06/23/2015 10:55 am ET Updated Dec 06, 2017

The Dreams of Little Children: Hellishly Boring, or Scientifically Valuable?


I don’t go out of my way to hear about other people’s dreams. I don’t find them particularly meaningful, nor meaningfully different than my own subconscious’ crackpot stories. Dreams are, by their very nature, surreal. But that doesn’t make them interesting.

In taking an anti-dreams stance, however, I occasionally overlook their value in considering congition from a scientific perspective. When dream analysis sheds light on biological unknowns, I’m more than willing to jump in. Case in point, a topic that initially made my eyes glaze over: little kids’ dreams.

As painful as it sounds, surveying the dreams of waist-high humans can actually contribute to our understanding of overlapping mental processes. How preschoolers dream, and subsequently relay their dreams, speaks to their emotional maturation, capacity for abstract thought and grasp on reality.

We don’t entirely understand how younger children dream. For a long time, prevailing wisdom said that youngsters mainly dream in static images; and those images were of animals rather than other humans. Seldom did the children incorporate themselves into their dreams in any active or self-referential way.

Over the years, several studies have challenged this model, offering instead a more complex view of the kiddie subconscious.

The divergent theories speak to the difficulty of conducting children’s dream research, according to a recent study published in Frontiers of Psychology. Basically, kids make for easily agitated, unreliable research subjects. For the study in question, a team of behavioral scientists from Hungary, Germany and the U.S. sought to avoid the pitfalls of previous research; they studied children in their natural environments without sacrificing scientific soundness.

The verdict? We owe kids some credit. Even four-year-old children, who can’t necessarily distinguish between dreams and waking fantasies, dream more like adults than previously thought.

Dreams of the littlest humans

In the past, researchers used two basic set-ups for pediatric dream studies. In the first, children spent a night sleeping in a lab, where researchers monitored their sleep patterns through EEG, waiting for the high levels of brain activity indicative of REM-stage sleep. This is when most vivid dreaming occurs. When REM activity died down, researchers woke children up and asked about their dreams.

Immediate awakening allows for the shortest lag-time between dreaming and re-telling of dreams, which increases the likelihood of thorough, accurate reporting. Waiting until children rise in the morning allows for the possibility of kids splicing their dreams with waking fantasies. Kids have a thorny relationship with the truth — they can be awkwardly honest, but they also lie indiscriminately. The less wiggle room they get, the more plainly factual their accounts can be.

The problem is, the lab is an artificial, unfamiliar environment. This may affect how children sleep and dream in the first place. For that reason, other researchers have opted to study dreaming children in their natural habitat.

In these studies, kids sleep at home and report on their dreams the following morning, either to their parents or to teachers at school. Unlike neutral research assistants, parents and teachers have a habit of swaying children’s answers, through asking leading questions and filling in recollection gaps. At-home studies may be more natural, but they’re less controlled.

Lab and home studies depicted children’s dreams somewhat differently. In seminal lab work performed by David Foulkes in 1970, typical dreams of pre-schoolers were “I was sleeping in the bathtub” and “A fish in a bowl on the riverside.” The dreams lacked motion, emotion, active self-characters, clear narratives and interactions between multiple characters.

In the at-home studies, on the other hand, dream reports contained emotional and cognitive language, descriptions of social interactions and complex narratives. From what we know, “real” adult dreams begin between ages nine and 11. But the at-home research suggested that children began dreaming similarly to adults before they even stepped foot in elementary school.

In the lab, Foulkes found eight percent of three- to five-year-olds’ dreams to contain emotion. For at-home studies, the figure was around 75 percent.

Controlled processes, home environment

In the current study, the Hungarian-lead research team set out to create a controlled experiment in which children could still sleep at home, rather than endure abrupt mid-slumber awakenings in science labs.

They recruited 40 Hungarian children, who ranged in age from four- to eight-and-a-half, to participate in a six-week study. All children came from middle-class, educated households in which at least one parent held a higher degree. Researchers trained parents to conduct tape-recorded dream interviews within 20 minutes of their children waking up. Parents also rated dream reports on a scale of one to 10, based on the likelihood that kids were describing waking fantasies rather than dreams.

Researchers independently rated dreams on the same scale. In cases where the ratings strongly diverged, or where parents didn’t follow protocol, researchers excluded dreams from their analysis.

In the end, researchers picked apart 349 dreams (an average of eight or nine per child). The resulting dreams weren’t quite as dramatic and narratively complex as reported in previous at-home studies. Still, researchers concluded that, “even at a preschool age, the level of most measures of dreams are significantly closer to adult standards than the laboratory based approach had concluded.”

Among four- to eight-year-olds, dreams became more sophisticated as kids aged, but even the youngest children described non-static, highly eventful dreams.

In the study, they wrote:

“This leads us to the conclusion that children's cognitive architecture is already functional in spontaneously constructing motion imagery. This latter phenomenon contradicts Foulkes' speculations who found young children's dreams to be static and hypothesized that preschoolers are unable to imagine motion in space.”


  • The average four-year-old’s dream contains multiple characters, who are more often human than animal. The average number of characters-per-dream, 3.13, didn’t change as kids got older. In fact, that’s comparable to adults’ dreams.
  • Eighty-six percent of all dreams contained motion.
  • Ninety percent of dreams involved a self-initiated activity, in which a character acted with intention.
  • Children themselves appeared as active characters in more than 75 percent of their dreams, meaning they reported personal involvement in the narrative — e.g., “We went to the city park, papa and you [mom] and Lili and Bende [siblings] we went for a walk and we arrived at a garage…”
  • More than half of all dreams included social interactions between two characters. As children got older, the interactions became more aggressive and less friendly.
  • Cognitive verbs appeared in 28 percent of dreams, and increased significantly between the youngest and oldest children in the study. For example, a five-year-old girl dreamt, “A bad person came into our house … and she pretended to be our mother … and we really thought she was the real mother…”
  • Across the board, girls dreamt more about females and boys more about males. Among girls, the frequency of recalling dreams increased with age, whereas the opposite was true of boys. So, among the oldest kids, girls recalled their dreams almost twice as often as boys did.

Some parents sit in rapt attention when their children recount their dreams. Others (like me, I imagine, eventually) would rather chew on nails. But maybe those of us in the second camp shouldn’t be so quick to dismiss. The subconscious of a child may actually be a useful place to spend some time.

-- Theresa Fisher

Read more at Van Winkle's