MIND, BRAIN, AND EDUCATION
Block Talk: Spatial Language
During Block Play
Katrina Ferrara1, Kathy Hirsh-Pasek1,NoraS.Newcombe
1, Roberta Michnick Golinkoff
Wendy Shallcross Lam1
ABSTRACT—Spatial skills are a central component of intel-
lect and show marked individual differences. There is evidence
that variations in the spatial language young children hear,
which directs their attention to important aspects of the
spatial environment, may be one of the mechanisms that con-
tributes to these differences. To investigate how play affects
variations in language, parents and children were assigned to
1 of 3 conditions: free play with blocks, guided play, or play
with preassembled structures (Study 1). Parents in the guided
play condition produced signiﬁcantly higher proportions of
spatial talk than parents in the other two conditions, and
children in the guided play condition produced signiﬁcantly
more spatial talk than those in the free play condition. Study 2
established baselines of spatial language during activities not
involving spatial materials. Proportions of spatial words were
lower than those in any of the conditions of Experiment 1. In
sum, interaction with blocks naturally elicits elevated levels
of spatial language, especially in the context of guided play,
suggesting simple-to-execute educational interventions.
Spatial skills are a crucial component of human intellect. They
allow us to encode information about small and large-scale
objects—such as the location of our watch under a book, or
which way to turn to reach a destination. They also allow
us to mentally transform this information, such as imagining
what we might see if approaching an intersection from an
alternative direction. Spatial skills provide a foundation for
learning in the disciplines of science, technology, engineering,
and mathematics (STEM) (Burnett, Lane, & Dratt, 1979;
Casey, Nuttall, & Pezaris, 1997; Casey, Nuttall, Pezaris, &
Benbow, 1995; Tracy, 1987). For example, mental rotation and
1Department of Psychology, Temple University
2School of Education and DepartmentsofPsychologyandLinguisticsand
Cognitive Science, University of Delaware
Address correspondence to Katrina Ferrara, Johns Hopkins University,
Room 237 Krieger Hall, 3400North Charles Street, Baltimore, MD 21218;
spatial visualization are related to geometric problem solving
in high school (Battista, 1990; Delgado & Prieto, 2004; Kersh,
Casey, & Mercer Young, 2008), to mathematics achievement
(Fennema & Tarte, 1985; Guay & McDaniel, 1977; Hegarty &
Kozhenikov, 1999), and to success in chemistry (Stieff, 2007;
Wu & Sha, 2004).
How do spatial skills develop? One important answer may
lie in the relationship between human spatial cognition and
the symbol systems we use to describe spatial concepts.
In particular, the representational system afforded by
spatial language may provide an accessible introduction
to spatial concepts, such as the relationship between
objects, as illustrated by words like under and next to.By
directing children’s attention to spatially relevant aspects of
their environment, language highlights patterns that might
otherwise go unnoticed, for example, how one block is
situated under another is a tower. This spatial language
offers a categorical label that emphasizes qualitative divisions
in what is otherwise continuous space. As such, spatial
language might support spatial reasoning ability. The role
of vocabulary as a guide for future behavior and learning
has already been demonstrated in the area of literacy
(Christie & Enz, 1992; Christie & Roskos, 2006). With
regard to early spatial development, Casasola (2008) suggests
that, as infants acquire spatial terms, they form more
perceptually diverse spatial categories. In addition, individual
differences in children’s spatial language production predicts
performance on a variety of spatial skills assessments (Pruden,
Levine, & Huttenlocher, 2010). Gentner and colleagues
(Gentner, 2003; Gentner & Loewenstein, 2002) suggest that
spatial vocabulary may prove central for developing spatial-
relational understanding; ‘‘...relational language fosters the
development of representational structures that facilitate
mental processing—that is, that relational language provides
tools for thought (p. 316).’’
Despite its relevance to the development of spatial skills,
little is known about the contexts in which children may
be exposed to rich spatial language, or the settings in which
they are prone to use spatial language on their own. Research
suggests that the amount of exposure to different words
2011 the Authors
Volume 5—Number 3 Journal Compilation 2011 International Mind, Brain, and EducationSocietyandBlackwellPublishing,Inc. 143
predicts vocabulary development (Hart & Risley, 1995; Hoff,
2006; Huttenlocher, Haight, Bryk, Seltzer, & Lyons, 1991),
especially when the words are used in a way that helps the
child understand their meaning (Weizman & Snow, 2001).
But when are spatial words used, and in what contexts?
Block play is one common spatial activity in which spatial
language might naturally occur. Blocks have been frequently
mentioned as contributing to the development of spatial skills
(Brosnan, 1998; Caldera et al., 1999; Ginsburg, 2007; Ness &
Farenga, 2007). During the second and third years of life,
children pile blocks on top of one another (Shutts, Ornkloo,
von Hofsten, Keen, & Spelke, 2009). As their play becomes
moresophisticated, children pay specialattention to the colors,
shapes, and sizes of blocks. They may also compare the relative
sizes of the towers they create (Leeb-Lundberg, 1996). Reifel
(1984) suggests that blocks allow children to play directly with
spatial concepts, which in turn could assist their developing
representations of spatial relationships between objects in the
physical world (e.g., into, out, together, on top, beside, etc.).
In an analyses of open-ended forms of block play, researchers
concluded that the inherent geometric properties of blocks
encourage logico-mathematical thinking in young children
(Kamii, Miyakawa, & Kato, 2004). A relationship has also
been found between 3- and 5- year-olds’ block building skill
and their spatial visualization abilities (Caldera et al., 1999).
Furthermore, Wolfgang, Stannard, and Jones (2003) identiﬁed
a signiﬁcant relationship between complex LEGO building
during preschool years and later achievement in middle and
high school mathematics. Most recently, in an experimental
study that incorporated a story-telling context, Casey et al.
(2008) concluded that block building interventions have
an impact on spatial visualization and block building
Review of the current literature highlights the point that
there are several kinds of block play. Children sometimes
engage in free play with blocks, but they may also strive
to copy a structure depicted on a box or follow step-by-
step instructions. Does the context of block play signiﬁcantly
impact the amount of spatial language that children and
parents are apt to use in joint play sessions? It seems likely
that block play will encourage the use of more spatial language
than simple free play, but free play with blocks may still
elicit more spatial language than playing with materials that
do not involve construction of any kind. To investigate this
question, in Experiment 1, we focused on three common
contexts of play. In the free play condition, parents and children
played with blocks without any form of guidance. In the
guided play condition, the parent and child were given ﬁve
numbered photographs that pictorially depicted the steps
to build a particular ﬁnal structure. In the preassembled
play condition, a glued-together model was given to the
pair, and the prefabricated structure served as a prop for
Thirty-six, 3- to 4.5-year-old children (18 females, mean age
46 months, range 36.5–69 months) and 36, 4.5- to 5-year-
old children (18 females, mean age 63, range 54–71 months)
participated, with either a parent or a guardian. Two of the
parents who participated in the study were fathers and 74 were
mothers. Data from 10 additional children had to be discarded
due to failure to complete the task (8) or experimental error
(2). Potential participants were identiﬁed through a direct
marketing list and were contacted by an introductory letter,
followed by a phone call. Additional participants responded
to advertisements or heard about the study through word
of mouth. As the area in which recruitment was targeted
is predominantly middle to upper-middle class Caucasian,
the majority of children who participated came from families
belonging to this demographic group.
A set of MegaBloks containing various sized blocks, as well
as vehicles and ﬁgures, was used. From these blocks, two
critical structures could be created (a garage or a helipad).
A full set contained 114 building blocks, 8 ﬂat block panels
that could create ground surfaces, 2 window-shaped blocks, a
lamp-shaped block, a ladder-shaped block, 4 ﬁgures depicting
different occupations (a ﬁreﬁghter, a pilot, and 2 civilians),
and 2 vehicles (a ﬁre truck and a helicopter). A video camera
set unobtrusively in the corner of the room, approximately 5 ft
from where parent and child were seated, was used to record
the session and to later transcribe the footage.
The study contained two 10-min phases. Participants were
randomly assigned to one of three conditions in Phase 1. In
the free play condition, parents and children were told to play
with the set of blocks as they would at home. In the guided
play condition, the parent and child were given ﬁve numbered
photographs depicting the steps to build either the garage
or the helipad (much like the instructions one receives for
IKEA furniture assembly). In Step 1, the picture illustrated
the placement of the board upon which to build. Step 2
portrayed the ﬁrst ground layer of blocks, and so forth, up to
Step 5 which showed the ﬁnished structure, complete with
placement of the ﬁgures around or in the building. Parents
and children were encouraged to build the structure using
the pictures for guidance. In the preassembled play condition, a
glued-together model of the garage or the helipad was given to
the dyad, as well as the vehicles and ﬁgures. No extra blocks
were provided for building. The dyad was instructed to play
as they naturally would. Thus, all three play conditions in
144 Volume 5—Number 3
Katrina Ferrara et al.
Phase 1 offered parents and children an opportunity to play
with blocks. However, each differed in terms of the structure
of the play situation. Children in each of the conditions had
access to the same number of blocks, ﬁgures, and vehicles. The
experimenter left the room during building times.
After 10 min of play, Phase 2 began. In this phase, all dyads
were assigned to the guided play condition. The purpose of
this manipulation was to assess whether spatial language is
affected positively or negatively by prior styles of interaction
with block materials. Participants were given a picture of the
structure that they had not played with or attempted to build
during Phase 1 (e.g., those who had built the heliport were
now asked to build a garage). The order of which structure
the groups built and played with ﬁrst was counterbalanced.
Parents were told that they could help, but that the goal was
for the child to build. After 10 min, the experimenter returned
and told the children that they could keep playing or stop.
All videotaped interactions were transcribed for Phases 1
and 2. The transcripts were analyzed for child and parental
spatial language using the spatial categories of the University
of Chicago spatial language coding system (Cannon, Levine,
& Huttenlocher, 2007). Speciﬁcally, coders identiﬁed terms
and phrases that described the following spatial categories:
(1) spatial locations (up, down), (2) deictic terms (here, there),
(3) dimension (long, tall), (4) spatial features or properties
(curvy,straight), (5) shapes (rectangle,square), and(6) spatial
orientations or transformations (‘‘turn it around,’’ ‘‘the man
is facing the block’’). Coders only identiﬁed words that were
spatial in context and avoided idioms and/or spatial terms
that were used in a metaphorical way (e.g., ‘‘that building is on
ﬁre,’’ ‘‘the doctor performs surgery on the patient’’). Repetitive
statements were only considered once, such that if the parent
of child repeated the same exact statement twice in a row,
only the ﬁrst would be included in the analysis.
For each building phase, the total number of individual
spatial words (e.g., down) and phrases (a collection of words
that describe a single spatial meaning, e.g., in the middle,on each
side) were tallied. A ratio of spatial terminology to non-spatial
talk was calculated for each parent–child dyad. The ratio is
Spatial words +Spatial phrases
Spatial words +Spatial phrases +Nonspatial ndividual words
To take into account the variation in verbosity across
different pairs, this ratio captures the proportion of total
utterances in each phase that were spatial in nature.
Spatial words and spatial phrases were independent of one
another—no spatial content was counted more than once.
Proportions of child and parental spatial language were
calculated and averaged separately for Phase 1 and Phase 2.
Three independent coders followed the criteria of the coding
system and coded transcripts from identical participants (7%
of the total transcripts included in the dataset) and were in
high agreement (r=.89) concerning the proportion of spatial
language present in both phases. A third coder second-coded
17% of the 72 transcripts and was in high agreement with both
primary coders (r=.87 and r=.95).
Results and Discussion
As is shown in Table 1, individuals varied a great deal in both
verbosity and use of spatial language. To elucidate the degree
to which block play in the three play contexts elicited spatial
language from parents, a repeated measures ANOVA was
performed. Spatial language in both phases (2) was the within-
subject variable, and condition (3), sex (2), and age (2) were
the between-subjects variables. The dependent variable was
the proportion of parent spatial language in Phases 1 and 2.
A signiﬁcant effect was found for condition, F(2, 60)=10.62,
p<.05, η2=.26, as well as a signiﬁcant phase by condition
interaction F(2, 60)=13.99, p<.05, η2=.32. Post hoc
comparisons (Tukey HSD) indicated that parents in the
guided play condition demonstrated signiﬁcantly higher
proportions of spatial talk than those in the free play (mean
difference =.025, p<.001), or preassembled conditions
(mean difference =.022, p<.001). The preassembled and
free play conditions showed no signiﬁcant difference from
one another (mean difference =.003, p=.866). To clarify
the main effect for play condition in Phase 1, independent
sample ttests were conducted. Parents in the guided play
condition (M=.096) demonstrated signiﬁcantly more spatial
language than parents in the free play condition (M=.057),
t(46)=−5.454, p<.05, or parents in the preassembled
condition (M=.057), t(46)=5.77, p<.05. No signiﬁcant
differences were found between parents in the free play and
preassembled play condition, t(46)=.062, p=.95.
An additional question was whether children’s spatial lan-
guage would differ with the condition assigned. An ANOVA
analogous to the one above was conducted with the propor-
tions of child spatial language in Phases 1 and 2. A signiﬁcant
main effect was found for condition, F(2, 60)=4.65, p<.05,
η2=0.90. Independent sample ttests showed that children
in the guided play condition (M=.067) demonstrated signif-
icantly more spatial language than children in the free play
condition (M=.046), t(45)=−2.13, p<.05. Just as with the
parents, no signiﬁcant differences were found between the pre-
assembled(M=.059) and free playconditions, t(46)=−1.86,
p=.07. However, children in the guided play condition did
not signiﬁcantly differ in their spatial talk from those in
Volume 5—Number 3 145
Mean, Standard Deviation, and Range of Total Word and Spatial
Word Counts for Parents and Children in Phases 1 and 2, Shown for
the Guided, Free Play, and Preassembled Conditions of Study 1
Phase 1 Total words 724(197) 415–1220
Spatial words 71(32) 27–145
Phase 2 Total words 638(188) 318–1025
Spatial words 58(22) 24–112
Phase 1 Total words 197(91) 32–428
Spatial words 14(9) 2–38
Phase 2 Total words 234(118) 84–441
Spatial words 15 2–34
Condition: Free play
Phase 1 Total words 496(165) 191–899
Spatial words 29(15) 11–69
Phase 2 Total words 578(188) 180–947
Spatial words 45(22) 12–95
Phase 1 Total words 252(81) 115–403
Spatial words 12(7) 4–29
Phase 2 Total words 231(81) 88–367
Spatial words 12(6) 1–26
Phase 1 Total words 567(272) 128–1164
Spatial words 32(16) 7–66
Phase 2 Total words 582(248) 207–1120
Spatial words 15(7) 3–28
Phase 1 Total words 253(95) 68–428
Spatial words 51(28) 8–102
Phase 2 Total words 240(78) 42–401
Spatial words 17(8) 2–29
the preassembled condition, t(45)=.88, p=.38. Post hoc
comparisons (Tukey HSD) indicated that children in the
free play condition demonstrated signiﬁcantly lower propor-
tions of spatial talk than those in the preassembled (mean
difference =−.017, p=.023) and guided conditions (mean
difference =−.016, p=.033). The difference between guided
play and preassembled conditions was not signiﬁcant (mean
difference =.0011, p=.98). Overall, the observed differences
in child spatial talk indicate the beginnings of a trend similar
to that demonstrated by parents, but does not signiﬁcantly
follow through to the assembled and guided play conditions.
Perhaps if the parameters of the study had been altered slightly,
such that the play session lasted for longer than 10 min, or if the
guided condition had contained more steps, children’s spatial
language would follow the pattern of parents in this respect.
To see if the condition assigned in Phase 1 bore a relationship
to the amount of spatial language demonstrated in Phase 2
(in which all dyads engaged in guided play), correlations
were computed across each of the three conditions, for
both parents and children. For each of the conditions
in Phase 1, the proportion of parent spatial language in
Phase 1 correlated with the same measure in Phase 2;
free play (r=.62, p<.001), guided (r=.66, p<.001), and
preassembled (r=.41, p<.05). This relationship was not
found for children in any of the conditions (rs<.15, ps>.05).
Proportions of parent and child spatial talk did not correlate
in the Phase 1 free play condition (r=.32, p>.05), or the
guided play condition (r=−.17, p>.05) but did correlate in
the preassembled condition (r=.49, p<.05).
In recognition of the possibility that a particular type of
spatial word category could be driving the observed results for
both parents and children, separate ANOVAs were conducted
to determine whether the use of certain kinds of spatial
word categories (as deﬁned by the coding system, such as
deictic terms, location terms, or descriptions of orientation and
transformation) varied by condition. Spatial word type (6) was
the within-subject variable, and condition (3) and sex (2) were
the between-subjects variables. The dependent variable was
the proportion of spatial language as categorized by the spatial
word categories. For parents, a signiﬁcant effect was found for
condition, F(2, 60)=20.63, p<.001, η2=.385, as well as a
signiﬁcant phase by condition interaction F(2, 60)=3.35,
p<.001, η2=.09. Post hoc comparisons (Tukey HSD)
conﬁrmed that the instances of the spatial word categories
were signiﬁcantly greater in the guided play condition than the
free play (mean difference =7.54, p<.001) and preassembled
play (mean difference =6.50, p<.001) conditions; the
preassembled play and guided play conditions did not
signiﬁcantly differ from one another (mean difference =
−1.04, p=.70). For children, a signiﬁcant phase by condition
interaction was found, F(2, 60)=10.20, p<.001, η2=.236.
Tables 2 and 3 illustrate the means for each spatial category
for parents and children, for both Phases 1 and 2. As was found
in the analyses of overall proportions of spatial talk, children
do not show as dramatic an effect of condition as their parents.
Were children differentially engaged in play during these
three conditions? To answer this question, video footage was
coded for the amount of time 80% of the subjects spent
building, talking about, and generally interacting with the
block materials. On average, out of the 10-min session, children
spent 9 min, 37 s engaged with the blocks in Phase 1, and 9 min,
25 s in Phase 2. To asses whether condition had an effect
on levels on engagement, an ANOVA was conducted with
time spent playing in Phases 1 and 2 as the within-subjects
variable, gender (2) and condition (3) asthe between-subjects
variables. No signiﬁcant differences were found among the
conditions assigned in Phase 1 (Mfree play =9.75, Mguided =
9.81, Mpreassembled =9.43), F(2, 55)=0.43, p=.84, η2=.001.
In summary, the analyses demonstrate that play context
(free, guided, and preassembled) impacts the spatial
vocabulary that children are apt to hear. The strength of
146 Volume 5—Number 3
Katrina Ferrara et al.
Means for Spatial Word Categories Demonstrated by Parents in
Phases 1 and 2
Spatial word category Free play Guided Preassembled
Location 13.58 22.88 18.25
Deictic 6.67 17.59.58
Dimension 5.33 10.67 1.88
Feature/property 4.17 15.67 5.2
Shape 0.33 4.83 0.75
Location 17.75 24.17 19.71
Deictic 11.38 14.92 13.13
Dimension 4.71 7.25 5.29
Feature/property 12.58 13.42 13
Shape 1.04 1.79 1.04
the effect of construction contexts on parental talk is further
evidenced by the fact that no explicit instructions to use
spatial language were given. The guided play context in
particular elicited the most spatial language from parents,
perhaps because the dyad was most inﬂuenced by the shared
goal of the condition. For parents, the guided play condition
was also shown to elicit signiﬁcantly higher amounts of spatial
words per category. Child data indicated the beginnings of
this trend, in that the free play context contained signiﬁcantly
lower proportions of spatial talk; however the guided play and
preassembled play conditions did not differ from one another
in spatial proportions of speech or spatial word categories.
Study 1 leaves open the question of what baseline levels of
spatial language might be present when children and parents
interact in situations without blocks or other types of spatial
materials. We therefore sought to analyze the language of
Means for Spatial Word CategoriesDemonstratedbyChildrenin
Phases 1 and 2
Spatial word category Free play Guided Preassembled
Location 5.29 3.21 9.58
Deictic 3.08 5.63 6.04
Dimension 2.13 3.67 0.46
Feature/property 1.58 1.83 1.33
Shape 0.13 1.33 0.04
Orientation/transformation 1.75 0.75 2.25
Location 4.52 5.292 6.25
Deictic 3.75 5.36.67
Dimension 2.17 2.63 2.42
Feature/property 2.54 2.33 2.88
Shape 0.17 0.38 1.08
Orientation/transformation 1.58 1.08 1.38
parents and children during other everyday non-spatial play
activities. In order to capture a naturalistic and heterogeneous
array of language in diverse non-spatial contexts, Study 2 used
the CHILDES database to investigate the quantity of spatial
language demonstrated by parents and children in other types
of play and daily life activities that do not involve construction.
Recruitment Method and Materials
To remain consistent with the methods of Study 1, transcripts
from the CHILDES database were selected to ﬁt the following
criteria: (1) children were within the age parameters of Study 1
(3 through 5 years of age), (2) the transcribed interaction
was restricted to one caregiver speaking to one child (e.g.,
avoided group dialogue), and (3) the interaction involved
task(s) that did not involve play with construction toys.
Furthermore, when time information was included, 10-min
interactions were used to parallel the methods of Phase
1 in Study 1. If no time information was given, efforts
were made to closely match total word counts to those
demonstrated in Study 1. Ultimately, the collection of
transcripts represented the following activities (which were
not mutually exclusive across transcripts): lunch with parent,
play with puppets, drawing, playing house, playing store,
dressing up, playing ‘‘zoo’’ with animal ﬁgurines, pretending
to talk on a telephone, playing tea party with dolls, playing
with pretend food and kitchen utensils, playing ‘‘school,’’ and
throwing a ball.
Thirty-one transcripts were gathered and analyzed. In keeping
with the age groups of Study 1, 14 transcripts were obtained for
children approximately between the ages of 3 and 4.5 (6 males,
8 females, mean age 43 months, range 31–43 months) and 17
from children approximately between the ages of 4.5 and 5 (7
males, 10 females, mean age 63 months, range 54–71 months).
Procedure and Coding
Transcriptswereanalyzed according to the codingsystemused
in Study 1. Two reliable coders from Study 1 calculated the
proportions of parent and child spatial language demonstrated
in the CHILDES transcripts.
To compare the spatial content of the language that preschool
children hear from caregivers when participating in other
types of play (CHILDES control) to that of the language
elicited when playing with construction toys like blocks, a
one-way ANOVA was conducted. As illustrated in Figures 1
Volume 5—Number 3 147
and 2, signiﬁcant differences were found between the spatial
language proportions in Study 1 and the proportions calculated
from the set of CHILDES transcripts, F(1, 102)=55.02, p<
.001. A series of independent samples ttests were performed
to probe the difference between the block construction play
groupsand the general play group.Itwas found that in the three
block play contexts, parents used signiﬁcantly more spatial
language than those depicted in the CHILDES transcripts
(MCHILDES =.03, SE =.003): in the preassembled condition
(Mpreassembled =.057, SE =3.32), t(53)=6.29, p<.001, the
guided play condition (Mguided =.096, SE =6.46), t(53)=
10.59, p<.001, and the free play condition (Mfree play =.057,
SE =3.15), t(53)=5.71, p<.001.
Similarly, children themselves used spatial language
more frequently in the conditions of Study 1 than
those depicted in the CHILDES transcripts, F(1, 101)=
23.15, p<.001 (Figure 2). As with parents, independent
t-tests revealed that, when broken apart, each block play
condition elicited signiﬁcantly more child language than
the selected CHILDES activities (MCHILDES =.03, SE =.003):
preassembled (Mpreassembled =.059, SE =5.79), t(53)=5.65,
p<.001; guided (Mguided =.067, SE =1.79), t(52)=4.23, p<
.001; and free play (Mfree play =.047, SE =4.53), t(53)=3.00,
p=.005. These results lead us to conclude that introducing
blocks to a play context is likely to elicit conversation
containing a host of spatial vocabulary above and beyond
what is used in other types of play.
The purpose of these studies was to explore whether children’s
learning of spatial language might be enhanced in settings
Proportion of spatial language
Free play Guided Preassembled CHILDES
Fig. 1. Proportions of parent spatial language in Phases 1 and 2 as a
function of play condition group (Study 1), in comparison to parent
spatial language proportions demonstrated in selected CHILDES
transcripts (Study 2).
Free play Guided Preassembled CHILDES
Proportion of spatial language
Fig. 2. Proportions of child spatial language in Phases 1 and 2 as a
function of play condition group (Study 1), in comparison to child
spatial language proportions demonstrated in selected CHILDES
transcripts (Study 2).
involving play with blocks, paying particular attention to the
role of verbal descriptions of spatial concepts. The ﬁrst study
revealed that parents and children do use spatial language
in block play, and even more importantly, that different play
contexts were more or less supportive of the use of spatial
language. In particular, the guided play context in Study 1
promoted more spatial talk from parents than the other two
Child spatial language followed a similar trend, in that
children in the guided play condition demonstrated more
spatial language than those in the free play condition, but not
in the assembled condition. Based upon what was observed in
the play conditions, it seems that the preassembled structure
afforded play that involved different types of spatial relations
apart from those used in construction contexts, for example:
‘‘Mommy let’s drive the cars in and out of the garage,’’ or, ‘‘The
ﬁreman is climbing all the way up to the top of the building.’’
Parents and children often elicited and commented on spatial
actions with the ﬁgures and vehicles provided. In contrast, in
the free play condition, children would put blocks together
somewhat randomly and seemed to not pay as much attention
to the ﬁgures and vehicles. This was also observed in the guided
play condition at times. If parents played less of an assertive
role, children would sometimes became more absorbed with
the activity of building, rather than building in a way that
followed the prescribed steps (it should be noted that in the
guided play condition, no strict rules were imposed such that
the building would have to be ﬁnished by the end of the play
session.) When children became absorbed with the materials,
not as much conversation about spatial conﬁgurations was
elicited as in the preassembled condition. We hypothesize
that stretches of this type of play may have contributed to
the lack of a difference between proportions of children’s
148 Volume 5—Number 3
Katrina Ferrara et al.
spatial talk in the preassembled and guided play conditions.
It may be that with a slightly older age group of children
who have a wider range of conversational and block building
skills (perhaps 6–7 year olds), the guided play condition would
show signiﬁcantly higher proportions of spatial talk for both
parents and children alike.
The second study allowed us to conclude that ordinary,
everyday interactions of parents and preschoolers do not
necessarilyinvolve spatial talk toas great an extentasplay with
construction materials. In these forms of play, it is notable that
the mean proportions of spatial phrases and words were found
to be the same for parents and children alike (MCHILDES =.03).
Thus, it seems likely that block play confers an advantage for
children’s exposure to spatial language and encourages them
and their parents to use spatially relevant terms.
These ﬁndings hold substantial educational implication
for the enhancement of spatial instruction. Many current
organizations have recognized the need to bolster this area in
the classroom. In 2006, the National Research Council’s report
entitled Learning to Think Spatially, outlined the importance of
developing spatial skills not only for success in the STEM
disciplines, but also for normal functioning in everyday life.
Despite its fundamental role, spatial learning is not speciﬁcally
addressed in many preschool and kindergarten curriculums
(National Research Council, 2006). The Council’s report
speciﬁcally advocated for greater inclusion of direct spatial
ability training, and deﬁned spatial ability as the ability to
process, manipulate, and visualize spatial information. A
recent meta-analysis of the efﬁcacy of such spatial ability
training showed that it is malleable and can beneﬁt from
training (Hand, Uttal, Marulis, & Newcombe, 2008). As set
forth in their Geometry Standards for Pre-K to Grade 2, the
National Council of Teachers of Mathematics (2000) highlight
the following spatial reasoning objectives: Describing spatial
relationships (e.g., the ability to use words such as top and
bottom), and creating mental images of geometric shapes.
Blocks may be one such educational tool that provide
young children with an accessible and playful introduction to
these spatial concepts and abilities. It has been demonstrated
that words embedded in playful contexts are learned better
and faster (Neuman & Roskos, 1992). The current study
indicates that play with blocks in a semi-structured guided
play context, in the company of a more experienced partner,
is especially beneﬁcial for children’s exposure to spatial
language. The particular beneﬁts of a guided play approach
have additionally been demonstrated in children’s learning of
the properties of shapes (Fisher et al., 2009). This hearkens
back to Vygotsky’s classic theory of scaffolded learning within
the zone of proximal development, in which a novice’s learning
is facilitated by an expert instructor (Vygotsky, 1978). Block
play in the preschool years may be additionally appropriate as
an educational tool in that it coincides with observed trends
in language development. It is between the ages of 2 and 5 that
children begin to express and understand relational concepts
such as big/little, wide/narrow, tall/short, in/on, high/low,
and here/there (de Villiers & de Villiers, 1979, 1992). Around
the age of 3, children use these terms to make appropriate
judgments of function (Gelman & Ebeling, 1989). With the
guidance of an adult partner, children will be exposed to
new and more sophisticated forms of spatial language to add
to their growing lexicons, granting them opportunities to
elaborate and expand upon their developing knowledge of
spatial relations and corresponding categorical labels.
The fact that no gender differences were found in the
current study additionally speaks to the educational validity
of blocks as a tool in the classroom. Although prior research
has typiﬁed a greater male preference for play with blocks
(Farrell, 1957; Farwell, 1930; Saracho, 1994, 1995), it seems
that girls may be just as motivated to play with construction
materials. Indeed, for all of the conditions in the current study,
it was found that girls and boys did not differ from one another
in the amount of time spent engaging with the blocks. In
assessments of competency in block building skills, boys have
been shown to have no advantages over girls (see review by
Kersh et al., 2008). Thus, teachers may incorporate blocks
into their pedagogical techniques without overt concern for
alienating female members of the class.
The delivery of spatial language through block play may also
prove particularly powerful as an educational tool for children
of low-SES (socioeconomic status) households, who may face
speciﬁc contextual challenges in acquiring language (Case,
Grifﬁn, & Kelly, 2001; Whitehurst, 1997). It has been found
that although low- and high-SES children start out with the
same number of spatial words at 30 months of age, the linear
rate of growth of spatial words is slower for low-SES children
when compared to middle- and high-SES children (Pruden
et al., 2010). Block play may also prove to be a particularly
useful and accessible tool in introducing at-risk children to
fundamental mathematical concepts (Park, Chae, & Foulks,
2008), and address differences in mathematics achievement
scores(Jordan, Huttenlocher, & Levine,1992; Saxe, Guberman,
Blocks offer one play context that may enhance learning.
Here, we examined block play between parents and children,
and found that, in comparison to many other types of play
activities, blocks encourage parents and children to use
signiﬁcantly more spatial terms in conversations with one
another. The fact that the guided play condition elicits more
spatial language suggests that experimental and educational
interventions may follow such a model to increase the
frequency of spatial language children hear and come to
use on their own. Pre-K and kindergarten teachers may not
fully recognize the educational value of block play (National
Association for the Education of Young Children [NAEYC]
1997; Park, Chae, & Foulks, 2008; Wellhousen & Keoff, 2000;
Zacharos, Koliopoulos, Dokomaki, & Kassoumi, 2007). These
Volume 5—Number 3 149
ﬁnding bear direct relevance to implementation in classrooms,
in which a teacher may use goal-directed block play as a
means of introducing and acting out spatial concepts and
A recent meta-analysis (Alﬁeri, Brooks, Aldrich, &
Tenenbaum, 2010) of discovery-based learning speaks to this
issue. The authors deﬁne discovery learning as occurring when
‘‘the learner is not provided with the target information or
conceptual understanding and must ﬁnd it independently and
with only the provided materials....there is an opportunity
to provide the learners with intensive or conversely, minimal
guidance, and both types can take many forms’’ (p. 2). The
guided condition of this study ﬁts within this approach, in
which the end product of a complete structure is obtained
by sequential stages at which children receive guidance and
feedback from their parents as needed. Based upon their
ﬁndings in the meta-analysis, Alﬁeri et al. (2010) suggest that
pedagogical approaches that employ scaffolded tasks with
predeﬁned objectives confer particular beneﬁts for learners.
Teachers may adopt this strategy in the context of guided play
with blocks, providing children with explanation and support,
yet also allowing them the space to build and discover on
Because of the unique language it elicits, playing with
blocks may be one of the means by which young children
begin to develop the spatial abilities that have been found
to be linked to a number of academic achievements later in
life (Humphreys, Lubinski, & Yao, 1993; Shea, Lubinski, &
Benbow, 2001). These data are among the ﬁrst to show that
naturalistic interactions between parents and children can
build a foundation for important spatial concepts and the
means of expressing them through language. Future research
will further elucidate the way in which block play may be
utilized as a mode that fuses together playful learning and
Acknowledgments—This research was supported in part by
funding from the National Science Foundation Science of
Learning Center (Grant #SBE 0541957) to Nora S. Newcombe,
the National Institutes of Health Stimulus Grant (Grant
#1RC1HD0634970-01) to Roberta Michnick Golinkoff and
Kathy Hirsh-Pasek, as well as a gift from MEGA Brands.
We are grateful to the contributions to this work made by the
members of the Temple Infant Lab and the Research in Spatial
Alﬁeri, L., Brooks, P. J., Aldrich, N. J., & Tenenbaum, H. R. (2010).
Does discovery-based instruction enhance learning? Journal of
Educational Psychology, 80,424–436.
Battista, M. T. (1990). Spatial visualization and gender differences in
high school geometry. Journal for Research in Mathematics Education,
Brosnan, M. J. (1998). Spatial ability in children’s play with Lego
blocks. Perceptual and Motor Skills, 87,19–28.
Burnett, S. A., Lane, D. M., & Dratt, L. M. (1979). Spatial visualization
and sex differences in quantitative ability. Intelligence, 4,233–242.
Caldera, Y. M., C ulp, A. M., O’Brien, M., Truglio, R. T., Alvarez, M., &
Huston, A. C. (1999). Children’s play preferences, construction
play with blocks, and visual-spatial skills: Are they related?
International Journal of Behavioral Development, 23,855–872.
Cannon, J., Levine, S., & Huttenlocher, J. (2007). A system for analyzing
children and caregivers’ language about space in structured and
unstructured contexts. Spatial Intelligence and Learning Center
(SILC) technical report.
Case, R. Grifﬁn, S., & Kelly, W. M. (2001). Socioeconomic differences
in children’s early cognitive development and their readiness
for schooling. In S. L. Golbeck (Ed.) Psychological perspectives on
early childhood education: Reframing dilemmas in research and practice.
Mahwah, NJ: Erlbaum.
Casasola, M. (2008). The development of infants’ spatial categories.
Current Directions in Psychological Science, 17(1), 21–25.
Casey, B. M., Andrews, N., Schindler, H., Kersh, J. E., Samper, A.,
involving block building activities. Cognition and Instruction, 26,
Casey, B. M., Nuttall, R., Pezaris, E., & Benbow, C. P. (1995). The
inﬂuence of spatial ability on gender differences in mathematics
college entrance test scores across diverse samples. Developmental
Casey, M. B., Nuttall, R. L., & Pezaris, E. (1997). Mediators of
gender differences in mathematics college entrance test scores:
Acomparison of spatial skills with internalized beliefs and
anxieties. Developmental Psychology, 33, 669–680.
Christie, J. F., & Enz, B. (1992). The effects of literacy play
interventions on preschoolers’ play patterns and literacy
development. Early Education and Development, 3,205–225.
Christie, J., & Roskos, K. (2006). Standards, science, and the role of
play in early literacy education. In D. Singer, R. M. Golinkoff,
&K.Hirsh-Pasek(Eds.),Play = learning: How play motivates and
enhances children’s cognitive and social- emotional growth (pp. 57–73).
New York: Oxford University Press.
Delgado, A. R., & Prieto, G. (2004). Cognitive mediators and sex-
related differences in mathematics. Intelligence, 32(1), 25–32.
de Villiers, P. A., & de Villiers, J. G. (1979). Early Language. Cambridge,
MA: Harvard University Press.
de Villiers, P. A., & de Villiers, J. G. (1992). Language development. In
M. H. Bornstein & M. E. Lamb (Eds.), Developmental psychology:
An advanced textbook (3rd ed.). Hillsdale, NJ: Erlbaum.
Farrell, M. (1957). Gender differences in clock play in early childhood
education. Journal of Educational Research, 51,279–284.
Farwell, L. (1930). Reactions of kindergarten, ﬁrst-and-second grade
children to constructive play materials. Genetic Psychology
Monographs: Child Behavior, Animal Behavior, and Comparative
Fennema, E., & Tartre, L. A. (1985). The use of spatial visualization in
mathematics by girls and boys. Journal for Research in Mathematics
Fisher, K., Ferrara, K., Hirsh-Pasek, K., Newcombe,N., & Golinkoff, R.
(2009, October). Transforming preschoolers’ geometric shape
150 Volume 5—Number 3
Katrina Ferrara et al.
knowledge: Exploring verbalizations and behaviors during a
categorization task. Poster presented at the Biennial Cognitive
Development Society Conference, San Antonio, TX.
Gelman, S. A., & Ebeling, K. S. (1989). Children’s use of nonegocentric
standards in judgments of functional size. Child Development, 60,
Gentner, D. (2003). Why we’re so smart. In D. Gentner & S. Goldin-
Meadow (Eds.), Language in mind: Advances in the study of language
and cognition. Cambridge, MA: MIT Press.
Gentner, D., & Loewenstein, J. (2002). Relational language and
relational thought. In E. Amsel & J. P. Byrnes (Eds.), Language,
literacy, and cognitive development:Thedevelopmentandconsequencesof
symbolic communication (pp. 87–120). Mahwah, NJ: Erlbaum.
Ginsburg, H. (2006). Mathematical play and playful mathematics:
K. Hirsh-Pasek (Eds.), Play =Learning: How play motivates and
enhanceschildren’scognitive andsocial- emotional growth (pp. 145–165).
New York: Oxford University Press.
Guay, R. B., & McDaniel, E. D. (1977). The relationship between
mathematics achievement and spatial abilities among elemen-
tary school children. Journal for Research in Mathematics Education,
Hand, L. L., Uttal, D. H., Marulis, L., & Newcombe, N. S. (2008).
AMeta-analysis oftraining effects onspatialskills. Presented
at the annual meetings of the Association for Psychological
Hart, B.,&Risley, T.(1995).Meaningfuldifferences intheeverydayexperience
of young American children. Baltimore, MD: Brookes.
Hegarty, M., & Kozhevnikov, M. (1999). Types of visual-spatial
representations and mathematical problem solving. Journal of
Educational Psychology, 91,684–689.
Hoff, E. (2006). Environmental supports for language acquisition. In
D. K. Dickinson & S. B. Neuman (Eds.), Handbook of early literacy
research (Vol. 2,pp.163–172).NewYork:GuilfordPress.
Humphreys, L. G., Lubinski, D., & Yao, G. (1993). Correlates of some
curious regressions on a measure of intelligence. Journal of School
Huttenlocher, J., Haight, W., Bryk, A., Seltzer, M., & Lyons, T. (1991).
Early vocabulary growth: Relation to language input and gender.
Developmental Psychology, 27,236–248.
Kamii, C., Miyakawa, Y., & Kato, Y. (2004). The development of
logico-mathematical knowledge inablock-buildingactivityat
ages 1–4. Journal of Research in Childhood Education, 19,44–57.
Kersh, J., Casey, B. M ., &M ercerYoung, J. (2008). Research on spatial
skills and block building in girls and boys: The relationship to
later mathematics learning. In O. N. Saracho & B. Spodek (Eds.),
Contemporary perspectives on mathematics in early childhood education
(pp. 233–252). Charlotte, NC: Information Age.
Leeb-Lundberg, K. (1996). The block builder mathematician. In
E. S. Hirsh (Ed.), The block book (pp. 34–60). Washington, DC:
National Association for the Education of Young Children.
National Association for the Education of Young Children. (1997).
Developmentally appropriate practices in early childhood programs.
Washington, DC: Author.
National Council of Teachers of Mathematics. (2000). Principles and
standards for school mathematics. Reston, VA: Author.
National Research Council. (2006). Learning to think spatially.
Washington, DC: The National Academies Press.
Ness, D., & Farenga, S. J. (2007). Knowledge under construction: The
importance of play in developing children’s spatial and geometric thinking.
Lanham, MD: Rowman & Littleﬁeld.
Neuman, S., & Roskos, K. (1992). Literacy objects as cultural tools:
Effects on children’s Literacy behaviors during play. Reading
Research Quarterly, 27,203–223.
Pruden, S. M., Levine, S. C., & Huttenlocher, J. (2010, March).
Individual differences in children’s spatial language use predicts
later spatial cognition. Paper presented at in S. Roseberry &
un (Chairs), When representational systems collide:
Aligning space and language. International Society on Infant
Studies, Baltimore, MD.
Reifel, S. (1984). Block construction: Children’s developmental
landmarks in representation of space. Young Children, 40,61–67.
Saracho, O. N. (1994). The relationship of preschool children’s
cognitive style to their play preferences. Early Child Development
and Care, 97,21–33.
Saracho, O. N. (1995). Preschooler’s cognitive style and their selection
of academic areas in their play. Early Child Development and Care,
Shea, D. L., Lubinski, D., & Benbow, C. P. (2001). Importance of
assessing spatial ability in intellectually talented young
adolescents: A 20-year longitudinal study. Journal of Educational
Shutts, K., Ornkloo, H., von Hoftsen, C., Keen, R., & Spelke, E.
(2009). Young children’s representations of spatial and func-
tional relations between objects. Child Development, 80, 1612–1627.
Stieff, M. (2007). Mental rotation and diagrammatic reasoning in
science. Learning and Instruction, 17,219–234.
Tracy, D. M. (1987). Toys, spatial ability, and science and mathematics
achievement: Are they related? Sex Roles, 17,115–138.
Vygotsky, L. S. (1978). The role of play in development. In M. Cole, V.
John-Steiner, S. Scribner, & E. Souberman (Eds.), Mind in society:
The development of higher mental processes. Cambridge, MA: Harvard
University Press. (Original work published in 1930, 1993, 1935.)
Weizman, Z. O., & Snow, C. E. (2001). Lexical output as related
to children’s vocabulary acquisition: Effects of sophisticated
exposure and support for meaning. Developmental Psychology, 37,
Wellhousen, K., & Keoff, J. E. (2000). A constructivist approach to block
play in early childhood. New York: Thomas Delmar Learning.
Whitehurst, G. J. (1997). Language processes in context: Language
learning in children reared in poverty. In L. B. Adamson &
M. A. Romski (Ed.), Research on communication and language disor-
ders: Contribution to theories of language development (pp. 233–266).
Baltimore, MD: Brookes.
Wolfgang, C. H., Stannard, L. L., & Jones, I. (2003). Advanced con-
structional play with LEGOs among preschoolers as a predictor
of later school achievement in mathematics. Early Child Develop-
ment and Care, 173(5), 46–75.
Wu, H., & Shah, P. (2004). Exploring visuospatial thinking in chem-
istry learning. Science Education, 88,465–492.
Zacharos, K., Koliopoulos, D., Dokomaki, M. & Kassoumi, H. (2007).
Views of prospective early childhood education teachers,
towards mathematics and its instruction. European Journal of
Teacher Education, 30,305–318.
Volume 5—Number 3 151