ArticlePDF Available

Linguistic Mediation of Children's Performance in a New Symbolic Understanding Task

Authors:

Abstract and Figures

The effects of language on symbolic functioning were examined using the boxes task, a new symbolic understanding task based on DeLoache's model task. Children (N = 32; ages 2;4–3;8) observed an object being hidden in a stack of four boxes and were then asked to retrieve a similar object in the same location from a set of four target boxes. Each box was identified with a different object sitting on a small ledge in front of it. Language use was manipulated by providing linguistic scaffolding (naming vs. standard) and by using objects to identify the boxes whose names were either known or unknown to the children (familiar vs. language control). Home language environment (monolingual vs. bilingual) and verbal age (measured via the Peabody Picture Vocabulary Test-Third Edition) were also examined. Main effects were found for type of object (with familiar objects yielding better performance), home linguistic environment (with bilingualism resulting in better performance), and verbal age. There was also a verbal age × linguistic scaffolds interaction: Verbal age was related to correct retrievals only in the naming condition. These results provide further evidence that language mediates children's symbolic development and indicate that the new boxes task is an effective way of evaluating young children's symbolic competencies.
Content may be subject to copyright.
ARTICLE
Linguistic Mediation of Children’s Performance in
aNewSymbolicUnderstandingTask
Bruce D. Homer and Natalya Petroff
The Graduate Center, City University of New York
Elizabeth O. Hayward
New York University
The effects of language on symbolic functioning were examined using the boxes task, a new
symbolic understanding task based on DeLoache’s model task. Children (N¼32; ages 2;4–3;8)
observed an object being hidden in a stack of four boxes and were then asked to retrieve a similar
object in the same location from a set of four target boxes. Each box was identified with a different
object sitting on a small ledge in front of it. Language use was manipulated by providing linguistic
scaffolding (naming vs. standard) and by using objects to identify the boxes whose names were
either known or unknown to the children (familiar vs. language control). Home language environ-
ment (monolingual vs. bilingual) and verbal age (measured via the Peabody Picture Vocabulary
Test-Third Edition) were also examined. Main effects were found for type of object (with familiar
objects yielding better performance), home linguistic environment (with bilingualism resulting in
better performance), and verbal age. There was also a verbal age "linguistic scaffolds interaction:
Verbal age was related to correct retrievals only in the naming condition. These results provide
further evidence that language mediates children’s symbolic development and indicate that the
new boxes task is an effective way of evaluating young children’s symbolic competencies.
From birth, children are surrounded by symbols in the form of pictures, numbers, and text. Of
utmost importance, infants are immediately surrounded by the symbolic system of language.
While adults have little trouble interpreting the meaning of the symbols that surround them,
it takes time for young children to acquire an understanding of symbol use. Infants and young
children learn to make sense of symbols slowly, incrementally gaining access to the ‘‘symbolic
world’’ around them and the information that is represented by these symbols (Homer &
Hayward, 2008; Nelson, 2003).
Correspondence should be sent to Bruce D. Homer, Program in Educational Psychology, The Graduate Center, City
University of New York, 365 5th Ave., New York, NY 10016, USA. E-mail: bhomer@gc.cuny.edu
JOURNAL OF COGNITION AND DEVELOPMENT, 14(3):455–466
Copyright #2013 Taylor & Francis Group, LLC
ISSN: 1524-8372 print=1532-7647 online
DOI: 10.1080/15248372.2012.689268
To use symbols successfully, one must understand the relationship between the sign and its
referent. Saussure (1959) describes signs as two-sided psychological entities, made up of the sig-
nifier and the signified. The signifier is the property of a sign that relates to the object, while the
signified is the property of a sign that relates to the meaning represented by the sign. For example,
in spoken language, the signifier is the sound heard when a word is uttered, while the signified is
the concept being conveyed by the word. This dual nature of signs is a particular challenge for chil-
dren. Symbolic understanding requires a child to grasp both aspects of the sign, the signifier as well
as the signified. DeLoache (1995, 2002) has argued that young children cannot hold in mind the
sign as both a representation and an objectan ability that she has called dual representation.
In an effort to examine symbolic understanding in young children, DeLoache and her col-
laborators developed the model task, which led to groundbreaking research (e.g., DeLoache,
1991, 1995, 2000; DeLoache & Burns, 1994; DeLoache, Kolstad, & Anderson, 1991; DeLoache
& Marzolf, 1992; DeLoache, Miller, & Rosengren, 1997). In this task, children are shown a
model of a room and told that it corresponds in every way to an adjacent, real room. They
are then shown a toyusually a miniature dollthat is hidden in the model room. Children
are then told that there is a similar, larger toy in the corresponding location in the real room
and are asked to find the big toy in the real room.
In numerous studies with this task, DeLoache has found that although 3-year-olds can use the
model to find the toy in the large room, 2.5-year-olds cannot (e.g., DeLoache, 1987, 1991,
2000; DeLoache et al., 1991). Success on the model task requires representational insight, the abil-
ity to mentally represent the relation between a symbol (i.e., the model) and its referent (i.e., the
room). According to DeLoache, children gain representational insight once they can engage in dual
representation (i.e., seeing a symbol as both a concrete object and an abstract representation).
Several factors affect children’s performance on the model task. When the experimenter
emphasizes the symbolic relation between the model and the room, drawing a connection between
the two, performance is supported. For example, 2.5-year-olds do markedly better on this task if
each object in the model is identified and the child is asked to point to the corresponding object in
the room (Troseth & DeLoache (2003), as cited by DeLoache & Sharon, 2005). Children have
been found to do better when there is greater perceptual similarity between the model and room
(DeLoache et al., 1991). However, the task becomes more difficult when the model is identical to
the room, as this seems to impede children from holding the sign and referent in mind as two
distinct entities (DeLoache & Sharon). Practice with symbolic retrieval tasks for multiple trials
also seems to consistently enhance performance in young children (Marzolf & DeLoache, 1994).
When the ‘‘object’’ properties of the model are particularly salient to the child, it is more dif-
ficult for them to use the model as a symbol. DeLoache (2000) found that 2.5-year-olds perform
significantly better on the task if they view the model through a window (thereby reducing the
physical salience of the model) or view the model in two-dimensional pictures, and found that
they perform worse when they are permitted to play with the model first (emphasizing the physi-
cal salience of the model). This may explain why it is easier for children to understand symbolic
information in two-dimensional pictures than from three-dimensional models, as pictures are not
particularly salient as objects (DeLoache, 1991).
While it is clear from the research of DeLoache and colleagues that the specific task demands,
as well as individual cognitive development, affect the development of symbolic representation,
other research has emphasized the importance of adult social and linguistic interaction in
symbolic competency. The work of Tomasello, Striano, and Rochat (1999) indicates that adults
456 HOMER, PETROFF, AND HAYWARD
scaffold children’s early symbol use through modeling and verbal scripts. Similarly, Striano,
Tomasello, and Rochat (2001) found that symbolic play increased when mothers were present,
which they attributed to maternal modeling and using language to encourage children to interact
with objects symbolically. Sharon (2005) used the model task to explore how intentionality
contributes to early symbol use and understanding in 2.5-year-olds and 3-year-olds using
a within-subjects design. There were two conditions: the standard condition and the intentional
condition. In the standard condition, children received the standard instructions as in DeLoache’s
work. In the intentional condition, children received additional verbal instructions that high-
lighted the intention of the model to serve as a guide to the room (e.g., ‘‘I made the small room
like the big room to help you find Big Teddy Bear’’). There was a main effect for treatment such
that both age groups scored significantly higher in the intentional condition. In other words,
performance was supported by a verbal explanation of how the model was supposed to function.
Evidence of language mediating children’s symbolic functioning also comes from work by
Callaghan (2000), who investigated whether language supports children’s ability to understand
graphical symbols. Children aged 2.5 to 3 years old were shown a picture (e.g., a line drawing)
and were asked to identify which of two objects was depicted in the picture. Callaghan manipu-
lated children’s ability to use verbal labels to perform this task either by using a target object of
which the name would be unknown (e.g., fungus) to the children or by having two referents with
the same basic label (e.g., two ‘‘dogs’’). Callaghan found that children performed significantly
worse when verbal labels were unavailable and concluded, therefore, that language, specifically
labels, facilitates young children’s understanding of the symbolic nature of pictures.
Recently, Homer and Nelson (2009) examined the role of language in the model task. Children
aged 2.5 to 3.5 years old were given a standardized language measure as well as one of two
versions of the model task, a standard version (based on DeLoache, 1995) or a naming version,
which provided linguistic scaffolds. In the naming version, children were asked to name the
location where ‘‘Little Mickey’’ was hiding in the model before searching for ‘‘Big Mickey’’
in the actual room. Homer and Nelson (2009) found that language predicted performance on
the model task. Furthermore, children performed significantly better in the naming version of
the task, supporting Homer and Nelson’s (2009) claim that language serves a number of cognitive
functions that facilitate children’s representational and symbolic abilities. Specifically, Homer
and Nelson (2009) argue that language provides a social context for representation and that it
scaffolds behavior and thought (Vygotsky, 1978). Basing their arguments on Vygotsky’s notion
of mediation, the authors argue that psychological tools, such as language, mediate ‘‘higher’’
mental functions, such as symbolic representation.
Studies employing the model task have shed a great deal of light on young children’s
developing symbolic understanding; however, the original task has some practical limitations.
The model task requires a good deal of space and is not readily portable; it requires the use
of a dedicated room necessitating that participants be seen in a laboratory setting. In addition,
the specific objects and symbols that can be used in the model task are difficult to manipulate,
as each object needs to make sense in the context of a room.
In an effort to address these limitations, a new symbolic understanding task, the boxes task, has
been developed and was tested in the current study. Similar to the three-tiered boxes apparatus
developed by Loewenstein and Gentner (2005) to study the use of relational language and spatial
reasoning in 4- and 5-year-olds, the boxes task employs two sets of boxes in which a ‘‘Mickey’’
doll is hidden. The boxes task presents two sets of four boxes: one smaller set, analogous to the
LINGUISTIC MEDIATION OF SYMBOLIC UNDERSTANDING 457
model in the model task, and a larger set, analogous to the room in the model task. In this new
task, children observe an object hidden in a stack of the smaller four boxes and are then asked
to retrieve a similar object in the same location from a set of bigger target boxes. The bigger target
boxes are identified with a distinct object on a small ledge in front of each box, while the small
model boxes are identified with a corresponding line drawing on a small ledge in front of each
box. Line drawings were employed in an effort to facilitate dual representation by reducing the
physical saliency of the objects (DeLoache, 2000). Although inspired by the DeLoache task
and attempting to capture the same conceptual developments, this new task requires less space
than the original model task and is easily transported, which therefore allow it to be taken out
of the laboratory setting and into the field. Furthermore, the boxes task provides the opportunity
for more comprehensive theory testing, as the target objects used are easily manipulated.
The goal of the current study was to investigate language and symbolic understanding in young
children with the boxes task to expand on the previous findings on language and symbolic under-
standing by Homer and Nelson (2009). This study examined the role of language in children’s
symbolic understanding in two ways: by comparing performance with and without linguistic scaf-
folds (Homer & Nelson, 2009) as well as by manipulating children’s access to verbal labels
(Callaghan, 2000). Homer and Nelson (2009) found that asking children to name hiding spots sig-
nificantly improved their performance on the model task, and therefore, this mode of linguistic
scaffolding was included as a factor in the current study by giving half of the children a standard
version of the boxes task (no linguistic scaffolding) and half a naming condition (with linguistic
scaffolding). Additionally, Callaghan found that children’s ability to comprehend the symbolic
function of realistic line drawings was enhanced when verbal labels for the objects being depicted
were available to children and was hindered when the availability of verbal labels was compro-
mised. Therefore, the availability of verbal labels for hiding spots was examined as another factor
in the current study by manipulating the types of objects used to identify the boxes. Half of the
hiding spots were identified by familiar objects that could easily be given unique names by young
children (e.g., ‘‘cow,’’ ‘‘chicken’’), and the other half were identified with control objects that are
much more difficult for children to give unique names (i.e., different seashells, such as ‘‘clam
shell,’’ ‘‘snail shell’’ that would all be labeled as ‘‘shell’’ by the children). It was hypothesized
that linguistic scaffolding (i.e., children naming the hiding locations), as well as the availability
of unique labels of the hiding spots, would result in improved performance in the boxes task
and that language ability would predict children’s success in this task.
A secondary goal of the study was to assess the viability of the boxes task as a measure of
early symbolic competencies. It was hypothesized that children’s performance on the boxes task
would reveal similar developmental patterns as has been found with the model task.
METHOD
Participants
Children (N¼32) ages 2;4 to 3;8 (M¼3;1, SD ¼4.6 months) participated in the study. There
were 18 boys. The children were recruited from a large urban center via flyers and Internet
postings. The children came from various ethnic and socioeconomic backgrounds; however,
as a prerequisite for participation in this study, all spoke English in the home.
458 HOMER, PETROFF, AND HAYWARD
For about 30%of the children (n¼9), at least one of their caregivers also spoke a language
other than English to the child at home. Although bilingualism was not originally intended to be
studied as a factor, due to the relatively large number of children who came from bilingual
homes and the fact that bilingualism has been shown to have a positive effect on certain cogni-
tive abilities, such as executive functions (Bialystok, 1999; Carlson & Meltzoff, 2008) and meta-
linguistic awareness (Bialystok, 2007; Homer, 2009), it was included as a factor in the current
study. Defining bilingualism is inherently difficult (Bialystok & Hakuta, 1994), and furthermore,
we did not measure children’s competencies in any language other than English, and so we could
not identify the extent to which any of these children were actually bilingual. For the current
article, this group of children is referred to as ‘‘bilingual,’’ but this is intended solely to identify
that they came from a bilingual home. The other languages spoken in the bilingual homes
included Spanish for three children, Chinese for two children, and one child each from homes
in which the following languages were spoken: Hebrew, Italian, Madinco (the primary language
of The Gambia), and Portuguese. As a token of appreciation, all children received a small gift;
their parents received gift certificates, in the amount of $20, redeemable at local bookstores.
Procedure
The study took place in a university child study laboratory. Children were randomly assigned to
either a control group or an experimental group. The control group (standard; n ¼16) received
the standard version of the boxes task, which was similar to the typical DeLoache model task
(e.g., DeLoache et al., 1991). The experimental group (naming; n ¼16) received the naming
version of the boxes task, which was identical to the standard version except that it included
a linguistic scaffold (described in the Procedure section below). In both versions, children were
asked to complete two sets of searches with two types of objects: familiar, which included four
different small plastic farm animals (a cow, sheep, pig, and chicken), and language control,
which included four different seashells (shells from a sea snail, clam, conch, and auger). This
manipulation was similar to that used by Callaghan (2000) to manipulate children’s access to
verbal labels. The order of familiar and control sets was counterbalanced. All children were
given a standardized language test, the Peabody Picture Vocabulary Test-Third Edition
(PPVT-III; Dunn & Dunn, 1997). Lastly, to ensure that the children were able to generate unique
labels for the farm animals (familiar objects) but were only able to generate a basic label (e.g.,
‘‘shell’’) for the seashells (control objects), children were asked to name the objects used to
identify the boxes.
The basic procedure for both groups, standard (no linguistic scaffolds) and naming (linguistic
scaffolds), was the same. The task was set up within a play area in the psychology lab. The play
area was divided by a white cardboard screen, 41 inches high "40 inches wide. On one side of
the screen, four large white boxes (6 inches "6 inches "18 inches) were stacked on the floor,
forming the target boxes. On the other side of the screen, four corresponding small white boxes
(2 inches "2 inches "4 inches) were stacked on top of a small table forming the model boxes.
Each of the four large boxes was equipped with a small ledge on the front side to hold an
identifying object. The small boxes in the model had comparable ledges to hold a line drawing
of an identifying object. The boxes, in both the model and the target, were closed on all sides
except for the right end, which was open but covered with a white curtain (see Figure 1).
LINGUISTIC MEDIATION OF SYMBOLIC UNDERSTANDING 459
As an introduction to the boxes task, children were shown two toys: a stuffed toy, Big Mickey,
and a small plastic replica of Big Mickey, Little Mickey. All children were shown the target
boxes and were told, ‘‘These are Big Mickey’s play boxes where he likes to hide.’’ The children
were then introduced to the model boxes, Little Mickey’s play boxes, and the correspondence
between the target and the model was explained: The model boxes were brought over to the
other side of the screen, and then the experimenter would hold the model boxes next to the target
boxes and name each pair of objects=drawings and highlight the correspondence: ‘‘Look, Little
Mickey’s pig box is just like Big Mickey’s pig-box, only smaller.’’ The identifying objects and
corresponding line drawings were either four farm animals (familiar condition, i.e., unique labels
available) or four seashells (language-control condition, i.e., unique labels unavailable).
The children were told, ‘‘Big Mickey and Little Mickey always do the same thing: Whenever
Little Mickey does something in his play-boxes, Big Mickey does the same thing in his play
boxes.’’ The children then watched as the experimenter hid Little Mickey in one of the four
small boxes, while pointing to the hiding location: ‘‘Little Mickey is going to hide now; look,
he is hiding here.’’ At the same time, on the other side of the screen, a second experimenter hid
Big Mickey in the corresponding box in the target boxes. Children in the naming condition
FIGURE 1 Experimental apparatus: the target boxes on the left (i.e., ‘‘Big Mickey’s play boxes’’) and the model boxes
on the right (i.e., ‘‘Little Mickey’s play boxes’’).
460 HOMER, PETROFF, AND HAYWARD
received linguistic scaffolding: They were prompted to identify Little Mickey’s hiding spot by
labeling the corresponding drawing (e.g., ‘‘Can you tell me where Little Mickey is hiding?’’). If
a child pointed instead, she would be encouraged to label the hiding place, ‘‘Can you say where
Little Mickey is hiding? What’s the name of the box where Little Mickey is hiding?’’ Regardless
of whether the child produced the correct label or not, the experimenter would then explicitly
state Little Mickey’s hiding spot: ‘‘Little Mickey is hiding in the pig-box.’’ Children in the stan-
dard condition received no linguistic scaffolding.
Children in both conditions were then told, ‘‘Remember, Big Mickey is hiding in the same
place in his boxes as Little Mickey; let’s go find Big Mickey now.’’ They were brought to the
other side of the screen and were asked to find Big Mickey (symbolic retrieval). All children
performed eight symbolic retrievals in total: two sets of four trials, one using familiar objects
(farm animals) and one using language-control objects (seashells). The order of objects within
each set was randomized, as was the position of the correct hiding spot. As a memory control,
after finding Big Mickey, children were asked to go back to the model and find Little Mickey
(memory retrieval task).
Each child was scored for the total number of correct symbolic retrievals based on the first
place where the child looked in the target boxes, for a maximum of eight correct symbolic retrie-
vals (four familiar object trials plus four language-control object trials). Also, each child was
scored on the total number of correct memory retrievals based on the first place where the child
looked in the model, for a maximum of eight correct memory retrievals. Each child was assessed
for language ability and was scored for their verbal age, based on the PPVT-III (Dunn & Dunn,
1997). For the PPVT-III task, children were shown a series of four pictures and were asked to
point to the target vocabulary word (e.g., ‘‘point to ball’’). After establishing that a child was
able to follow instructions, testing continued to the next page of four pictures. Testing continued
until the child gave 8 incorrect answers out of 12 consecutive target words.
Finally, at the end of the session, all children were asked to name the familiar and
language-control objects one at a time. Their responses were recorded and coded. Children were
considered correct in their labeling of an object when the label generated was unique and spe-
cific to the object. For example, both ‘‘lamb’’ and ‘‘sheep’’ were deemed correct, as they
uniquely identified the sheep object from the other 3 animals, whereas ‘‘shell’’ was deemed
incorrect, as this label does not distinguish between the 4 seashells. Children were able to label
an average of 3.58 of the 4 farm animals (SD ¼0.87), with 72%of children providing correct
labels for all 4 animals. In contrast, children were able to label an average of 0.23 of the 4
seashells (SD ¼0.43), with 84%failing to provide a correct label for any of the 4 seashells. None
of the children were able to correctly label more than 1 of the 4 seashells. Therefore, the assump-
tion that children would generate unique verbal labels for familiar objects and would be unable
to do so for language-control objects was confirmed.
RESULTS
Preliminary Analyses
To ensure that all the children were able to follow the task and remember the hiding location,
scores on the memory retrievals were examined. The average memory retrieval score overall
LINGUISTIC MEDIATION OF SYMBOLIC UNDERSTANDING 461
was 6 out of 8, which is comparable to previous studies (e.g., DeLoache, 2000; Homer &
Nelson, 2009), indicating that any difficulty with the task was not due to memory issues.
A second set of preliminary analyses was conducted to determine if the standard and naming
groups differed significantly in chronological age, verbal age (based on PPVT-III scores), or
performance on the memory-control questions. The standard and naming groups did not differ
with regards to chronological age, t¼0.04, p¼.970, but did differ marginally in regards to their
verbal age, t¼2.02, p¼.053, such that children in the standard condition had higher verbal ages
(standard, M¼47, SD ¼14.4; naming, M¼37.25, SD ¼12.8). Naming and standard groups did
not differ significantly in the number of correct memory retrievals in the boxes task, t¼#0.09,
p¼.929 (M¼5.88, SD ¼2.19 for standard, M¼5.94, SD ¼1.73 for naming), indicating no
significant group difference in their understanding of the task.
A final set of preliminary analyses was conducted to determine if the children from monolin-
gual homes differed from those from bilingual homes in chronological age, verbal age (based on
PPVT-III scores), or performance on the memory-control questions. The monolingual and
bilingual children did not differ significantly in chronological age, t¼1.51, p¼.143, but did
differ significantly in their verbal age, t¼2.05, p¼.049, such that monolingual children had
a higher average verbal age (M¼45, SD ¼14.1) than did the bilingual children (M¼34,
SD ¼12.0). This difference in English-language ability is not unexpected in children learning
to speak more than one language: Bilingual children are often found to have smaller vocabul-
aries in each of the specific languages that they are learning compared with monolingual children
(Oller & Eilers, 2002). The monolingual and bilingual children did not differ significantly in the
number of correct memory retrievals, suggesting that there was no difference in their under-
standing of the task, t¼1.19, p¼.243.
Effects of Language
To examine the association between language and symbolic understanding in the boxes task,
a repeated-measures analysis of covariance was conducted with the number of correct symbolic
retrievals as the dependent measure and type of objects as a within-subject measure (familiar vs.
control), linguistic scaffolds (standard vs. naming condition), order of objects (familiar objects
first vs. control objects first), and home language environment (monolingual vs. bilingual) as
between-subjects factors, and verbal age (measured via the PPVT-III) as a covariate. The
inclusion of verbal age as a covariate allowed for an examination of the main effect of this factor
and also controlled for the between-group difference in verbal age. The model included main
effects for all factors, the interaction between verbal age and linguistic scaffolds, as well as
the interaction between objects, order of objects, and linguistic scaffolds.
There was a significant main effect of type of object, F(1, 24) ¼4.96, p¼.036, partial
g
2
¼.17, such that when familiar objects were employed, children had more correct symbolic
retrievals (M¼2.22, SD ¼1.18) than when the control objects were employed (M¼2.03,
SD ¼1.15). The effect of the linguistic scaffolds was not found to be significant, F(1,
24) ¼3.54, p¼.072. There was, however, a significant main effect of verbal age, F(1,
24) ¼7.08, p¼.014, partial g
2
¼.23. The effect of verbal age was qualified by an interaction
between verbal age and linguistic scaffolds, F(1, 24) ¼4.70, p¼.040, partial g
2
¼.16. Further
analyses revealed that verbal age was positively correlated with the number of correct symbolic
462 HOMER, PETROFF, AND HAYWARD
retrievals for the linguistic scaffolds group (R
2
¼.16), but not for the control group (R
2
¼.01; see
Figure 2). Finally, there was a significant main effect for home linguistic environment, F(1,
24) ¼9.55, p¼.001, with the children from bilingual homes having a significantly greater
mean number of symbolic retrievals (M¼5.22, SD ¼2.05) compared with the children from
monolingual homes (M¼3.87, SD ¼1.74).
DISCUSSION
The primary goal of the current study was to examine the role of language in children’s performance
on a new symbolic understanding task. Overall, the current results provide further evidence that
language plays an important role in children’s symbolic understanding. A significant main effect
was found for the type of object that was used to indicate the hiding spot in the boxes task. Children
had more correct symbolic retrievals when the hiding locations were indicated with familiar objects,
for which they could generate unique labels, than with the control objects, for which they could not
generate unique labels. This suggests that one way language may support children’s symbolic
competencies is in the access to verbal labels for hiding spots in a symbolic understanding task, such
as the model task or the new boxes task.
The current study adapted Callaghan’s (2000) methodology to investigate the extent to which
children would be more successful in their retrievals when they were able to generate a unique label
for the hiding spot (familiar objects) compared with when they were unable to generate a unique
label (control objects). Similar to Callaghan’s finding that children’s understanding of graphical
FIGURE 2 Number of correct symbolic retrievals by verbal age in linguistic scaffolding versus control conditions.
LINGUISTIC MEDIATION OF SYMBOLIC UNDERSTANDING 463
symbols is supported when they have access to verbal labels, children in the current study were
more successful in their symbolic retrievals when object labels were available (familiar) as com-
pared with when they were unavailable (control), despite the fact that performance in memory trials
between the two conditions did not differ. Therefore, access to verbal labels seems to facilitate
symbolic competencies beyond simply supporting memory for the hiding spot. However, other
factors, such as children’s familiarity with the objects, the discriminability of the objects, etc.,
may also be affecting performance. Further research is needed to isolate these possible effects from
the effects of children having access to unique labels in tasks of symbolic understanding.
Similar to previous findings (Homer & Nelson, 2009), children’s language ability (indicated
by verbal age in the current study) was found to be significantly associated with performance
on the symbolic understanding task. However, this association was only found for children’s
performance in the linguistic scaffolding condition, suggesting that verbal age was predicting
how well children were able to take advantage of the linguistic scaffolds (i.e., being asked to
name the hiding locations) in the boxes task.
Contrary to expectations, there was not a main effect of linguistic scaffolds (although it
approached significance, p¼.072). In this way, the current findings diverge from those of
Homer and Nelson (2009). Homer and Nelson (2009) found that linguistic scaffolds enhanced
performance on the model task, and they argued that these scaffolds helped children form a
linguistic representation that supported their symbolic understanding. As Homer and Nelson
(2009) employed the original model task in their study, the linguistic scaffolding condition
involved the naming of both a spatial preposition (such as on, under, or in) as well as an object
(such as chair, table, or bookshelf). Both the preposition and the object varied across retrievals.
In the current boxes task, only the object varied (e.g., pig box vs. cow box) from trial to trial,
while the preposition stayed the same: The toy was always in a box. Therefore, it is possible
that linguistic scaffolds are most effective if more linguistic information needs to be tracked
to guide behavior. Loewenstein and Gentner (2005) found that hearing spatial-relational lan-
guage (such as on, under, or in) helped to guide searching behavior in preschoolers; therefore,
it is plausible that 3-year-olds are best able to benefit from the linguistic scaffolds when spatial
prepositions are included. However, the interaction between linguistic scaffolds and verbal age
mentioned earlier suggests that children with higher verbal ages were able to benefit from the
linguistic scaffolds in the boxes task. This effect warrants further investigation.
Finally, a significant effect of home linguistic environment was found. Despite having lower
English verbal ages and there being no difference in chronological age, children from bilingual
homes had significantly more correct symbolic retrievals compared with their monolingual
counterparts. Cognitive advantages have been found for bilingual children as young as the part-
icipants in the current study. For example, Bialystok and her colleagues have found that 2- and
3-year-old bilingual children show advantages in executive functions compared with monolin-
gual children (Bialystok, Barac, Blaye, & Poulin-Dubois, 2010; Poulin-Dubois, Blaye, Coutya,
& Bialystok, 2011). Older bilingual children have been shown to have advantages in metalin-
guistic awareness (Bialystok, 2007). The current finding that children from bilingual homes per-
form better on the boxes task calls for further research into what is responsible for this effect. It is
possible that either executive functions or metalinguistic awarenessor bothare responsible.
It may also be that young bilingual children are better able to engage in dual representation
because they are required to hold multiple linguistic representations for the same object
(e.g., ‘‘ball’’ and ‘‘bola’’). Further research is required to examine these factors.
464 HOMER, PETROFF, AND HAYWARD
A secondary goal of the current study was to establish the viability of the new boxes task for
assessing young children’s symbolic competencies. Children’s performance on the boxes task is
similar to how children of this age perform on the model task. The newly developed boxes task has
some practical advantages over DeLoache’s original model task. The boxes task is readily port-
able and space-efficient and could therefore be used in a variety of settings. The space demands of
the original model task constrain data collection to the laboratory environment. In contrast, the
boxes task could be taken into the field, such that studies could be conducted in schools or even
in the homes of participants. The fact that this apparatus is easily transported makes extensive
longitudinal research in the area of symbolic understanding more feasible. Furthermore, unlike
the model task, where hiding places must make sense in the context of a room, the boxes task
allows for the manipulation of the objects used. The freedom to manipulate hiding places allows
for closer investigation and theory testing of the factors at play as children develop symbolic under-
standing. In the current study, a manipulation of the objects used to indicate hiding places allowed
for a closer investigation of how language interacts with the development of symbolic understand-
ing, as indicated by previous theory and research (Homer & Nelson, 2005, 2009).
Future research on the development of children’s symbolic competencies should not only
further illuminate how language ability and bilingualism improve performance in experimental
tasks, but should also investigate the ways in which language mediates the use of everyday
symbols and representational tools such as maps and diagrams. Furthermore, understanding
symbolic representation is also the foundation for early literacy acquisition. Cognitive flexibility,
or the ability to consider simultaneously multiple representations such as symbols, is central
in integrating multiple aspects of language, which is essential for comprehending written signs
(Homer & Hayward, 2008). As children learn to read, they are learning to decrypt written
representations, similarly to how at an earlier age, they learned to understand symbolic repre-
sentations in the world around them. Early symbolic understanding is also involved in the
process by which young children transfer information learned in picture books to the real world
(Simcock & DeLoache, 2006). In an increasingly digital age, more research needs to address
how language and symbols come into play when young children interact with new technologies,
such as touch screens and smart phones. Longitudinal research should be conducted to see how
children’s symbolic understanding changes as language develops and how these developments
are influenced by exposure to new media technology. The new boxes task, introduced here,
provides a useful tool for exploring young children’s symbolic understanding and competencies
as they learn to navigate their complex symbolic worlds.
REFERENCES
Bialystok, E. (1999). Cognitive complexity and attentional control in the bilingual mind. Child Development,70,
636–644.
Bialystok, E. (2007). Acquisition of literacy in bilingual children: A framework for research. Language Learning,57,
45–77.
Bialystok, E., Barac, R., Blaye, A., & Poulin-Dubois, D. (2010). Word mapping and executive functioning in young
monolingual and bilingual children. Journal of Cognition and Development,11, 485–508.
Bialystok, E., & Hakuta, K. (1994). In other words: The psychology and science of second language acquisition.
New York, NY: Basic Books.
Callaghan, T. (2000). Factors affecting children’s graphic symbol use in the third year: Language, similarity, and
iconicity. Cognitive Development,15, 185–214.
LINGUISTIC MEDIATION OF SYMBOLIC UNDERSTANDING 465
Carlson, S. M., & Meltzoff, A. N. (2008). Bilingual experience and executive functioning in young children.
Developmental Science,11, 282–298.
DeLoache, J. S. (1987). Rapid change in the symbolic functioning of very young children. Science,238, 1556–1557.
DeLoache, J. S. (1991). Symbolic functioning in very young children: Understanding of pictures and models. Child
Development,62, 736–752.
DeLoache, J. S. (1995). Early understanding and use of symbols: The model model. Current Directions in Psychological
Science,4, 109–113.
DeLoache, J. S. (2000). Dual representation and young children’s use of scale models. Child Development,71, 329–338.
DeLoache, J. S. (2002). The symbol-mindedness of young children. In W. Hartup & R. A. Weinberg (Eds.), Child
psychology in retrospect and prospect: In celebration of the 75th anniversary of the Institute of Child Development
(Vol. 32, pp. 73–101). Mahwah, NJ: Erlbaum.
DeLoache, J. S., & Burns, N. M. (1994). Symbolic functioning in preschool children. Journal of Applied Developmental
Psychology,15, 513–527.
DeLoache, J. S., Kolstad, V., & Anderson, K. N. (1991). Physical similarity and young children’s understanding of
scale models. Child Development,62, 111–126.
DeLoache, J. S., & Marzolf, D. P. (1992). When a picture is not worth a thousand words: Young children’s understanding
of pictures and models. Cognitive Development,7, 317–329.
DeLoache, J. S., Miller, K. F., & Rosengren, K. S. (1997). The credible shrinking room: Very young children’s
performance with symbolic and nonsymbolic relations. Psychological Science,8, 308–313.
DeLoache, J. S., & Sharon, T. (2005). Symbols and similarity: You can get too much of a good thing. Journal of
Cognition and Development,6, 33–49.
Dunn, L. M., & Dunn, L. M. (1997). Peabody Picture Vocabulary Test (3rd ed.). Circle Pines, MN: American Guidance
Service, Inc.
Homer, B. D. (2009). Literacy and metalinguistic development. In D. R. Olson & N. Torrance (Eds.), The Cambridge
handbook of literacy (pp. 487–500). Cambridge, MA: Cambridge University Press.
Homer, B. D., & Hayward, E. O. (2008). Cognitive and representational development in children. In K. B. Cartwright
(Ed.), Literacy processes: Cognitive flexibility in learning and teaching (pp. 19–41). New York, NY: Guilford Press.
Homer, B. D., & Nelson, K. (2005). Seeing objects as symbols and symbols as objects: Language and the development
of dual representation. In B. D. Homer & C. S. Tamis-LeMonda (Eds.), The development of social cognition
and communication (pp. 29–52). Mahwah, NJ: Lawrence Erlbaum Associates.
Homer, B. D., & Nelson, K. N. (2009). Naming facilitates young children’s understanding of scale models: Language
and the development of symbolic understanding. Journal of Cognition and Development,10, 115–134.
Loewenstein, J., & Gentner, D. (2005). Relational language and the development of relational mapping. Cognitive
Psychology,50, 315–353.
Marzolf, D. P., & DeLoache, J. S. (1994). Transfer in young children’s understanding of spatial relations. Child
Development,64, 1–15.
Nelson, K. (2003). Making sense in a world of symbols. In A. Toomela (Ed.), Cultural guidance in the development of
the human mind (pp. 139–162). Westport, CT: Greenwood.
Oller, D. K., & Eilers, R. E. (Eds.). (2002).Language and literacy in bilingual children.Tonawanda,NY:MultilingualMatters.
Poulin-Dubois, D., Blaye, A., Coutya, J., & Bialystok, E. (2011). The effects of bilingualism on toddlers’ executive
functioning. Journal of Experimental Child Psychology,108, 567–579.
Saussure, F. D. (1959). Course in general linguistics. New York, NY: The Philosophical Library, Inc. (Original work
published 1915).
Sharon, T. (2005). Made to symbolize: Intentionality and children’s early understanding of symbols. Journal of
Cognition and Development,6, 163–178.
Simcock, G., & DeLoache, J. (2006). Get the picture? The effects of iconicity on toddlers’ reenactment from picture
books. Developmental Psychology,42, 1352–1357.
Striano, T., Tomasello, M., & Rochat, P. (2001). Social and object support for early symbolic play. Developmental
Science,4, 442–455.
Tomasello, M., Striano, T., & Rochat, P. (1999). Do young children use objects as symbols? British Journal of
Developmental Psychology,17, 563–584.
Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes. Cambridge, MA: Harvard
University Press.
466 HOMER, PETROFF, AND HAYWARD
ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
Surface similarity generally promotes reasoning by analogy and physical similarity has been shown to have a powerful positive effect on very young children's use of a scale model as a source of information about another space. The research reported here investigated 2!-year -old children's performance in an object retrieval task when asked to reason from one space to a second, perceptually identical one. Con- trary to the expectation that increased surface similarity would improve the chil- dren's performance, it actually had a negative effect. Children were less successful in 3 different tasks in which 2 spaces were identical than a task in which the spaces were merely similar in size and appearance. Their performance appeared to be affected by 4 factors: representational insight, memory for the most recent hiding place, the abil- ity to resist responding perseveratively, and their level of analysis of the task. The re- sults suggest that part of the power of symbols derives from the fact that they are vir- tually never identical to their referents, thereby making it possible to use one to draw inferences about the other without risk of confusing them. The identification of similarity between separate entities is a pervasive and powerful tool of human cognition. Its power derives in part from its remarkable flexibility. We can recognize similarities between bald heads and golf balls, digital clocks and sun- dials, dobermans and dachshunds, the solar system and an atom. We interpret things as alike based on appearance, function, kind, and relational structure, as in the previ- ous examples, as well as an unlimited list of other factors—material composition, size, smell, ownership, geographical origin, and so on. Without an appreciation of similarity, fundamental cognitive processes such as categorization, induction, gen-
Article
Full-text available
Becoming a proficient symbol user is a universal developmental task in the first years of life, but detecting and mentally representing symbolic relations can be quite challenging for young children To test the extent to which symbolic reasoning per se is problematic, we compared the performance of 2 1/2-year-olds in symbolic and nonsymbolic versions of a search task. The children had to use their knowledge of the location of a toy hidden in a room to draw an inference about where to find a miniature toy in a scale model of the room (and vice versa) Children in the nonsymbolic condition believed a shrinking machine had caused the room to become the model They were much more successful than children in the symbolic condition, for whom the model served as a symbol of the room The results provide strong support for the role of dual representation in symbol understanding and use.
Article
Full-text available
Two studies examined language and understanding of scale models. First, children (N = 16; ages 2;4 to 3;5) received either the standard DeLoache model task or a naming version (in which children are asked to name the hiding location before retrieving a hidden object). Language ability positively correlated with performance on the model task, and children had significantly more errorless retrievals in the naming condition. Study 2 examined naming and transfer effects by giving children (N = 18; ages 2;5–3;1) both versions of the model task. Language ability correlated significantly only with performance in the naming condition. Both prior experience and naming condition significantly improved performance, with the effects being additive. Language and children's development of symbolic understanding are discussed.
Article
This book sets a high standard for rigor and scientific approach to the study of bilingualism and provides new insights regarding the critical issues of theory and practice, including the interdependence of linguistic knowledge in bilinguals, the role of socioeconomic status, the effect of different language usage patterns in the home, and the role of schooling by single-language immersion as opposed to systematic training in both home and target languages. The rich landscape of outcomes reported in the volume will provide a frame for interpretation and understanding of effects of bilingualism for years to come.
Article
The hallmark of human cognition is symbolization: There is nothing that so clearly distinguishes us from other creatures as our creative and flexible use of symbols. Cultural creations such as writing systems, number systems, maps, and models-to name a few-have enabled human knowledge and reasoning to transcend time and space. My working definition of an external, artifactual symbol is that it is any entity that someone intends to stand for something other than itself. Note that this definition is agnostic about the nature of symbols; virtually anything can be a symbol, so long as some person intends that it be responded to not as itself, but in terms of what it represents. Adults are so experienced and skilled with symbols and symbolic reasoning that they simply assume that many of the novel entities they encounter will have symbolic import. They appreciate that such entities should be responded to as representations of something other than themselves and readily do so. My research reveals that children only gradually adopt this assumption. Despite the centrality of symbolization in human cognition and communication, young children are very conservative when it comes to detecting and reasoning about symbol-referent relations.