See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/247513687

Children’s Drawings: A mirror to their minds

Article in Educational Psychology · February 2006

DOI: 10.1080/01443410500344167

CITATIONS READS
213 17,622

4 authors, including:

Isabelle D. Cherney
Mount Mary University
49 PUBLICATIONS 1,761 CITATIONS

SEE PROFILE

All content following this page was uploaded by Isabelle D. Cherney on 28 August 2015.

The user has requested enhancement of the downloaded file.

Educational Psychology
Vol 26, No. 1, February 2006, pp. 127–142

Children’s Drawings: A mirror to their

minds
Isabelle D. Cherney*, Claire S. Seiwert, Tara M. Dickey and
Judith D. Flichtbeil
Creighton University, USA
10IsabelleCherney
[email protected]
00000February
Educational
10.1080/01443410500344167
CEDP_A_134399.sgm
0144-3410
Original
Taylor
2006
26 and
&Article
Francis 2006
(print)/1469-5820
Francis
Psychology
Ltd (online)

Children’s drawings are thought to be a mirror of a child’s representational development. Research

suggests that with age children develop more complex and symbolic representational strategies and
reference points become more differentiated by gender. We collected two drawings from 109 5–
13-year-old children (three age groups). Each child drew their family and their school and partici-
pated in an independent recall task. The results indicated significant gender and age differences in
the number of details depicted in the family drawings. There were also significant differences
between boys’ and girls’ stereotyped drawings, usage of proportionality, and clothing. With age,
children tended to draw more aerial views of their school. The results are discussed in terms of the
contribution children’s drawings can make to the study of cognitive development and vice versa, as
well as their importance for education.

Understanding children’s cognitive development has implications for many fields,

and in particular for education. While there are many possible approaches to the
study of cognitive development, assessing a child’s drawings can provide a window
into their representational world. For example, the ability of children to depict
spatial elements from their environment through an understanding of where an
object is located in comparison to others is an important aspect of child development
as well as an important aspect of geography, geometry, and graphic design. The
purpose of the present study is to contribute to our understanding of representa-
tional development by studying two different types of children’s drawings – a family
drawing and a school drawing – and how these representations may correlate with
the children’s (working) memory capacity. Children’s drawings are often used for
cognitive, personality, and diagnostic assessment (Knoff & Prout, 1985; Naglieri,

*Corresponding author. Department of Psychology, Creighton University, Omaha, NE 68178,

USA. Email: [email protected]

ISSN 0144-3410 (print)/ISSN 1469-5820 (online)/06/010127–16

© 2006 Taylor & Francis
DOI 10.1080/01443410500344167
128 I. D. Cherney et al.

1988), and it is therefore important to establish a reliable baseline of their abilities,

particularly with regard to gender- and age-related differences.
Research into children’s drawings has focused on three main areas: (a) the internal
structure and visual realism of children’s depictions (e.g., Cox, 1985, 1992); (b) the
perceptual, cognitive, and motor processes involved in producing a drawing (e.g.,
Freeman, 1980); and (c) the reliability and validity of the interpretation of children’s
drawings (e.g., Hammer, 1997). Very young children produce simple scribbles, and
later demonstrate representational intentions. With maturation and increased
dexterity, children draw things as they are known rather than as they are actually
perceived. Finally, they develop visual realism that includes consideration of
perspective (Tallandini & Valentini, 1991). The present study was designed to
examine possible relationships between the visual realism of children’s graphical
depictions and their underlying cognitive processes. More specifically, we examined
the relationship between children’s representation of spatial and familial relation-
ships and their independent recall.
How objects, thoughts, and events in the world are represented is one of the
fundamental questions in cognitive development. There is more than one way to
represent a thing or information, and children of different ages seem to use different
ways to represent their world. How they represent knowledge and encode events in
their world changes developmentally. The question is whether there is a different
representation for different kinds of information, or whether all kinds of information
are represented in some common format.
A picture of an object is a stand-in for the real thing – it is a concrete yet symbolic
representation of an object that is not physically present. A challenge to children in
developing competence with symbols stems from the dual nature of symbols; every
symbolic artifact is an object in and of itself, and at the same time it also stands for
something other than itself. To understand and use a symbol, a child must mentally
represent its concrete characteristics and its abstract relation to what it stands for
(DeLoache, Pierroutsakos, & Troseth, 1996). This dual representation can be
achieved if the child can keep in mind both the object and its referent. DeLoache
(1987) has demonstrated that basic understanding of the analogical space–object–
symbol relation emerges by about 3 years. That is, 3-year-olds can understand that a
little Snoopy hidden under the little couch in a scale model of a room stands for big
Snoopy being hidden under the big couch in the real room. With age, children
develop the (memory) capacity and the motor skills necessary to represent and re-
create an aspect of their everyday experiences. The ability to create images meaning-
fully is an important human capacity that has been exploited in a number of
methods for improving children’s comprehension and memory (Pressley, 1977).
Case (1985) argued that the development of working memory controls the growth
of cognitive structures. Working memory is therefore a useful construct with which
to measure developmental changes in the information-processing capacity which
drawing requires. Because individuals can accurately perceive, remember, and
manipulate the environment, they clearly have the ability to represent it. Recall can
be facilitated by a more richly represented image, which in turn can presumably
 Children’s Drawings 129

serve as a mirror into a child’s mental representation of that image. Thus, exploring
children’s mental representations through drawings may provide researchers with a
better understanding of their developing knowledge structures.
Children must acquire fine motor skills and develop eye–hand coordination to
carry out the motor functions necessary for drawing. Toddlers’ scribbles appear
unintentional and largely determined by the mechanical structure of the arm, wrist,
and hand. Although initially awareness and planning may be absent, through their
scribbles, children have created a record of their actions. However, researchers
suggest that drawing skills may not be required to create a mental representation on
paper (e.g., Golomb, 2004). Young children with no prior drawing experience tend
to draw human figures spontaneously when given a pencil and paper. Golomb
(2004) identifies children’s drawings of the human figure as the most informative in
terms of cognitive development. This is studied more frequently than other repre-
sentations and allows researchers to assess the development of children’s use of
figural differentiation and representation. Children typically progress from scribbles
to first representational shapes and forms, to complex representations of the human
figure (Golomb, 2004; Harris, 1963). In other words, as children mature and
develop cognitively, their representations move from simple pictures to differenti-
ated, complex ones. A young child will often draw a “tadpole” figure with a circle as
the body/head and sticks as appendages (Golomb, 2004). This representation is a
generic creature shaped to symbolise a person. As children develop, they experiment
with their representation of the human figure and eventually arrive at a differentiated
form that is specific and includes their personal drawing style.
Children’s drawings are greatly influenced by the art of their society and by
schooling. Children in cultures with little interest in art produce simpler forms
(Wales, 1990). Schooling provides opportunities to draw and write, see pictures, and
to understand that artistic forms have meanings that are shared by others (Cox,
1993). Children’s cognitive development as represented in drawings of the human
figure can also reflect a child’s social world. La Voy et al. (2001) explored the idea
that, because cultural differences permeate children’s representations of people,
children from different cultural backgrounds may represent these differences in their
drawings. Cultural differences were examined by looking at differences in social
factors such as smiling, and at details and perception of societal worth expressed by
the height of the figures. Results indicated that American children drew more smiles
and Japanese children drew more details as well as larger figures (La Voy et al.,
2001). Similarly, Case and Okamoto (1996) showed that there are cultural differ-
ences between Chinese and Canadian children’s drawings. These findings suggest
that children’s drawings not only reflect representational development but a child’s
understanding of self and culture as well.
Children’s human figure drawings and in particular family drawings are also used
by psychologists to assess personality (Knoff & Prout, 1985; Naglieri, 1988). The
assumption is that children’s drawings reveal their inner world as reflected in their
representation of experiences with their own family. The advantage of using draw-
ings is that they represent a means of gathering children’s self-report data without
130 I. D. Cherney et al.

some of the limitations of questionnaires or interviews (Pianta, Longmaid, &

Ferguson, 1999). However, individual differences in fine motor skills can confound
both the drawing variables and outcomes of the personality assessments which they
are believed to predict (Pianta & McCoy, 1997). Because of gender differences in
fine motor skills (Halpern, 2000), females may be at an advantage if interpretations
are based on criteria that may be confounded with motor skills. For example,
Koppitz (1968) found that girls tended to include more body parts and clothing in
their figures than boys. Because drawings are often determined and interpreted, in
part, on the basis of realism (Pianta et al., 1999), which is identified through
completeness of the human figure(s), the additional details drawn by the girls could
be misinterpreted as manifesting superior intelligence (Goodnow, 1977; Naglieri,
1988). If drawings are used for school placement purposes, boys may be at a disad-
vantage. In addition, sex differences in children’s free drawings may indicate that
boys and girls have different feelings and ways of perceiving objects when they draw
pictures. It is therefore important to examine gender differences further and to
establish a baseline for the various types of drawings that are frequently used for
diagnostic purposes. In addition, fine motor skills develop rapidly in children
between 5 and 9 years old. Therefore, chronological age is an important factor to
consider in drawing performance.
Children’s drawings show dramatic gains in organisation, detail, and representation
of depth during middle childhood, when some depth cues such as making distant
objects smaller than near ones begin to appear (Braine, Schauble, Kugelmass, &
Winter, 1993). Compositional development, the arrangement of the different parts of
the picture, is evident in children’s drawings of their families. Golomb (2004) suggests
that most children across age groups align family members side by side facing forward
although, with increasing age, composition becomes more relevant to a child’s draw-
ing; 3–4 year olds do not completely align family figures and do not randomly place
the figures on the page, but in older children, alignment improves and they display an
increased awareness of the coordinates of the paper. Attending to a page’s quadrants
allows children to experiment with the placement of the figures. Thus framing, align-
ment, and compositional tasks are strategies children use to group their figures
together.
The most frequently used strategy is that of alignment. It reveals which figures are
grouped together (Golomb, 2004). Gender grouping is also a strategy used by many
children. For example, children will draw all girls similarly and yet differently from
the boys. Symmetrical ordering of family members is another grouping device used
by children that increases with age. Symmetrical ordering refers to how children
arrange the figures. For example, grouping older and younger children separately, or
placing parental figures in the middle of the family with children on either side, can
be seen increasingly with age. In using the symmetrical ordering strategy, children
reveal information about their familial relationships. In addition, information about
who the figures are is evident from the size relations, with parents being taller than
the children. Thus, children’s family drawings hold promise as a method for inter-
preting representations of this important emotional relationship.
 Children’s Drawings 131

Another important aspect of a child’s development is the ability to depict spatial

elements from his or her environment through an understanding of where an object
is located in comparison to him- or herself. Liben and Downs (1993, 2001) found
that children as young as three could recognise maps as symbolising referent spaces
but that their understanding was limited due to their developing symbolic and
spatial concepts. They also showed that by the age of five or six, children can identify
spatial relations on a map, but have difficulties with its geometry. Moreover, boys
tended to perform better than girls in correctly identifying the direction of an object
when the map was turned 180° in comparison to how the real room was laid out
(Liben & Downs, 1993). This finding is consistent with other studies that have
shown males outperforming females on tasks necessitating visual–spatial and mental
rotation abilities (e.g., Cherney & Collaer, 2005; Cherney & Neff, 2004; Halpern,
2000).
A child’s understanding of spatial tasks can also be explored through his/her draw-
ings. Very young children fail to comprehend a picture’s reference point of view and
are unable to understand individual references despite understanding that the
picture or map represents another object (Liben & Downs, 1993). In addition,
young children have difficulty understanding a picture’s viewing angle. For example,
Liben and Downs (2001) found that when shown an aerial photograph of Chicago
and asked to identify grass, young children tended to identify the bottom of the page
as grass. Similarly, many young children reported that representational roads could
not be roads because two cars would not be able to fit on them.
Young children tend to perform better on topological mapping tasks whereas
older children (8–12-year-olds) tend to perform better on projective and Euclidean
relations, thus demonstrating a better understanding of reference points (Golomb,
2004; Liben & Downs, 2001). Golomb notes that 7- and 8-year-olds begin to
portray near–far relations using the vertical axis of the page. Moreover, an important
aspect of spatial development is the ability to view the world aerially. Iijima, Arisaka,
Minamoto, and Arai (2001) found that row or frontal arrangements were most
commonly used by girls, whereas the incidence of aerial composition was signifi-
cantly higher in boys’ drawings. Boys also tended to draw dynamic objects (e.g.,
cars, machines, robots) in bold colours whereas girls tended to draw nature and life-
oriented objects (e.g., humans, flowers, butterflies) using warm colours. These sex
differences in drawings mirror some of the sex differences found in toy preferences
and leisure activities (e.g., Cherney & London, submitted).
Representational development is also evident when examining children’s drawings
of schools. Tallandini and Valentini (1991) found that children’s representations of
schools were expressed in three different pictorial components: (a) the building
structure (a global component); (b) the building access; and (c) the windows. Using
these components, Tallandini and Valentini observed children’s drawings of school
buildings and found a progression in representational strategy, with a clear associa-
tion between age and strategy for all these pictorial components. Children’s most
rudimentary strategy, labelled failed symbolism, either did not have these pictorial
components or they were unrecognisable. The second strategy, generic symbolism,
132 I. D. Cherney et al.

had pictorial components similar to a house, with no evidence of it being a school

building. Specific symbolism, the third strategy, had components belonging to any
school building. Finally, realism included pictorial components that made unequivo-
cal reference to a specific, real school, usually the school attended by the child
(Tallandini & Valentini, 1991). Children aged 5–7 draw schools like a simple house
without any particular indications that it represents a school. As children conceptu-
ally differentiate categories of buildings, they modify the picture into a more specific
prototype.
In addition to understanding space locations, a child’s cognitive development is
also revealed by the complexity of his or her drawings. Bensur, Elliot, and Hedge
(1997) found age differences in the developmental complexity of children’s draw-
ings. These differences were related to their recognition and memory of various
objects found in their environment. Although it was implied that a more developed
working memory was responsible for more complex drawings, it is also evident that
increased spatial ability also influences complexity. Bensur et al.’s study concluded
that increased drawing complexity reveals advanced working memory and spatial
abilities. In addition, Case and Okamoto (1996) demonstrated that training or addi-
tional drawing experiences contribute to the complexity of the drawings (e.g.,
Chinese schools provide specific instruction on how to render certain objects, such
as the human figure). The present study explored these assumptions by examining
the relationship between children’s recall and the complexity of their family and
school drawings. It was hypothesised that children who portrayed more details in
their drawings would also recall more stimuli (toy pictures) and have more drawing
experience.
Taken together, these studies suggest that children’s drawings illustrate an array of
mental representations and cognitive abilities. In the present study we attempted to
juxtapose two types of drawings that are used to index children’s feelings and under-
standing concerning the objects depicted: (a) one that depicts an important
emotional relationship, i.e., the family; and (b) one that depicts a familiar spatial
object/location, i.e., the school. We examined whether children’s representations
would differ between these two types of relationships and how age and gender would
influence the 5–13 year-olds’ representations of the pictures. Because females tend
to be more attentive to and have more experience with familial relationships (Matlin,
2004) and males tend to be more attentive to and have more experience with spatial
tasks (e.g., Cherney & London, submitted), we reasoned that, similar to La Voy
et al.’s (2001) cultural differences in children’s drawings, we would find gender
differences between children’s family and school drawings. In particular, we hypoth-
esised that girls would draw more essential and inessential details than boys in their
family drawings and that boys would draw more essential and inessential details than
girls in their school drawings. We also hypothesised that boys would be more likely
use an aerial perspective than girls (e.g., Iijima et al., 2001). Based on the findings
that with age children’s drawings become more complex and more realistic, we
hypothesised that older children would draw more aerial views of their school build-
ings than younger children and that the latter would be more likely than older
 Children’s Drawings 133

children to use stick figures. In general, with age, we expected children’s drawings to
contain more details.

Method
Participants
A total of 109 (54 boys and 55 girls) 5–13-year-old children’s drawings were analy-
sed. Each of the three age groups comprised a similar number of boys and girls (38
5–7-year-olds: M = 76.65, SD = 11.62; 34 8–10-year-olds: M = 115.90, SD
= 10.28; 37 11–13-year-olds: M = 147.98, SD = 8.52). Of these, 87% were
European-Americans, 7% were Asian Americans, 3% African Americans, and 1%
Hispanics. Participants were recruited from local childcare centres and a small
private school in the Midwest. Children received a small prize for their participation
in the study.

Procedure
Each child was tested individually. The drawing tasks were used as distracter tasks in
a study on the development of incidental memory (Cherney, 2005). Each child was
exposed to two sets of 18 toy pictures that were presented on a computer screen and
was asked to label each toy. After the first set of pictures, the participant received a
pencil and a piece of white, standard size paper. The child was given five minutes to
draw a picture following the instructions: “Please draw your family.” Then, s/he was
shown the second set of 18 toy pictures and asked to label them. Following this
encoding task, the participant was asked to draw his/her school on the reverse side of
the sheet of paper. After each drawing session, the participant was asked to recall the
pictures s/he had seen previously.

Coding
The family drawings were coded by three trained coders who were blind as to the
hypotheses of this study. Inter-rater reliability was 92%. They used 16 criteria that
illustrated key concepts in the drawings (e.g., Bensur et al., 1997; Golomb, 2004;
Harris, 1963; La Voy et al., 2001). All of the items, except four, were coded dichot-
omously. The dichotomous categories included whether or not the child used stick
figures, whether any figure was shown smiling, whether all figures were smiling,
proportionality (appendages vs. the trunk of figure), whether or not any pets were
included in the family picture, whether the figures were represented with clothing,
whether the clothing was stereotypic (e.g., included skirts or triangles for female
figures), whether the picture included a particular setting/composition, whether the
child wrote any words or letters to accompany the picture, whether members of
the extended family were included, whether some members were holding hands, and
the placement of figures on the page (immediate centre/whole page, top, bottom,
134 I. D. Cherney et al.

left, right, top and left, top and right, bottom and left, bottom and right). Categories
that were not coded dichotomously included the number of essential and inessential
details, and the average heights of male and female figures. Essential details were
defined as the hair, nose, mouth, eyes, hands, feet, arms, and legs; each item was
attributed a point for inclusion, to a maximum of eight points. Inessential details
were defined as any additional detail that was not previously categorized (e.g., acces-
sories, fingernails, etc.). The heights of all male and all female figures were averaged
to create an average male height and average female height.
The school drawings were coded using 10 criteria. Six of these criteria were coded
dichotomously. The dichotomous categories included the point of reference chosen
by the child (i.e., frontal or aerial views), whether the child chose to draw the interior
or exterior of the school, the proportionality (doors and window proportionate to the
building and each other), whether or not the name of the school was written on the
drawing, whether there were any other words included in the picture, and the place-
ment. Non-dichotomous categories included essential and inessential details, the
number of windows or desks (depending on view), and the number of people
included in the picture. Essential details, defined for the exterior view as a door and
windows and for the interior view as desks and chairs, were counted individually
(range 0–2). There was no range limit for inessential details. The placements of the
figures on the page were coded using the same scheme as with the family drawings.
The number of correctly identified toy pictures was summed across the two recall
tasks.

Results
The drawings were coded by three trained experimenters. Interrater reliability was
89%. Drawings were analysed separately. A total of 109 family drawings and 110
school drawings were analysed (see Figure 1 for examples – one for each age group,
for each type of drawing). Separate 2 (sex) × 3 (age group) analyses of variance
(ANOVA) on the continuous variables were performed. For the number of essential
details in the family drawings, there was a significant main effect of age (F[2,103]
= 4.37, p = .015, η2 = .11). Post-hoc tests (Tukey’s HSD; p < .05) showed that the
youngest children (M = 6.11, SD = 1.54) drew significantly fewer essential details
than the 8–10-year-olds (M = 6.82, SD = 1.58) and the oldest children (M = 7.05,
SD = 1.20). The two older age groups did not differ significantly in their drawing of
essential details. A 2 (sex) × 3 (age) ANOVA on the inessential details showed signif-
icant main effects of sex (F[1,103] = 8.25, p = .005, η2 = .07) and age (F[2,103]
= 11.95, p = .001, η2 = .18). As hypothesised, girls (M = 3.37, SD = 2.94) drew
significantly more inessential details than boys (M = 2.07, SD = 2.21) and the
youngest children drew significantly fewer inessential details (M = 1.25, SD = 1.81)
than the middle (M = 3.05, SD = 2.52) and oldest children (M = 3.87, SD = 2.89;
Tukey’s HSD, p < .05). In addition, there was a marginal sex difference in the aver-
age female height drawn (F[1,103] = 3.84, p = .05, η2= .04). On average, girls (M
= 7.37, SD = 3.80) tended to draw the female characters taller than did the boys (M
 Children’s Drawings 135

Figure 1. An example of family and school drawings for each age group

= 5.87, SD = 4.25). For the school drawings, the only significant main effect was a
sex difference in the number of desks or windows drawn (F[1,103] = 3.96, p = .049,
η2= .04). On average, males drew 7.59 (SD = 17.29) windows/desks and females
13.32 (SD = 12.82).
Separate chi square tests on the dichotomous variables by sex and age were
Figure 1. An example of family and school drawings for each age group

computed. For the family drawings, results indicated significant sex differences for
clothing (χ2[1, n = 109] = 6.31, p = .012), the use of stereotyping (χ2[1, n = 109]
= 13.16, p = .001) and proportionality (χ2[1, n = 109] = 4.81, p = .028; see Table
1). Girls were more likely than boys to use clothing and stereotyped features in their

Table 1. Gender differences (n) in family drawings

Boys Girls Total

With clothing 30 43 73
Without clothing 24 12 36
Stereotypic 6 23 29
Not stereotypic 48 32 80
Proportional 2 9 11
Not proportional 52 46 98
136 I. D. Cherney et al.

Table 2. Age differences in family drawings

Number of children in different age

groups

Age 5–7 years 8–10 years 11–13 years Total

Stick figures 19 8 9 36
No stick figures 19 28 28 75
With clothes 18 25 30 73
No clothes 20 9 7 36

drawings and they were more likely to draw proportionate figures. In terms of age,
there were significant differences in the use of stick figures (χ2[2, n = 109] = 7.60,
p = .022) and in the use of clothing (χ2[2, n = 109] = 10.59, p = .005; see Table 2).
Not surprisingly, the youngest children were more likely to draw stick figures and
less likely to use clothing in their depictions. The only significant difference between
the school drawings was that of the reference point children of different ages chose
(χ2[2, n = 109] = 8.69, p = .013); 38% of the older children used an aerial view, as
did 18% of the 8–10-year-olds and 10% of the youngest age group. There were no
sex differences for the school drawings.
Finally, to test whether drawing complexity increased with independent memory
recall, correlational analyses of the essential and inessential details in both drawings
and the total recall scores revealed a significant correlation between essential
details and memory scores (r[109] = .24, p = .012) as well as between the inessential
details and memory scores (r[109] = .34, p = .001) for the family drawings only. No
correlations were found between the details of the school drawings and memory
scores. To explore further the relationship between detailed drawings and recall, the
essential and inessential details for both drawings were summed to form a “detail
index”. The simple regression with recall scores as the dependent variable and the
index as the predictor was significant (F[1,107] = 7.89, p = .006; R2 = .069; b =
.25). The index accounted for only 7% of the variance in memory.

Discussion
Understanding children’s representational development is an essential component
for constructing a more complete picture of cognitive development. The representa-
tion of knowledge in long-term memory has been portrayed as an important explan-
atory component of memory performance and cognitive development (Schneider &
Pressley, 1985). One key issue that all theories of cognitive development must
address concerns age differences in how children represent experience. A critical
issue is the ability to understand, use, and produce symbols. They allow us to
communicate; to abstract, simplify, and generalise; to hold information far from the
referent; to manipulate and transform information, etc. (e.g., Liben & Downs,
 Children’s Drawings 137

1993). The purpose of the present study was not only to examine and establish a
baseline for 5–13-year-old children’s family and school drawings, but also to investi-
gate how memory performance would be correlated with the children’s graphical
representations. Because children’s drawings are often used for diagnostic and refer-
ral purposes (e.g., Pianta et al., 1999), it is crucial to assess children’s drawing skills
in a non-clinical sample. We chose to examine two types of drawings that are used
with clinical samples. The results for each type of drawing are described in turn.
In general, the results of the current study support the hypothesis that with age
children’s ability to create more complex drawings increases and, in particular, that
the difference in representational complexity is linked, at least in part, to an
increased working memory capacity as measured by an independent memory recall
task (Bensur et al., 1997; Scott, 1981). The family drawings showed that, consis-
tent with other studies and our hypothesis, older children included significantly
more essential and inessential details than younger children (e.g., Bensur et al.,
1997; Case, 1985; Golomb, 2004; La Voy et al., 2001); they were more likely to
draw figures with clothes and were less likely than younger children to use stick
figures. In other words, with age, children’s drawings included more distinctive
features, making the pictures appear more realistic. They frequently wrote the
names of the family members, indicating that they were drawing an experienced
relationship. Younger children made comments suggesting that they were also
drawing from experience. The fact that they included pets also suggests that they
were aware that they were drawing themselves. When drawing their family, the
children first used symbolic strategies and later a realism strategy (Tallandini &
Valentini, 1991). In comparison, children in the youngest age group produced
more stick figures.
These findings may illustrate the development of differentiation abilities
(Golomb, 2004). With increased cognitive (working memory) capacity, children are
more easily able to differentiate between representations (Case, 1985), moving from
simple (e.g., stick figures) to complex representations (clothed individuals with
detailed facial features). This increased differentiation is also manifested in the find-
ing that the number of details in the children’s family drawings was significantly
correlated with their recall of toy pictures. It is possible that asking children to recall
aspects of their own drawings rather than using an independent recall measure
would have yielded higher recall scores. The complexity of children’s drawings is
probably influenced by children’s ability to recall figural configurations and to hold
them in working memory. The results may also simply reflect children’s maturation
and development of motor skills. However, it is interesting to note that there was no
association between chronological age and details depicted in the children’s school
drawing, suggesting that these two types of drawings may tap differential cognitive
resources. Similarly, children’s placement of family figures was the same across the
age groups, suggesting that children as young as 5 years old have developed a
schema or mental framework about the alignment of family members.
The current findings also showed significant sex differences among the family
drawings. Girls were more likely than boys to draw clothes, add stereotyped details
138 I. D. Cherney et al.

(e.g., fingernails, hairstyles, jewellery) to their figures, and draw proportionate

human figures. These findings are related to the fact that, on average, girls drew
significantly more inessential details than boys, and they are consistent with previous
studies (e.g., Koppitz, 1968) showing that girls tend to depict more details in their
family drawings. Together, these results suggest that the girls’ drawings may repre-
sent their experiences with family relationships. In general, girls tend to value
relationships more than boys and they tend to pay more attention to what other
same-sex friends wear (Matlin, 2004). Thus, the depictions may not necessarily
reflect a different stage of intellectual development. Furthermore, there were no age-
related differences in the number of children who used stereotyped items in their
drawing, suggesting that girls may develop a prototype or schema for what it means
to be female. This schema remains rather invariant over time. This prototype may
reflect their attempt to draw things as they are known rather than as they are actually
perceived. It may be a means for differentiating females from males. The finding that
girls were more likely than boys to draw proportionate human figures may be due to
their expertise in drawing and family relationships. Girls tend to spend more time
drawing than boys (Cherney & London, submitted).
Interestingly, girls also tended to draw the female figures taller than did the boys.
Golomb (2004) suggests that a child’s composition of a drawing may be reflective of
the relationships in his or her life. For example, the hierarchy of the family may be
made explicit by children drawing parents first followed by the children (Golomb,
2004). The tendency for girls to draw larger female figures may indicate the
children’s high self-esteem. The current sample was composed of children with a
middle to high SES background. The children reported spending several hours each
day in extra-curricular activities. It is therefore reasonable to assume that these girls
were likely to have high self-esteem. However, it is also likely that the boys had
equally high self-esteem. Another possible reason for the taller female figures may be
that girls adopt more sophisticated and controlled drawing techniques in relation to
emotional characters than boys. Hammer (1997) suggests that girls draw in a more
controlled manner than boys at most ages. However, if boys were drawing in a less
controlled manner than girls, this may have resulted in larger figures, in part because
of the reduced level of production control involved. La Voy and her colleagues
(2001) suggest that the size of figures may reflect a child’s view of his or her worth
within a culture. Burkitt, Barrett, and Davis (2003) found that children drew human
figures taller than the baseline following a positive characterisation, and shorter
following a negative characterisation. In particular, boys tended to draw larger posi-
tive drawings than girls, but there were no gender differences between the negative
drawings. Overall, Burkitt et al.’s results suggest that the size of children’s drawings
may reflect an affective component of their representation. That study also found
that the oldest group drew taller and larger drawings than the youngest group,
running counter to previous findings that children’s drawings become smaller with
age (Cox, 1992, 1993). Our current findings did not show any age-related size
differences. Lack of power may be one reason why there were no age differences,
although absolute numbers showed a trend toward drawing larger figures with age.
 Children’s Drawings 139

Future studies should continue to examine the effects of affective characterisations

on the size of children’s drawings.
With regard to the school drawings, we hypothesised that older children would
be more likely than younger children to draw an aerial view of their school build-
ing, indicating a better understanding of spatial relations. The findings were consis-
tent with our hypothesis and Liben and Downs (2001) findings that as children
mature, they develop more complex spatial abilities, allowing them to view the
world from different points of view. As Piaget and Inhelder (1956) theorised,
children first develop a sense of topological concepts, including relations such as
proximity, order, and separation. Next, children develop projective and Euclidean
concepts – that is, they are able to represent spatial relations such as left or right,
front or behind, front or side, and they develop a reference system that allows them
conservation of distance and angle. Similarly, researchers suggest that young chil-
dren rely upon local landmarks or local configurations before becoming increas-
ingly more flexible in using alternative frames of reference and finally mastering
precise metrics of space (e.g., Liben & Downs, 2001; Newcombe, Huttenlocher,
Drummey, & Wiley, 1998). It is therefore not surprising that older children are
more likely to draw their school building(s) from a different vantage point.
However, it is worthy to note that the participants in the current study, although
they drew their school, have never seen or experienced their school from a bird’s-
eye view. No aerial picture of the school exists. Thus, the children had to create a
mental image/concept of the school grounds, use symbols to convey that image,
transform this large, irregular, three-dimensional surface to a small, regular, two-
dimensional surface, and translate the symbols into fine motor movements.
Presumably, these skills improve with a higher working memory capacity. Children
may have used symbolism and realism strategies simultaneously (Tallandini &
Valentini, 1991). This significant age difference may also be due, in part, to expo-
sure to alternative reference points during geography, algebra, and/or art classes.
As Case and Okamoto (1996) showed, additional training and exposure is often
translated into increased realism.
There were no gender-related differences in the way in which the school was
portrayed, with the exception of the difference in desks/windows production. Girls
drew significantly more desks/windows than boys. This finding was contrary to our
hypothesis, but was consistent with other findings (e.g., Koppitz, 1968) showing
that girls tend to depict more details in their pictures. These results suggest that girls
may have a more developed schema for school buildings than boys and/or that sex
differences in motor skills may confound the findings. Remarkably, there were no
correlations between essential and inessential details and the children’s recall. In
general, the school drawings were very similar in that the children used few details
and tended to use a general pictorial prototype of a house. The private school’s main
architectural site is a large, red-bricked colonial house. In addition, many children
chose to draw their classroom rather than the school building. A lack of variance
may therefore be the cause of this finding; the range of scores was more restricted
than that of the family drawing. Alternatively, the drawings may reflect the fact that
140 I. D. Cherney et al.

the school building is an inanimate object that does not change over time whereas
family sizes (and heights) change.
In conclusion, the present findings demonstrate that the relationship between
children’s depictions of families and school buildings is complex. Understanding
how a child sees him- or herself within the context of the family can expand our
knowledge about the development of identity and relationships. The sex differences
in family drawings suggest that girls and boys represent families differently.
Although there were many age-related differences in the family drawings, with chil-
dren’s drawings becoming more realistic with age (as measured by an increase in the
number of details), there was only one age-related difference in the school drawings.
With age, children’s representations of their school changed from a frontal perspec-
tive to an aerial one, suggesting a change from relying upon landmarks and local
configurations to alternative frames of references (Newcombe et al., 1982). Similar
to Bensur (1997), the present findings show an increasing complexity in children’s
drawings requiring manipulation of spatial components. These could be influenced
by three developmental variables: (a) the children’s capacity to attend to the appear-
ance of objects regardless of the internal representation a child may possess; (b) their
ability to recall figural configurations and to hold them in working memory; and (c)
the effects of fine motor skills (Bensur & Eliot, 1993). One third of the older
children chose to use this new spatial perspective. However, it is difficult to offer an
uncontroversial interpretation of the correlational data.
As Liben and Downs (2001) note, the process of spatial thinking is important not
only to acquire the skills to read maps, but also for the graphic representations of
places and space. Map skills and spatial thinking play a crucial role in geography, but
they are also related to other educational subjects such as geometry, physical educa-
tion, art, etc. Computer graphics rely on spatial and graphic representational skills.
These graphics are increasingly used for the visualisation of data. Scientific visualisa-
tion is central to contemporary work in fields as diverse as biology, chemistry,
physics, anthropology, history, demography, geology, meteorology, psychology, and
astronomy to name a few (Liben & Downs, 2001). It is therefore crucial to provide
children with the spatial training that will enable them to represent space from a
different perspective. Children’s drawings may be a particularly useful tool for
assessing their development in these spatial and familial representations.

Author Note
This study was presented at the 2003 3rd Biennial Meeting of the Cognitive Devel-
opment Society, Park City, UT. We would like to thank the children of the
Brownell-Talbot school for their participation in this study.

References
Bensur, B., & Eliot, J. (1993). Case’s developmental model and children’s drawings. Perceptual and
Motor Skills, 76, 371–375.
 Children’s Drawings 141

Bensur, B. J., Eliot, J., & Hedge, L. (1997). Cognitive correlates of complexity of children’s
drawings. Perceptual and Motor Skills, 85, 1079–1089.
Braine, L. G., Schauble, L., Kugelmass, S., & Winter, A. (1993). Representation of depth by
children: Spatial strategies and lateral biases. Developmental Psychology, 29, 466–479.
Burkitt, E., Barrett, M., & Davis, A. (2003). The effect of affective characterizations on the size of
children’s drawings. British Journal of Developmental Psychology, 21, 565–584.
Case, R. (1985). Intellectual development: Birth to childhood. Orlando, FL: Academic Press.
Case R., & Okamoto, Y. (1996). The role of central conceptual structures in the development of
children’s thought. Monographs of the Society for Research in Child Development, 246(61), 1–2.
Cherney, I. D. (2005). Children’s and adults’ recall of sex-stereotyped toy pictures: Effects of
presentation and memory task. Infant and Child Development, 14, 11–27.
Cherney, I. D., & Collaer, M. L. (2005). Sex differences in line judgment: Relationship to mental
rotation, academic preparation and strategy use. Perceptual and Motor Skills, 100, 615–627.
Cherney, I. D., & London, K. L. (submitted). Gender-linked differences in the toys, television shows,
computer games, and outdoor activities of 5–13-year-old children.
Cherney, I. D., & Neff, N. L. (2004). Role of strategies and prior exposure in mental rotation.
Perceptual and Motor Skills, 98, 1269–1282.
Cox, M. V. (1985). One object behind another: Young children’s use of array-specific or view-
specific representation. In N. H. Freeman & M. V. Cox (Eds.), Visual order: The nature and
development of pictorial representation (pp. 188–200). Cambridge UK: Cambridge University
Press.
Cox, M. V. (1992). Children’s drawings. Harmondsworth, UK: Penguin.
Cox, M. V. (1993). Children’s drawings of the human figure. Hove, UK: Erlbaum.
DeLoache, J. S. (1987). Rapid change in the symbolic functioning of very young children. Science,
238, 1556–1557.
DeLoache, J. S., Pierroutsakos, S. L., & Troseth, G. L. (1996). The three ‘R’s of pictoral compe-
tence. In R. Vasta (Ed.), Annals of child development (Vol. 12, pp. 1–48). Bristol PA: Jessica
Kingsley.
Freeman, N. H. (1980). Strategies of representation in young children: Analysis of spatial skills and
drawing processes. London: Academic Press.
Golomb, C. (2004) The child’s creation of a pictorial world. Mahwah, NJ: Lawrence Erlbaum.
Goodnow, J. J. (1977). Children’s drawings. Cambridge, MA: Harvard University Press.
Halpern, D. F. (2000). Differences in cognitive abilities (3rd ed.). Mahwah, NJ: Lawrence Erlbaum
Associates.
Hammer, E. F. (1997). Advances in projective drawing interpretation. Springfield, IL: Thomas.
Harris, D. B. (1963). Children’s drawings as measures of intellectual maturity: A revision and extension
of the Goodenough Draw-a-Man Test. New York: Harcourt, Brace and World, Inc.
Iijma, M., Arisaka, O., Minamoto, F., & Arai, Y. (2001). Sex differences in children’s free draw-
ings: A study on females with congenital adrenal hyperplasia. Hormones and Behavior, 40,
99–104.
Knoff, H. M., & Prout, H. T. (1985). Kinetic Drawing System for family and school: A handbook. Los
Angeles: Western Psychological Services.
Koppitz, E. M. (1968). Psychological evaluation of children’s human drawings. New York: Crune and
Stratton.
La Voy, S. K., Pederson, W. C., Reitz, J. M., Brauch, A. A., Luxenberg, T. M., & Nofisnger, C. C.
(2001). Children’s drawings: A cross-cultural analysis from Japan and the United States.
School Psychology International, 22, 53–63.
Liben, L. S., & Downs, R. M. (1993). Understanding person–space–map relations: Cartographic
and developmental perspectives. Developmental Psychology, 29, 739–752.
Liben, L. S., & Downs, R. M. (2001). Geography for young children: Maps as tools for learning
environments. In S. L. Golbeck (Ed.), Psychological perspectives on early childhood education:
Reframing dilemmas in research and practice (pp. 220–252). Mahwah, NJ: Lawrence Erlbaum.
142 I. D. Cherney et al.

Matlin, M. W. (2004). The psychology of women (5th ed.). Belmont, CA: Thomson/Wadsworth
Publishers.
Naglieri, J. A. (1988). Draw-a-Person: A quantitative scoring system. New York: Psychological
Corporation.
Newcombe, N., Huttenlocher, J., Drummey, A. B., & Wiley, J. G. (1998). The development of
spatial location coding: Place learning and dead reckoning in the second and third years.
Cognitive Development, 13, 185–200.
Piaget, J., & Inhelder, B. (1956). The child’s conception of space. New York: Basic Books.
Pianta, R. C., Longmaid, K., & Ferguson, J. E. (1999). Attachment-based classifications of
children’s family drawings: Psychometric properties and relations with children’s adjustment
in kindergarten. Journal of Clinical Child Psychology, 28, 244–255.
Pianta, R. C., & McCoy, S. (1997). The first day of school: The predictive utility of an early
school screening program. Journal of Applied Developmental Psychology, 18, 1–22.
Pressley, M. (1977). Imagery and children’s learning: Putting the picture in developmental
perspective. Review of Educational Research, 49, 319–370.
Schneider, W., & Pressley, M. (1997). Memory development between two and twenty (2nd ed.).
Mahwah, NJ: Lawrence Erlbaum Associates Publishers.
Scott, L. H. (1981). Measuring intelligence with the Goodenough-Harris Drawing Test. Psychological
Bulletin, 89, 483–505.
Tallandini, M. A., & Valentini, P. (1991). Symbolic prototypes in children’s drawings of schools.
Journal of Genetic Psychology, 152, 179–190.
Wales, R. (1990). Children’s pictures. In R. Grieve & M. Hughes (Eds.), Understanding children
(pp. 140–155). Oxford, UK: Blackwell.
View publication stats