There is a particular kind of patience a parent learns at a dinner table set for someone who isn’t there. A chair pulled out for a guest only the four-year-old can see. A seatbelt fastened, with great seriousness, around an empty stretch of back seat. A bath towel held ready for a friend who never drips. Plenty of families have done some version of this, half-charmed and half-uneasy, wondering what exactly it means that their child has conjured a companion out of nothing and folded it into the household routine.
The unease usually has a shape to it. Somewhere in the back of the mind sits an old picture of the child with an imaginary friend: solitary, a little fragile, short on real playmates, perhaps signaling a problem that the invisible friend has been recruited to solve. It is a sympathetic worry. It also turns out to run almost exactly backwards from what the research describes.
An old stereotype, and where it came from
The stereotype, as the psychologist Marjorie Taylor and her colleagues have described it, is of a shy, withdrawn child with some emotional difficulty who needs to go out and make some real friends. The image isn’t conjured from nothing. Children do sometimes lean on imagination to cope — a child can walk past a frightening dog more bravely with an invisible tiger at their side, or tell a pretend friend the secrets they can’t say aloud. The imagination is a genuine resource, and it is occasionally pressed into hard service.
But coping is not the usual reason a companion appears. Across decades of interviews with hundreds of children and parents, Taylor’s group at the University of Oregon and other researchers have found that pretending to have a companion happens far more often simply because it is enjoyable than because a child is in any distress. And when they have measured the children who do it against the children who don’t, the differences tend to lean the encouraging way.
What the research actually found
Children who create imaginary companions tend, as a group, to be sociable rather than solitary — less shy, more outgoing, and often quicker to grasp how other people think and feel. The associations are modest and they are correlations, not a guarantee about any one child, but they point consistently away from the lonely-loner picture. A child rehearsing a whole second personality, giving it opinions and moods and a voice of its own, is doing something a lot like practicing other minds.
The companions themselves are not always the obliging imaginary servants adults might expect, either. Taylor’s interviews are full of pretend friends who are bossy, argumentative, and inconveniently independent — who show up late, refuse to play the assigned game, or, in one child’s complaint, hit her on the head and put yogurt in her hair. A friend who won’t simply do as it is told is a friend a child has to negotiate with. That is social practice, not social retreat.
It is also a strikingly common thing to do. By one accounting that includes companions of every kind, roughly sixty-five percent of children up to the age of seven have had one at some point; narrow the definition to invisible friends alone and the figure settles closer to thirty-seven percent. A behavior this widespread is poorly described as a warning sign.
They know it isn’t real
The other half of the old worry — that a child who talks daily to someone invisible has lost track of what is real — fares no better under scrutiny. In an early study, Taylor and her colleagues interviewed twelve four-year-olds with imaginary companions and fifteen without, and found no difference between the two groups in their ability to tell pretend things from real ones. The children with vivid invisible friends were exactly as clear-eyed about the cookie-that-isn’t-a-cookie as everyone else.
When researchers looked specifically for confusion, they mostly failed to find it. In one sample of eighty-six children with invisible friends, eighty-three — about ninety-seven percent — gave no sign of mistaking the companion for a real person. The children volunteer this themselves, often in the same breath as an elaborate description: a house ghost named George who is, the child will tell you with a small smile, “just pretend.” Asked where a friend goes when she’s away, one child explained simply, “I pretend they’re real but they’re not.”
There is even a quiet comedy in how little parents tend to know about all this. By one early estimate, roughly a third of the parents of children with imaginary companions have no idea the companion exists. The friend at the dinner table is, often enough, a guest the grown-ups never properly met.
This is one body of research, not a verdict
None of this means an imaginary companion is always a sign of flourishing, and it would be a distortion to swing from one tidy story to its opposite. The findings on sociability and social understanding are associations drawn largely from interviews and questionnaires, and much of this work has been done with primarily Euro-American, middle-class families — a real limit on how far the numbers should be stretched. Researchers themselves are careful to say that some children genuinely do use companions to manage fear, loss, or loneliness, and that the same imaginative gift can serve a happy child and a struggling one alike.
What the evidence does is dismantle the assumption that the invisible friend is, by default, a problem. It does not replace that assumption with a promise that it is always a gift. The honest summary is narrower and more useful: across large numbers of ordinary children, inventing a companion travels with sociability and a sturdy grip on reality far more often than with their absence.
What this can and cannot do
We write here as parents and as readers of this research, not as clinicians, and an essay is a reading of studies rather than an assessment of any particular child. What the work can offer a worried parent is mostly permission — to watch an invisible friend arrive without reading it as a symptom, and to play along without fear of feeding a delusion the child doesn’t actually hold.
What it cannot do is diagnose. An imaginary companion sits alongside the rest of a child’s life, and it is that whole picture that matters. If a child seems genuinely withdrawn from real relationships, persistently distressed, or unable to set the pretend aside when it counts, those are reasons to talk with a pediatrician or a child psychologist — not because of the imaginary friend, but because a child’s distress deserves attention wherever it shows up. For most children, the companion is simply one of the more vivid rooms in an ordinary imagination.
So the next time a small voice asks you to wait while a friend climbs into the car, it is worth remembering which way the evidence runs. The invisible companion at the table is far less often a sign of the company a child cannot find than a sign of the company a child already knows how to keep.
