From Scribble to Figure: How Children's Drawing Actually Develops

Lead

Many parents quietly swap out the drawings on the refrigerator every week or two. At first: a tangle of lines that ran off the edge of the paper. Half a year later: loops that closed into circles. Half a year after that: two lines growing from a circle — and the child announced it was "daddy."

Judging those drawings as "good" or "bad" is probably the wrong lens. The path from scribble to human figure is the record of motor control, visual memory, and symbolic function developing in parallel, all of it written out on paper. There is no such thing, in this framework, as "a period when the drawing is poor."

This article draws on foundational research to map the stages children's drawing passes through, and takes a look at how the old idea that you can measure intelligence from a child's drawing fares under modern scrutiny.

Scribbles Contain 20 Basic Shapes

The first systematic account of children's drawing development belongs to Rhoda Kellogg. Between 1948 and 1966, Kellogg collected approximately one million drawings by young children from around the world and analyzed them into 20 basic forms — vertical lines, horizontal lines, diagonals, curves, dots, and others [1]. These elements combine into placement patterns, then radial forms, mandalas, suns, and finally recognizable human figures. The progression is consistent across cultures.

The art educator Viktor Lowenfeld, working in roughly the same period, proposed a six-stage model. The first stage — the scribbling stage, running from about age two to four — he subdivided into three sub-phases: disordered scribbling, controlled scribbling, and named scribbling [2]. Named scribbling is the phase in which the child, having drawn something, gives it a name: "This is Mama." That moment marks the transition from drawing as a motor activity to drawing as representation — the mark on the paper begins to stand for something [2,3].

The figure that appears around age three in children across cultures is the "tadpole figure" — a circle representing the head, with two lines descending directly from it as legs, and no trunk. Maureen Cox's survey of the literature on children's figure drawing notes that the tadpole appears universally across cultures, and interprets it as the child's extraction of an internal model: the head is the center of thought, the legs make the body stand [3,4]. The absence of the trunk is not a failure to observe the body — children this age know perfectly well that trunks exist. It reflects the selective salience of features in the child's internal representation.

Here the crucial interpretive point: these stages are a sequence of drawing modes, not a ladder from worse to better. There is no "clumsy phase." Scribbles have their own completeness; the tadpole figure has its own completeness. Each stage is the full expression of what the child's developing perceptual and motor system can put on paper at that moment.

What Happened to "Drawing Measures Intelligence"

There is a test many parents of young children will have heard mentioned: the Draw-a-Person Test, originally developed by Florence Goodenough in 1926 and revised by Dale B. Harris in 1963 [5]. The child is asked to draw one person; the drawing is then scored on the presence and accuracy of features — head, trunk, limbs, fingers, and so on — and the score is used as an index of cognitive development.

Modern validation research has been substantially less enthusiastic. Kirsten Imuta and colleagues published a systematic analysis of the Draw-a-Person Test's validity in PLoS One in 2013 [6]. The correlation between DAP scores and Wechsler-series IQ tests was approximately 0.27 — a weak relationship. As a screening tool, false-positive and false-negative rates were high enough that the test is unreliable for identifying children with intellectual disabilities or developmental delays [6]. A 2018 study by Aminabadi and colleagues confirmed that the Goodenough-Harris DAP is not a sound substitute for validated developmental screening instruments such as the Ages and Stages Questionnaires [7].

The test has not been abandoned entirely — it still appears in some clinical contexts — but the idea that drawing complexity is a direct window into cognitive capacity has been largely set aside. Drawing complexity is more strongly predicted by fine motor development and by the amount of drawing experience a child has had than by intellectual ability [6]. "My child's drawings are simpler than other children the same age" is, by current evidence, an observation about motor development or practice, not about cognitive potential.

A further observation comes from cognitive scientist Neil Cohn, who has analyzed cross-cultural variation in drawing skill from a linguistic perspective [8]. Japanese children reliably demonstrate more sophisticated drawing ability than their peers in Western countries — but the explanation is not talent. It is exposure. Growing up surrounded by manga means being exposed to a rich visual vocabulary: schematic conventions for eyes, mouths, hair, movement. Drawing, Cohn argues, is a system with a vocabulary and a grammar, as language is — and like language, it develops faster when the environment supplies more input [8]. If a child seems to "have no talent for drawing," the more useful question is whether they have had access to rich visual schemas to imitate.

What Parents Can Do — and Observe

The research points to a relatively uncomplicated set of practical orientations.

The first is to avoid rushing technical instruction. Kellogg's and Lowenfeld's stage accounts both describe a developmental sequence that unfolds over years, driven by internal maturation [1,2]. Intervening early — "here, this is how you draw a circle" — risks introducing noise into a self-directed motor acquisition process. Showing a model is different from guiding the child's hand. The former gives the child something to imitate; the latter substitutes adult control for the child's own motor exploration.

The second is to change the evaluative vocabulary. Instead of "that's nice" or "good job," try "the circle closed this time," or "the figure has two legs now," or simply "tell me about it." This last question is what Lowenfeld's named scribbling stage is about [2]: the child's act of attaching meaning to a mark. Encouraging that process — the connection between physical mark and symbolic intention — is thought to scaffold later narrative drawing and story-description ability. The child who tells you "this is Mama going to the park" is doing something more developmentally rich than drawing a technically accurate figure.

The third is to save the drawings over time. Children's drawing development is continuous, and changes within a single week or month can be invisible. Three months, six months, a year apart — that is the span at which the arc becomes readable: "there's the month the circle closed," "there's the week the trunk appeared." Keeping dated photographs of drawings in an app like Memori makes it possible to read that arc later, without having to save every sheet of paper. And the question of whether to keep or discard a drawing is best answered not by "is it good?" but by "is this a shape only this child could draw right now, at this age?"

Summary

Children's drawing is a site where motor development, symbolic function, and cultural input all unfold simultaneously — and no stage within that unfolding is a failed version of the next. Kellogg's 20 basic forms [1], Lowenfeld's named scribbling [2], Cox's tadpole figure [3,4], the eventual full human figure — each stage prepares the ground for what follows.

The old idea that drawing complexity maps onto cognitive ability has not survived modern empirical scrutiny [6]. What remains is the value of the record itself — of knowing when this particular child began to draw this, at this age, in this form.

Stop scoring the drawings on the refrigerator. Start writing the date.

References

1. Kellogg R. Analyzing Children's Art. Palo Alto, CA: National Press Books; 1969.
2. Lowenfeld V, Brittain WL. Creative and Mental Growth. 8th ed. New York: Macmillan; 1987. (First published 1947.)
3. Cox MV. Children's Drawings of the Human Figure. Hove: Lawrence Erlbaum Associates; 1993.
4. Picard D, Durand K. Are young children's drawings canonically biased? J Exp Child Psychol. 2005;90(1):48–64. doi:10.1016/j.jecp.2004.08.005.
5. Harris DB. Children's Drawings as Measures of Intellectual Maturity: A Revision and Extension of the Goodenough Draw-a-Man Test. New York: Harcourt, Brace & World; 1963.
6. Imuta K, Scarf D, Pharo H, Hayne H. Drawing a close to the use of human figure drawings as a projective measure of intelligence. PLoS One. 2013;8(3):e58991. doi:10.1371/journal.pone.0058991. PMID: 23527063.
7. Aminabadi NA, et al. Goodenough-Harris Draw-a-Man Test as a substitute of Ages and Stages Questionnaires for evaluation of cognition in children. Iran J Pediatr. 2018;28(4):e63946. PMCID: PMC6160634.
8. Cohn N. Framing "I can't draw": the influence of cultural frames on the development of drawing. Cult Psychol. 2014;20(1):102–117. doi:10.1177/1354067X13515936.