijeakumar et al. 2026 (Infant Behavior & Development); Wass et al. 2025 (Scientific Reports).