Construction Play 1
Children’s Play Preferences, Construction Play with Blocks, and Visual-Spatial Skills:
Are They Related?
Yvonne M. Caldera Anne McDonald Culp Marion O'Brien
Texas Tech University University of Alabama University of Kansas
Rosemarie T. Truglio Mildred Alvarez Aletha C. Huston
Children’s Television Workshop San Jose State University University of Texas
The research was supported by a grant from the General Research Fund of the University of
Kansas and by training grant 5T32HD07173 from the National Institute for Child Health and Human
Development. The authors are grateful to Mary Pinon for her assistance in data collection and coding,
for the editors for providing feedback on the manuscript, and to the children and teachers who
participated in the project. Address correspondence and reprint requests to Yvonne M. Caldera,
Department of Human Development and Family Studies, Texas Tech University, Box 41162,
Lubbock, TX 79409.
Running head: Construction Play
Construction Play 2
Fifty-one preschoolers' play preferences, skills at assembling block structures, and spatial abilities
were recorded in this study. There were no sex differences in children's visual-spatial skills, and play
with art materials and children's free and structured play with blocks were related to spatial
visualization. Two patterns emerged from the findings: 1) activity and performance representing skills
in spatial visualization and visual motor-coordination, and 2) creativity, or the ability to break set and to
produce varied solutions using visual materials. Future research might examine the extent to which
children’s play activities and experiences predict these types of skills.
Construction Play 3
Children’s Play Preferences, Construction Play with Blocks and Visual-Spatial Skills:
Are They Related?
It is widely recognized that play, a spontaneous, self-guided activity, provides children
opportunities to learn and practice important cognitive and social skills. Visual spatial skill is one area
of cognitive ability that has frequently been associated with sex differences (Huttenlocher & Levine,
1993) and with particular play preferences (Tracy, 1987). Interest in the determinants of visual-spatial
skills has been stimulated by consistent findings that spatial abilities are related to math and science
achievement (Casey, Pezaris, & Nuttall, 1992; Tracy, 1990) and that males tend to perform better than
females on most visual-spatial tasks, at least by adolescence (Burns & Reynolds, 1988; Geary, Gilger,
& Elliott-Miller, 1992; Harris, 1981; Maccoby & Jacklin, 1974; McGee, 1979; Tracy, 1990), as early as
middle childhood (Johnson & Meade, 1987; Kerns & Berenbaum, 1991) and even preschool
(Huttenlocher & Levine, 1993).
Spatial abilities, according to Linn and Petersen (1985) involve more than one component. In
a meta-analysis of studies investigating sex-differences in spatial abilities, they identified three primary
components: a) spatial perception, or determining spatial relations of objects with respect to one's
own body orientation; b) mental rotation, or identifying a model in a different orientation; and c) spatial
visualization, which involves multistep mental manipulations of spatially presented information. Linn
and Petersen (1985) found that the magnitude of sex differences in visual-spatial skills is largest for
mental rotation, medium for spatial perception, and smallest for spatial visualization tasks. These
findings have been further replicated with more recent studies in a meta-analysis by Voyer, Voyer, and
Bryden (1995). Voyer et al (1995) further discovered that effect sizes in sex differences found in the
literature vary as a function of partitioning of data (differences more marked in older ages) and some
aspect of how the tests were administered or scored (individually vs. in group).
The origin of this sex difference remains a source of considerable debate. Both genetic and
hormonal explanations have been proposed (Casey & Brabeck, 1989; Kimura & Hampson, 1994;
Lewis & Harris, 1988; McGee, 1979; Newcombe & Bandura, 1983; Resnick, Gottesman, Berenbaum,
Construction Play 4
& Bouchard, 1986; Voyer & Bryden, 1990). For example, females with congenital adrenal hyperplasia
(CAH) (Resnick et al., 1986) and right-handed females with nonright-handed relatives (Casey &
Brabeck, 1989) demonstrate high levels of spatial ability as compared to their female counterparts.
On the other hand, some investigators have stressed the importance of environmental factors
such as sex-typed play activities in the development of visual-spatial skills (Newcombe , 1982;
Sherman, 1967). This view emphasizes socialization differences in childhood (Fagot, 1977, 1978).
There is evidence that boys play more often than girls with blocks and other toys that provide practice
with two- and three-dimensional manipulations and transformations (Tracy, 1987), and this "practice
deficit" may contribute to girls' poorer performance on visual-spatial tasks. A meta-anlaysis in the role
of experience in spatial skills performance by Baenninger & Newcombe (1989) found that for both
males and females, participation in spatial activities is related to spatial ability. This relationship,
however, accounts for only 9% of the variance in spatial skills. The present study was designed to
further investigate the link between play preferences in the classroom, ability at block play and visual-
spatial skills of girls and boys in the preschool age when sex differences are lowest.
Spatial Abilities and Play Activities
Correlational studies have shown that participating in spatial activities such as ball playing and
playing with "manipulative" toys are positively related to preschoolers' visual spatial skills.
Preschoolers who frequently engaged in "masculine" play activities have been shown to perform
better on the Preschool Embedded Figures Test (PEFT) (spatial visualization) than those who spent
less time in masculine activities (Fagot & Littman, 1976). After coding preschoolers' free play
preferences as either masculine or feminine, Connor and Serbin (1977) found that boys' preferences
for masculine activities was positively related to their performance on the PEFT, but not to their
performance on the Block Design subtest of the Wechsler Preschool and Primary Scale of
Intelligence (spatial visualization with mental rotation component). In contrast, activity choices were
not related to girls' performance on these spatial tasks. A replication study with a larger sample,
however, demonstrated that preschoolers of both sexes who were rated as frequently engaging in
Construction Play 5
masculine activities performed better on the Block Design task than did those identified as "high
feminine" children (Serbin & Connor, 1979).
Anthropological studies conducted in Kenya also provide support for the hypothesis that
engaging in spatial activities enhances performance on spatial tasks (Munroe & Munroe, 1971;
Nerlove, Munroe, & Munroe, 1971). Children who had more experience exploring the physical
environment away from their home performed better on the Copying Blocks subtest of the
Stanford-Binet than did children who stayed close to home.
In addition to these correlational studies, several experimental studies have demonstrated
that visual spatial skills can be learned through training and practice and will generalize to a related
task (Connor, Serbin, & Schackman, 1977; Connor, Schackman, & Serbin, 1978; Matthews, 1987).
To determine if male superiority on visual spatial tests is a direct result of masculine play experiences,
Sprafkin, Serbin, Denier, and Connor (1983) randomly assigned preschoolers to a training session
during which they were exposed to masculine toys (e.g., blocks and tinker toys). Children exhibiting
high or low rates of male-preferred selections in free play were equally represented in both training
and control conditions. Two training sessions over a six week period consisted of three phases: (a)
children were instructed about how the materials may be used and shown relevant features (e.g.,
number of sides); (b) children were encouraged to make various structures out of the materials; and
(c) children played with the materials on their own during a free play session. Performance on the
Block Design subtest of the WPPSI and the Punched Holes Test was significantly greater at post-test
for children in the training condition, whereas pre- and post-test scores did not differ for the control
group, thus providing evidence for a causal relation between play with construction toys and spatial
visualization task performance.
To date, however, there are no studies in which children's structured and unstructured play
with blocks and similar manipulative toys has been observed closely enough to determine the
mechanisms by which such play might enhance visual-spatial skills. The present study examines the
role of preschoolers' play preferences and behavior as correlates of two types of visual spatial skills.
This study goes beyond other research in that it includes in-depth observation of children's
Construction Play 6
involvement and skill in complex block play – both structured and unstructured. Children's
spontaneous levels of interest, creativity and complexity of play was assessed during a free play
session with blocks. Children's skills in block construction and visualization were assessed in a
structured block modeling task. Two types of visual spatial skills were assessed using standardized
tests which measured the spatial visualization component of spatial skills and perceptual field
independence or ability to break set. Spatial visualization was selected for two reasons: 1) tests
which can be used with preschoolers assess only spatial visualization as the other two components
appear at later ages (Voyer et al., 1995); and 2) these tests do not demonstrate consistent sex
differences, thus, within-gender differences can be more readily investigated. Although performance
on embedded figures involves some visual-spatial ability, it also measures field independence or
ability to break set, both of which are components of nonverbal creativity. Finally, verbal skills were
assessed to control for individual differences in verbal skills.
The goals of this study were two-fold: (a) to investigate the relation between preschoolers' play
preferences in a classroom setting and their performance on standardized visual-spatial tasks; and (b)
to analyze their free and structured play with blocks and its relation to spatial skill performance. The
following hypotheses were tested in this project: 1) Children's play preferences in preschool,
particularly play with manipulative toys, will be related to their visual spatial skills as assessed by
standardized tests; 2) Children's play with blocks (involvement and complexity of free play with
blocks; and complexity and correctness of structured/modeling block play) will be related to their
visual spatial skills as assessed by standardized tests.
Sixty children attending nine university-affiliated preschool classrooms participated in this
study. The children ranged in age from 47 to 69 months (M = 56.5 mo). Data from nine of the
children are not included in this report because the children did not complete all measures due to
moves from the area during the study. The final sample thus consisted of 26 girls and 25 boys.
Construction Play 7
There was no difference in the ages of the two sexes (girls M = 56.3 mos, boys M = 56.7 mos, t < 1.0,
ns). Most of the children were from middle-class homes, and they were predominantly European-
American in ethnic background.
Procedures and Measures
Play preferences. The toy and play activity preferences of the children were measured using
a play preference teacher rating form completed by the lead teacher in each child's classroom. The
form consisted of a listing of 18 play activities typically observed in preschool classrooms. Teachers
reported how frequently each child chose each of the activities during free play periods using a 5-point
Likert-type scale ranging from "almost always" to "almost never." Twelve of the 18 activities were
selected to represent three categories of play: art (easel painting, using magic markers, crayons and
coloring books, and cutting paper); fantasy play (playing house, playing hospital, giving a puppet
show, and playing grocery store); and manipulative toy play (jigsaw puzzles, parquetry blocks, Legos,
and blocks). Scores on these subsets were summed to obtain a preference rating for each child in
Observations of Block Play
On two occasions, children were taken to a playroom where they were videotaped at play with
standard preschool wooden blocks.
Unstructured block play. During the first block play session, children were presented with a
large box filled with 70 blocks comprising a variety of shapes and were asked to "Build the best thing
you can with these blocks." No other instructions were given, and the adult, who remained in the
room to monitor the child and the video camera, was involved in paperwork and instructed to minimize
interaction with the child. The unstructured block play sessions lasted approximately 15 min, but
could be terminated earlier if the child wished.
Construction Play 8
The videotapes of the unstructured block play sessions were later coded for the complexity
and competency of children's use of blocks and their degree of interest and involvement. The total
time children spent in play was also recorded. Two observers scored the tapes and both scored 12 of
the tapes for reliability. The coding definitions and inter-observer agreements for each coding
category are shown in Table 1.
Insert Table 1 about here.
Structured block play. On a different day, children were taken to the same playroom and
shown a complex block model, illustrated in Figure 1, which they were asked to duplicate, using the
box of blocks that had been provided for unstructured play. Again, the adult remained in the room and
videotaped the session but engaged in minimal interaction. The structured block play sessions lasted
approximately 15 min and could be terminated earlier by the child.
Insert Figure 1 about here.
The videotapes of the structured block play sessions were coded for accuracy by the same
two coders who scored the unstructured play sessions. Each child received a score for each block
placed; a correct placement received a score of 2, a placement that was approximately correct a
score of 1, and an incorrect placement a score of 0. The child's total score was used as the primary
measure of accuracy; scores could range from 0 to 86.
In addition, the block model was partitioned into 6 sections, and the correct placement of
sections with regard to the entire model was scored from slides taken of the child's final structure. For
each section of the model, a child received a score of 3 for placement between the correct two
adjacent sections, a score of 2 for correct placement next to one adjoining section, a score of 1 for
placement of the section in the approximately correct position in the structure, or a score of 0 if the
section was not constructed or placed entirely incorrectly. A child's total score on this scale could
Construction Play 9
range from 0 to 18. The number of sections that were correctly constructed was also counted.
Finally, children's efforts to rebuild collapsed portions of the structure were counted. Interobserver
agreement, calculated on 25% of the structured block play sessions, averaged 85% accuracy of
individual block placements.
Visual-spatial skills were evaluated using three different standardized measures. Children's
verbal skills were also evaluated in order to control for general intellectual ability. Each test was
administered on a separate day in a testing room near the child's preschool classroom. The order in
which tests were administered was varied randomly.
Block Design. The Block Design (BD) subtest from the Wechsler Preschool and Primary
Scale of Intelligence (WPPSI; Wechsler, 1967) and the Wechsler Intelligence Scale for Children -
Revised (WISC-R; Wechsler, 1974) were administered to measure the children's ability to analyze
and reproduce abstract forms. These tests assess spatial visualization, but also contain some
mental rotation components. The WPPSI Block Design subtest uses flat (2-dimensional) red-and-
white blocks and consists of 10 items in which the child is asked to reproduce either a model
constructed by the examiner or a pictured design. The 11-item WISC-R Block Design subtest uses
red-and-white cubes and children are asked to reproduce patterns of increasing complexity.
All children began the block design tasks with the first WPPSI item and continued until they had two
consecutive failures. Both the WPPSI and the WISC-R tasks were used to ensure that all children
reached a ceiling of performance. The highest possible score on the combined tests was 44. The
block design subtests from the WPPSI and WISC-R are highly reliable, with test-retest correlations
reported at .82 and .85 respectively (Wechsler, 1967, 1974).
Embedded Figures Test. The Children's Embedded Figures Test (CEFT; Karp & Konstadt,
1971) was administered to assess children's ability to abstract a geometric figure embedded within a
more complex picture. This test also measures spatial visualization. The test consists of 11 items in
which children locate a triangle of specific size and 14 items in which children locate a 5-sided
"house"; each series is preceded by two practice trials. Children were given a laminated paper shape
Construction Play 10
to place on the embedded shape in the picture. The test was discontinued when children failed 5
consecutive items. The highest possible score was 25.
Test-retest reliability for the CEFT has been reported to range from .83 to .90 (Karp &
Kondstadt, 1971). Construct validity has been evaluated with older children through correlations with
scores on the adult version of the Embedded Figures Test (r = .85 for 11-year-old children) and with
visual-spatial subtests of the WISC (statistically significant r values ranging from .32 to .49; Witkin et
Copying Blocks. The block portion of the Copying subscale from the Stanford-Binet
Intelligence Scale, 4th edition (Thorndike, Hagen, & Sattler, 1986) was used to measure visual-motor
ability (CB). The CB contains aspects of spatial visualization as well as mental rotation. The test
consists of 12 items in which children are asked to reproduce 3-dimensional structures using solid-
color cubes. The test is discontinued after 4 consecutive failures, and the highest possible score was
12. Test-retest reliability for the entire copying subtest (which includes paper-and-pencil tasks that
were not administered in the present study) is reported at .87 (Thorndike et al., 1986).
Peabody Picture Vocabulary Test- Revised. The Peabody Picture Vocabulary Test-Revised
(PPVT-R; Dunn & Dunn, 1981) was used to measure the verbal abilities of the children. The test
consisted of 175 items that assessed the child's receptive vocabulary. The children were shown
groups of four pictures and then were instructed to show what was asked. The reliabilities for the test
range from .71 to .89 (Dunn & Dunn, 1981). The children's scores on this measure did not differ by
gender. Males and females scored similarly on their verbal abilities (male, M = 55.43, s.d. = 20.57 ;
female, M = 58.35, s.d. = 17.53).
All scores were submitted to multivariate analysis of covariance (MANCOVA) in order to test
for sex differences in play activities, unstructured and structured block play and standardized test.
Verbal ability (score on the PPVT-R) was used as the covariate in all analyses.
Construction Play 11
Play preferences. A 2 (sex) X 2 (age) multivariate analysis of covariance (MANCOVA)
controlling for score on the PPVT was carried out to test for sex differences in children's play
preferences as rated by their preschool teachers. Girls and boys were rated as having significantly
different preferences, F(3, 52) = 19.43, p<.001. Univariate tests indicated significant sex differences
in fantasy (boys M = 2.38, s.d. = .52; girls M = 2.98, s.d. = .63,F (1, 54) = 20.64; p<.001) and
manipulative toy play (girls M = 2.40, s.d. = .52; boys M = 3.23, s.d. = .50, F (1, 54) = 36.79, p<.001)
but no difference in preference for art activities (girls M = 2.20, s.d. = .46; boys M 2.40, s.d. = .45, F (1,
54) = 3.49, n.s.).
Unstructured block play. Separate MANCOVAs with the score on the PPVT as a covariate
were used to test for sex differences in building strategies, types of blocks used, quality of
construction play, and time spent in free block play (see Table 2). Boys and girls differed only on the
set of variables describing number and types of blocks used, F(5, 42) = 2.77, p<.05. Univariate
results indicated that girls were more likely than boys to use the "unique" or unusual-shaped blocks,
and boys were more likely than girls to build more than one structure during their free block play time.
There were no overall differences between boys and girls in the frequency with which they used
building strategies, F(7, 40) = 1.12, ns; the qualitative ratings of their play, F(3, 44) = 1.61, ns; or the
amount of time spent in block play, F(1, 46) = 2.89, ns.
Insert Table 2 about here.
Structured block play. A MANCOVA was run controlling for the score on the PPVT on the set
of structured block play variables that measured the accuracy with which children reproduced the
block model. No sex differences were found, F(5, 41) = 1.41, ns. (Correct blocks, girls M = 42.23, s.d.
= 25.12, boys M = 38.58, s.d. = 25.08; correct sections, girls M = 8.42, s.d. = 6.37, boys M = 8.08, s.d.
= 7.14; number of sections completed, girls M = 4.12, s.d. = 1.95, boys M = 3.50, s.d. = 2.15; rebuilds,
girls M = 1.50, s.d. = 2.00, boys M = 1.38, s.d. = 1.93.)
Construction Play 12
Spatial tests. A MANCOVA with the score on the PPVT as a covariate was carried out to test
for sex differences in children's performance on the three spatial tasks presented to them. There
were no overall differences between boys and girls, F(3, 52) = 0.86, ns. (Block Design, girls M =
18.41, s.d. = 13.04, boys M = 19.39, s.d. = 13.63; Copying Blocks, girls M = 8.06, s.d. = 2.47, boys M
= 7.36, s.d. = 2.70; Embedded Figures, girls M = 8.84, s.d. = 4.88, boys M = 9.29, s.d. = 5.77.)
Play Preferences, Block Play, and Spatial Skills
Variables used in analyses. A series of regression analyses was carried out to examine
associations among children's play preferences, their skill at unstructured and structured block play,
and their performance on spatial tasks. Because a large number of variables was obtained from the
unstructured play session, factor analysis was used to create clusters of variables representing
dimensions of children's play behavior. As shown in Table 3, three factors were found to account for
more than 50% of the variance. The first factor, labeled complexity of play, included use of block
interfaces, rotating and flipping blocks before placing them in a structure, and symmetrical placements
of blocks. A second factor, labeled interest/involvement, included total number of structural forms
used, the qualitative ratings of children's play, and the number of times children rebuilt structures that
collapsed. The third factor, labeled number and variety of blocks used, included the number of
different shaped blocks children used, use of unique-shaped blocks, and the total number of blocks
used. A composite variable was used in the regression equations to index each of these components
of children's free block play.
Performance on spatial tasks (dependent variable) was measured by the childrens’
performance on replicating a block model ( number of correct block placements and number of
correct completions of model sections); and on standardized subscales (block design, copying blocks,
and embedded figures).
Insert Table 3 about here.
Construction Play 13
For each set of analyses, age was entered first into the regression equation. Predictor
variables were then entered as a block to evaluate the significance of the relation for each set of
variables. Each analysis with a statistically significant R-squared was followed up by a series of
hierarchical analyses in which each independent variable was entered individually, in a stepwise
fashion, to evaluate the increase in R-squared attributable to each variable.
Play preferences and block play. The first set of regression analyses examined the
associations among children's play preferences and the sets of variables used as indices of skill at
unstructured and structured block play, and the relation between unstructured and structured play with
blocks. Significant results are shown in Table 4. Children's play preferences showed no relation to
the unstructured block play variables: complexity of play, F(4, 46) = 1.51; quantity and variety of
blocks used, F(4, 46) = 0.86. Preference for art activities and the complexity of play of children's free
play with blocks predicted their accuracy in structured block play.
Insert Table 4 about here.
Play preferences and spatial skills. The results of the regression analyses examining the role
of play preferences in spatial test performance are shown in Table 5. A preference for play with art
materials was associated with high performance on the BD test, and a trend toward a relation between
art and the CB test was found. A preference for fantasy play was significantly associated with
performance on the CB test. Manipulative toy play preference was not related to performance on any
of the spatial tests.
Insert Table 5 about here.
Block play and spatial skills. The results of the analyses relating children's skill at block play
and their performance on visual-spatial tests are shown in Table 6. The free block play variable
indexing complexity of play was a significant predictor of scores on the BD test and marginally related
Construction Play 14
to performance on the CEFT. The variable indexing interest and involvement in free block play
predicted CEFT scores. None of the unstructured block play variables was related to the CB task, but
accuracy of block placement during the structured block play session was strongly related to
performance on the BD and CB tests. Accuracy in reproducing the block model did not relate to the
Insert Table 6 about here.
The two main purposes of this study were to examine the relation between preschoolers' play
preferences in a classroom setting and performance on visual-spatial tasks, and the relation of play
with blocks and to visual-spatial skills. Previous research has typically found a difference between
girls and boys, favoring boys, in spatial ability. This difference has been attributed in part to
differences in play experiences of girls and boys. This study examined the play preferences and
spatial skills of preschoolers in an effort to provide further understanding of the role of experience in
this sex difference and in individual differences in visual spatial skills within sexes.
There were few differences between girls and boys on the measures used. Boys were rated
as preferring manipulative toys more than girls did. During play with blocks, girls and boys differed
only on two dimensions. Girls used more unique shapes in their structures and boys built more
structures, but these aspects of their play were not associated with visual spatial skills. Girls and boys
did not differ in performance on any of the visual-spatial tests. These findings are consistent with
previous research suggesting that girls and boys play with different toys, and that children's spatial
skills do not differ by gender in the preschool years, at lease as measured by psychometric tests; thus
further assessing the antecedents of spatial skills in future research seems warranted.
The prediction that children's play with manipulative toys would be related to their spatial
abilities was not supported. Play with manipulative toys was not related to any of the measures of
spatial skills either during block play or in the standardized tests. This finding is in contrast to Connor
Construction Play 15
and Serbin's (1977) finding that play in masculine activities, which included manipulative toys, was
positively related to spatial ability, but is consistent with Newcombe 's (1993) results showing that play
with masculine toys was not related to spatial ability. The Newcombe study and the present
investigation used different methods of classifying toys than Connor and Serbin’s study did. Connor
and Serbin classified their toys by sex-stereotype; Newcombe classified toys and play activities as
spatial, non-spatial, and rated by sex-stereotype; the present study classified toys by ability
requirements. Perhaps if individual types of toys and activities are compared individually across
studies, more similarities will be found.
Play with art materials, reproduction of a complex block structure, and performance on tests
of spatial visualization were interrelated. Although art was not expected to predict spatial skills, this
findings is consistent with Newcombe ’s (1993) investigation showing that participation in sex-neutral
activities, including art, was positively related to spatial ability. As art is generally considered sex-
neutral, sex-stereotyping of the play activity does not appear to be a relevant mediator of its
associations with spatial skills.
The common elements in art, reproducing a block structure, and the two tests of spatial
visualization appear to be skills in analyzing, visualizing, and visual-motor coordination. Engagement
in art activities may provide preschoolers with an opportunity to practice analyzing visually and
reproducing objects, which is a skill required by the block design test (BD), and experience with the
visual-motor coordination needed for the copying blocks test (CB). These findings are consistent with
the notion that practice with art materials develops eye-hand coordination and the ability to observe
the environment in more detail and thus to be better at reproducing observable designs.
Children’s performance with the blocks in free play and their interest and involvement
predicted performance on the embedded figures test, but not the tests of spatial visualization.
Performance on the CEFT requires the ability to abstract figure from ground, to break set, and to be
field independent. Children who created a variety of block forms, who demonstrated planning and
interest in building their free play structure appear to have been practicing skills involved in breaking
Construction Play 16
set and fluent production required to find the embedded shape in the CEFT. The common skills in
these activities are those often defined as creativity.
Even the patterns that were demonstrated here do not, of course, demonstrate a causal
relation. They are consistent with the hypothesis that experience in play activities influences children’s
visual-spatial skills, but an alternative interpretation is equally plausible. Individual differences in
visual-spatial ability or in breaking set that contribute to the observed correlations could result from
genetic differences or early experience or both. Nonetheless, it is important to understand how such
skills are manifested in play, and how different aspects of play and performance are configured.
In summary, the hypothesis that preference for manipulative play activities predicts visual
spatial ability was not supported. Instead, two patterns emerged. One of these appears to indicate a
configuration of activity and performance representing skills in spatial visualization and visual-motor
coordination. It consists of preference for art activities, skill at reproducing a complex block model,
and skill on two tests of spatial visualization. The second appears to represent some components of
creativity – ability to break set and to produce varied solutions using visual materials. It is represented
by interest and involvement in free play block construction, complex and varied structures in free play,
and skill on an embedded figures task that requires the child to break set and do perceptual
reorganization of a visual stimulus. Future research might profitably examine the extent to which
children’s play activities and experiences predict or enhance these two types of skills.
Construction Play 17
Baenninger, M., & Newcombe, N. (1989). The role of experience in spatial test performance: A
meta-analysis. Sex Roles, 20, 327-344.
Blakemore, J. E. O., La Rue, A. A., & Olejnik, A. B. (1979). Sex-appropriate toy preferences and
the ability to conceptualize toys as sex-role related. Developmental Psychology, 15(3), 339-340.
Burns, C. W., & Reynolds, C. R. (1988). Patterns of sex differences in children's information
processing with and without independence from g. The Journal of School Psychology, 26, 233-242.
Casey, M. B., & Brabeck, M. M. (1989). Exceptions to the male advantage on spatial tasks:
Family handedness and college major as factors identifying women who excel. Neuropsychologia, 27,
Casey, M. B., Pezaris, E., Nuttall, R. L. (1992). Spatial ability as a predictor of math
achievement: The importance of sex and handedness patterns. Neuropsychologia, 30, 35-45.
Connor, J. M., Schackman, M., & Serbin, L. A. (1978). Sex-related differences in response to
practice on a visual-spatial test and generalization to a related test. Child Development, 49, 24-29.
Connor, J. M., & Serbin, L. A. (1977). Behaviorally based masculine- and
feminine-activity-preference scales for preschoolers: Correlates with other classroom behaviors and
cognitive tests. Child Development, 48, 1411-1416.
Connor, J. M., Serbin, L. A., & Schackman, M. (1977). Sex differences in children's response to
training on a visual-spatial test. Developmental Psychology, 13(3), 293-294.
Dunn, l. M., & Dunn, L. M. (1981). Peabody Picture Vocabulary Test-Revised. Circle Pines, MN:
American Guidance Service.
Fagot, B. (1977). Consequences of moderate cross-gender behavior in preschool children.
Child Development, 48, 902-907.
Fagot, B. (1978). The influence of sex child on parental reactions to toddler children. Child
Development, 49, 459-465.
Fagot, B., & Littman, I. (1976). Relation of preschool sex-typing to intellectual performance in
elementary school. Psychological Reports, 39, 699-704.
Construction Play 18
Geary, D. C., Gilger, J. W., & Elliott-Miller, B. (1992). Gender differences in three-dimensional
mental rotation: A replication. The Journal of Genetic Psychology, 153, 115-117.
Harris, L. J. (1981). Sex-related variations in spatial skill. In S. Liben, A. H. Patterson, & N.
Newcombe (Eds.), Spatial representation and behavior across the life span. New York: Academic
Huttenlocher, J., & Levine, S. C. (1993). Unpublished test and data. Department of Psychology,
University of Chicago, IL.
Johnson, E. S., & Meade, A. C. (1987). Developmental patterns of spatial ability: An early sex
difference. Child Development, 58, 725-740.
Karp, S. A., & Konstadt, N. (1971). Children's embedded figures test. Palo Alto, CA: Consulting
Psychologists Press, Inc.
Kerns, K. A., & Berenbaum, S. A. (1991). Sex differences in spatial ability in children. Behavioral
Genetics, 21, 383-396.
Kimura, D., & Hampson, E. (1994). Cognitive pattern in men and women is infleunced by
flucuations in sex hormones. Current Directions in Psychological Science, 3, 57-61.
Lewis, R. S., & Harris, L. L. (1988). The relationship between cerebral lateralization and
cognitive ability: Suggested criteria for empirical tests. Brain & Cognition, 4, 122-145.
Linn, M. C., & Petersen, A. C. (1985). Emergence and characterization of sex differences in
spatial ability: A meta-analysis. Child Development, 56, 1479-1498.
Maccoby, E. E., & Jacklin, C. (1974). The psychology of sex differences. Stanford: Stanford
McGee, M. G. (1979). Human spatial abilities: Psychometric studies and environmental,
genetic, hormonal, and neurological influences. Psychological Bulletin, 86, 889-918.
Munroe, R. L., & Munroe, R. H. (1971). Effect of environmental experience on spatial ability in
an East African society. The Journal of Social Psychology, 83, 15-22.
Nerlove, S. B., Munroe, R. H., & Munroe, R. L. (1971). Effect of environmental experience on
spatial ability: A replication. The Journal of Social Psychology, 84, 3-10.
Construction Play 19
Newcombe , N. (1982). Sex-related differences in spatial ability: Problems and gaps in current
approaches. In M. Potegal (Ed.), Spatial abilities. New York: Academic Press.
Newcombe, N. (1993). The relation between preschoolers’ everyday activities and spatial
abilities. Poster presented at the Society for Research in Child Development, New Orleans, LA, April.
Newcombe , N., & Bandura, A. M. (1983). Effect of age at puberty on spatial ability in girls: A
question of mechanism. Developmental Psychology, 19, 215-224.
Resnick, S. M., Gottesman, I. I., Berenbaum, S. A., & Bouchard, T. J. (1986). Early hormonal
inflluences on congenital adrenal hyperplasia. Developmental Psychology, 22(2), 191-198.
Serbin, L. A., & Connor, J. M. (1979). Sex-typing of children's play preferences and patterns of
cognitive performance. The Journal of Genetic Psychology, 134, 315-316.
Sherman, J. A. (1967). Problem of sex differences in space perception and aspects of
intellectual functioning. Psychological Review, 74, 830-836.
Sprafkin, C. Serbin, L. A., Denier, C., & Connor, J. M. (1983). Sex-differentiated play: Cognitive
consequences and early interventions. In M. B. Liss (Ed.), Social and cognitive skills: Sex roles and
children's play. New York: Academic Press.
Tracy, D. M. (1987). Toys, spatial ability, and science and mathematics achievement: Are they
related? Sex Roles, 17(3-4), 115-138.
Tracy, D. M. (1990). Toy-playing behavior, sex-role orientation, spatial ability, and science
achievement. Journal of Research in Science Teaching, 27(7), 637-649.
Voyer, D., & Bryden, M. P. (1990). Mental rotation: A study of gender differences and sex role
identity. Canadian Psychology, 21(2a), 253.
Voyer, D., Voyer, S., & Bryden, M. P. (1995). Magnitude of sex differences in spatial abilities: A
meta-analysis and consideration of critical variables. Psychological Bulletin, 117(2), 250-270.
Wechsler, D. (1967). Wechsler Preschool and Primary Scale of Intelligence (WPPSI). New
York: Psychological Corporation;Harcourt, Brace, Jovonovich, Inc.
Wechsler, D. (1974). Wechsler Intelligence Scale for Children - Revised (WISC-R). New York:
Psychological Corporation; Harcourt, Brace, Jovonovich, Inc.
Construction Play 20
Witkin, H. S., Oltman, P. K., Raskin, E., & Karp, S. A. (1971). A manual for the Embedded
Figures Tests. Palo Alto, Calif.: Consulting Psychologists Press.
Table 1 Unstructured block play codes and inter-observer agreements.
placements Placing the same shaped block in a design, such as at both ends of a long block, or re-
same structure a second time
Rotations/flips Rotating or turning blocks for the purpose of making interfaces, symmetrical placements, or
Adjustments Moving a previously-placed block to another location on the structure
Trial placements Manipulating a new block as if to place it on the structure, actually touching the block to the stru
but not actually placing the block; or completely removing a previously-placed block
Saves Child acts to prevent a collapse
Interfaces Placement of two complementary blocks together, such as combining an arch with a semicircle or
two ramps together to form a rectangle
Collapses Block or structure falls or is moved out of place
TYPES OF BLOCKS USED
Total blocks Total number of different blocks placed
Different shapes used Total number of different block shapes
Unique shapes Number of unusually-shaped blocks used
Number of structures Number of different constructions built
Number of different
block forms Of 16 possible types of constructions (e.g., stacks, arches, enclosures)
QUALITY OF PLAY RATINGS (Agreement within one scalepoint)
Amount of planning in construction
Complexity of final structure
Construction Play 21
Means for unstructured block play variables for girls and boys
M SD M SD F
Building Strategies 1.63
Symmetrical placements 3.8 3.1 1.8 2.7 ---
Rotation/flips 1.5 2.0 0.6 1.0 ---
Adjustments 5.7 3.7 6.0 3.6 ---
Trial placements 3.1 2.6 2.6 3.3 ---
Saves 5.9 3.6 7.8 6.2 ---
Collapses 7.0 5.3 5.3 3.6 ---
Interfaces 1.8 1.9 1.3 2.4 ---
Types of Blocks Used 3.26*
Total blocks 58.9 19.3 51.2 23.7 1.61
Different shapes 8.9 1.9 8.3 2.5 0.96
Unique shapes 4.7 2.4 3.1 2.3 5.63*
No. of structures 1.1 0.3 1.5 0.9 5.15*
No. of block forms 7.8 1.8 6.9 2.4 2.40
Quality of Play Ratings 0.70
Interest 3.7 1.1 3.4 1.2 ---
Planning 3.3 1.2 2.8 1.5 ---
Complexity 3.1 1.3 2.7 1.4 ---
Time Spent in Play (Min.) 16.4 0.7 15.5 2.6 2.76
Construction Play 25
Factor loadings of variables from the unstructured block play session.
Factor 1 Factor 2 Factor 3
Complexity of Play Interest/Involvement Number and Variety of Blocks
Interfaces .89 No. block forms .76 Shapes .85
Rotations/flip .78 Complexity rating .76 Unique shapes .81
Symmetric placements .44 Rebuilds .69 No. of block .78
Planning rating .52
Interest rating .50
Construction Play 26
Preschoolers’s play preferences and free block play behavior as predictors of skill
Dependent Variables: Accuracy Blocks Accuracy Sections
Beta R2 Beta R2
Age .37*** .13 .31** .10
Art .26* .19 .36*** .22
Manipulative ns ns
Fantasy ns ns
Unstructured Block Play
Complexity of Play .29** .22 .23* .16
Interest/Involvement ns ns
Quantity and variety ns ns
*p<.10. **p<.05. ***p<.01.
Construction Play 27
Preschoolers' play preferences as predictors of spatial skills.
Block Design Copying Blocks Embedded Figures
Dependent Variables: Beta R2 Beta R2 Beta R2
Age .45*** .21 .47*** .22 .46*** .21
Art .33** .31 .23* .26 ns
Fantasy ns .34** .28 ns
Manipulative ns ns ns
*p<.10. **p<.05. ***p<.01.
Construction Play 28
Preschoolers' free block play and skill at copying a block model as predictors of spatial
Block Design Copying Blocks Embedded Figures
Dependent Variables: Beta R2 Beta R2 Beta R2
Age .45*** .21 .47*** .27 .46*** .21
Free Block Play
Complexity of Play .33*** .32 ns .24* .27
Interest/Involvement ns ns .39*** .38
Number & Variety ns ns ns
Structured Block Play
Accuracy Blo c ks .45*** .38 .37*** .34 ns
Accuracy S ections ns ns ns
*p<.10. **p<.05. ***p<.01.
Figure 1. Block model.