Building Blocks for Developing Spatial Skills: Evidence From a Large,Representative U.S. Sample

Abstract

There is evidence suggesting that children’s play with spatial toys (e.g., puzzles and blocks) correlates with spatial development. Females play less with spatial toys than do males, which arguably accounts for males’ spatial advantages; children with high socioeconomic status (SES) also show an advantage, though SES-related differences in spatial play have been less studied than gender-related differences. Using a large, nationally representative sample from the standardization study of the Wechsler Preschool and Primary Scale of Intelligence–Fourth Edition, and controlling for other cognitive abilities, we observed a specific relation between parent-reported frequency of spatial play and Block Design scores that was invariant across gender and SES. Reported spatial play was higher for boys than for girls, but controlling for spatial play did not eliminate boys’ relative advantage on this subtest. SES groups did not differ in reported frequency of spatial play. Future research should consider quality as well as quantity of play, and should explore underlying mechanisms to evaluate causality.

Keywords

spatial ability cognitive development learning socioeconomic status sex differences

The ability to reason about space is important for many aspects of daily functioning, such as navigating to a store, using diagrams to install a ceiling fan, or planning a garden (National Research Council, 2006). Spatial skills are also important for success in the science, technology, engineering, and math (STEM) domains (Uttal & Cohen, 2012; Wai, Lubinski, & Benbow, 2009). Hence, there has been a great deal of interest in supporting the development of spatial skills in early childhood. Research suggests that children’s play with certain types of toys (e.g., puzzles or blocks) is related to children’s spatial development (Verdine, Golinkoff, Hirsh-Pasek, & Newcombe, 2014), and that differential experience with these types of toys and activities could contribute to gender differences in spatial performance as well as to differences in spatial performance across socioeconomic (SES) levels (e.g., Levine, Vasilyeva, Lourenco, Newcombe, & Huttenlocher, 2005; Serbin, Zelkowitz, Doyle, Gold, & Wheaton, 1990). In the current study, we used a large, nationally representative sample to assess whether frequency of play with spatial toys is associated with spatial performance across various groups, even when other abilities and other types of play are statistically controlled. In addition, we assessed whether there are gender- and SES-based differences in spatial performance and reported frequency of spatial play, and if so, whether the differences in spatial play appear to be related to the differences in spatial performance.

Engagement with spatial toys and activities certainly seems likely to improve spatial skill. For example, when using a diagram to complete a Lego structure, a child practices one-to-one correspondence and spatial scaling. When completing a puzzle, a child uses mental rotation to match up pieces of pictures. Empirical support for the relationship between spatial play materials and spatial skills has been found for blocks (Brosnan, 1998; Caldera et al., 1999; Serbin et al., 1990), tangram puzzles and pentominoes (J. C. Yang & Chen, 2010), jigsaw puzzles (Levine, Ratliff, Huttenlocher, & Cannon, 2012), and mazes (Jirout & Newcombe, 2014). Numerical knowledge, which is closely related to spatial thinking (e.g., Dehaene, Bossini, & Giraux, 1993; Galton, 1880a, 1880b; Smith, 1964; T. Yang et al., 2014), was found to be related to experience with board games and card games (Ramani & Siegler, 2008). In a subsequent study, playing a linear, but not a circular, board game improved children’s math performance (Siegler & Ramani, 2009). Though the researchers focused on math learning, they considered the spatial components of the games (e.g., linearity, equal spacing between numbers, and the use of representational mapping and spatial alignment) to be important for learning (Siegler & Ramani, 2009). Thus, in the current study, we considered playing with puzzles, blocks, and board games to be spatial play that could be associated with the development of spatial reasoning. Additionally, we assessed whether parent-child activities related to mathematics, such as teaching number skills and playing math games, are associated with spatial performance.

We also examined reported frequency of play with spatial toys separately for boys and girls and for different SES levels. Researchers have found that people think of puzzles and block play as “masculine” and think of board games as more gender neutral (e.g., Connor & Serbin, 1977). Such ideas could influence children’s access to these toys and how often they are encouraged to engage in play with them. However, research suggesting that boys tend to play more often with these types of materials than girls do is somewhat dated (Serbin et al., 1990), and more recent work suggests that there is a gender difference in quality, but not frequency, of play with these materials (Levine et al., 2012). Ramani and Siegler (2008) found that middle-SES children had more experience with board games and card games than lower-SES children, but SES-related differences in frequency of other types of spatial play have been examined less often than gender-related differences.

Because spatial intelligence is malleable (Uttal et al., 2013), it is important to determine whether different groups of children receive differential spatial-play experience, which might account for group differences in spatial performance (Levine et al., 2005; Newcombe, 2007). For example, higher-SES children have shown an advantage in spatial performance (Levine et al., 2005). Also, at almost all ages, males have been found to outperform females on mental rotation tasks, though task characteristics also seem to influence performance (Miller & Halpern, 2014), and other variables—such as confidence or sex stereotypes— might mediate the gender differences (Estes & Felker, 2012; Lippa, Collaer, & Peters, 2010). Intervention, such as with spatial play, can potentially diminish gender differences in spatial performance in both children and adults (Newcombe & Stieff, 2012; Tzuriel & Egozi, 2010), and a recent study found that a spatial-play intervention improved spatial performance in economically disadvantaged populations (Grissmer et al., 2013). We explored whether spatial performance is related to more naturalistic play with spatial toys and whether the frequency of this type of play differs across groups.

The Current Study

Using a large and diverse sample, we investigated whether spatial play has a specific influence on spatial thinking by controlling for general intellectual ability and analyzing nonspatial play behaviors as well as spatial ones. Specifically, we used data from the normative sample for the fourth edition (the most recent revision) of the Wechsler Preschool and Primary Scale of Intelligence (WPPSI-IV; Wechsler, 2012), focusing on the Block Design test and a questionnaire with items about children’s play.

Our hypotheses were that children’s frequency of spatial play—specifically, play with puzzles, blocks, and board games—would relate to their spatial ability as measured by the Block Design test, even when we controlled for general IQ. We also expected to observe an effect of SES, with higher-SES children outperforming lower-SES children. Boys and girls do not typically perform differently on the Block Design test; however, because we were controlling for several other aspects of cognitive ability (e.g., verbal ability, processing speed, fluid reasoning, and working memory), we believed that we might detect gender differences not typically observed on this task but found for many other spatial-ability measures (Voyer, Voyer, & Bryden, 1995). That is, boys might excel at the Block Design test relative to the other WPPSI-IV subtests and therefore might show a gender advantage when we controlled for scores on other subtests. We expected that frequency of nonspatial play would not influence Block Design scores. Finally, we expected boys and higher-SES groups to have higher reported frequency of spatial play than girls and lower-SES groups.

Method

Participants

Participants were children that Pearson Education recruited for standardization of the WPPSI-IV (Wechsler, 2012). This initial sample included 1,700 children ages 2 through 7. Because several of the WPPSI subtests were designed for children ages 4 years and older, we included only 4- to 7-year-olds in our study, which resulted in a potential sample of 1,100 participants (mean age = 5.48 years, SD = 0.95, range = 4.00–7.60). Parents of 913 of these children responded to the frequency-of-play items on the questionnaire, but 66 of these parents did not report their income, so our final sample included 847 children.¹ The 2010 U.S. Census data were used to ensure a nationally representative sample; 51% of the children were female, and the racial-ethnic distribution was as follows: 52.5% White, 24.8% Hispanic, 15.2% Black, 3.1% Asian, and 4.4% other. Pearson Education compensated the parents of the participating children.

Measures

Our measures were the WPPSI-IV test (Wechsler, 2012) and the Home Environment Questionnaire (HEQ), a 41-item survey about children’s behavior, parent-child interaction, and family demographics developed for use in the WPPSI-IV standardization study. The subtests were administered according to the WPPSI-IV manual. Pearson Education developed the measures and collected all data.

Ability

Spatial ability was assessed using the Block Design subtest from the WPPSI-IV. This visual-spatial subtest is often seen as a relatively general indicator of spatial ability (Groth-Marnat & Teal, 2000). Children were asked to reproduce two-dimensional color designs using cubes with red, white, and half-red, half-white faces (Wechsler, 2012).

General intelligence was assessed using the Full-scale IQ score (FSIQ), which includes five core indices from the WPPSI-IV battery: Verbal Comprehension (two measures), Visual Spatial (Block Design), Fluid Reasoning, Working Memory, and Processing Speed (Wechsler, 2012). We also calculated a revised FSIQ (FSIQ-R) by summing scores on all the subtests included in the standard FSIQ except Block Design. FSIQ-R was used to control for general intellectual ability² when Block Design score was our dependent variable.

Play and parent-child activities

On the HEQ, parents reported how many times per week their children engaged in play with certain categories of toys and engaged in specific types of parent-child activities. There were four response options: “never,” “rarely” (i.e., once or twice per week), “sometimes” (i.e., three to five times per week), and “often” (i.e., six times or more). Because “never” responses were very rare, and for some items no parents gave this response, we collapsed “never” and “rarely” responses into a single category for analyses.

The toy categories were (a) puzzles, blocks, and board games; (b) drawing materials; (c) sound-producing toys (e.g., rattles, drums, guitars); (d) dolls, balls, cars, and trucks; and (e) bicycles, skateboards, scooters, and swing sets. The parent-child activities chosen for inclusion in this study involved math, spatial, and language activities: (a) teaching number skills (e.g., counting, adding), (b) teaching shapes, (c) playing math games, (d) playing word and spelling games, (e) telling stories (e.g., making up stories, sharing family history or folk tales), and (f) talking. The activity of talking with the child could not be included in our analyses because 94% of parents reported the highest level of frequency for this activity.

HEQ

Parents reported their education level on the HEQ by selecting from 8 multiple-choice options ranging from “less than high school” to “doctorate.” Income was reported using 12 options, from “Under $9,999,” “$10,000–$14,999,” “$15,000–$19,999,” and “$20,000–$29,999” through “$90,000–$99,999,” “$100,000–$199,000,” and “Over $200,000.”

Results

Parent education and household income were highly correlated, r(842) = .60, p < .001. Household income had higher correlations than parent education with Block Design, r(847) = .31 versus r(904) = .25; FSIQ-R, r(847) = .42 versus r(904) = .37; and FSIQ, r(847) = .40 versus r(904) = .36. Therefore, we focused on income in our analyses, collapsing the 12 response options into three categories: low (under $30,000), medium ($30,000–$70,000), and high (over $70,000). These categories were consistent with those used in previous research on SES effects on spatial ability (Levine et al., 2005). Our sample for analyses included 247 children in the low-income group (53% female), 259 children in the middle-income group (49% female), and 341 children in the high-income group (50% female).

Gender- and SES-related differences in spatial performance

We began by investigating the influence of gender and SES on children’s performance on the Block Design test, both with and without controlling for FSIQ-R as a covariate. Without FSIQ-R as a covariate, the main effect of SES was significant, F(2, 841) = 34.90, p < .001, η_p² = .077. The main effect of gender (p = .49) and the Gender × SES interaction (p = .78) were not significant. With FSIQ-R as a covariate, there was again a significant effect of SES, F(2, 840) = 3.03, p = .05, η_p² = .007, and additionally there was a significant main effect of gender, F(1, 840) = 7.17, p < .01, η_p² = .008, as well as of FSIQ-R itself, F(1, 840) = 303.49, p < .001, η_p² = .265. Males outperformed females, and the low-SES group scored significantly lower than both the middle- and the high-SES groups (both ps = .03; see Fig. 1 for mean Block Design scores for the gender and SES groups).

Fig. 1.

Mean Block Design score by socioeconomic status (SES) and gender, controlling for age and revised Full-scale IQ score. Error bars show 95% confidence intervals.

Relationship between spatial performance and spatial activities

Play with toys

To investigate the influence of play behaviors on Block Design score, we ran an analysis of covariance (ANCOVA) with reported frequency of the five categories of play as independent variables and gender, SES, and FSIQ-R as covariates (see Fig. 2 for mean Block Design score as a function of play behaviors). We again observed significant main effects of gender, F(1, 824) = 4.13, p = .04, η_p² = .005, and FSIQ-R, F(1, 824) = 280.01, p < .001, η_p² = .254, although the effect of SES was no longer significant (p = .12). As predicted, the frequency of playing with puzzles, blocks, and board games had a significant main effect on children’s Block Design score, F(2, 824) = 5.35, p < .01, η_p² = .013. Children who reportedly played “often” with such toys had significantly higher Block Design scores than did those who reportedly played with them “sometimes” (p < .01) or “rarely” or “never” (p < .01; the difference between the latter two groups was not significant, p = .98). Although the effect size (η_p²) was low, total adjusted R² was only .33, and—more important—the effect of play frequency was almost twice the size of the effect of SES in the original model. We calculated Cohen’s ds by dividing the difference between each pair of group means by the standard deviation of scores and found that the effect size for playing with puzzles, blocks, and board games “often” as opposed to “rarely” or “never” was 0.22, and the effect size for playing with such toys “often” as opposed to “sometimes” was 0.23. None of the other play types had significant main effects on Block Design score (drawing materials: p = .15; sound-producing toys: p = .17; dolls, balls, cars, and trucks: p = .78; and bicycles, skateboards, scooters, and swing sets: p = .33).

Fig. 2.

Relationship between play behaviors and Block Design score, controlling for revised Full-scale IQ score, age, and socioeconomic status. Error bars show 95% confidence intervals.

We reran the ANCOVA with frequency of play with puzzles, blocks, and board games as the only independent variable in order to test the full-factorial model (i.e., including interaction terms with gender and SES), and we observed the same results, with no significant interactions³ (ps > .18). There were again significant main effects of spatial play, F(1, 824) = 5.62, p < .01, η_p² = .013; gender, F(1, 824) = 5.75, p = .02, η_p² = .007; and FSIQ-R, F(1, 824) = 285.47, p < .001, η_p² = .26, and the effect of SES was not significant (p = .20).

Parent-child activities

We also ran an ANCOVA to test for main effects of reported frequency of the five categories of parent-child activities as well as gender and SES, controlling for FSIQ-R. As in the ANCOVA for play with toys, this model showed significant main effects of gender, F(1, 826) = 7.62, p < .01, η_p² = .009, and FSIQ-R, F(1, 826) = 289.02, p < .001, η_p² = .259; the parent-child activities did not have significant effects (telling stories: p = .52; playing word and spelling games: p = .06; playing math games: p = .15; teaching number skills: p = .73; teaching shapes: p = .33). Because playing word and spelling games had a marginally significant effect, we included this variable in a full-factorial model to investigate possible interactions, but found neither the main effect (p = .13) nor any interactions (ps > .34) to be significant; again, only gender and FSIQ-R had significant effects.

Group differences in frequencies of play with toys and parent-child activities

There is evidence that boys and girls have differential experience with spatial play (Brosnan, 1998; Connor & Serbin, 1977) and some support that spatial-play experience varies across SES levels as well (Levine et al., 2005; Ramani & Siegler, 2008). To explore this possibility in the current data set, we analyzed the influence of gender and SES on the reported frequency of the different types of play and parent-child activities. Because the dependent variables were ordinal, we explored group differences in the sample using ordinal regression. For each type of activity, we created a polytomous universal model (PLUM) to evaluate the association of frequency of the activity with gender and SES; data for males and high-SES children were baseline values. We included FSIQ (i.e., including Block Design) of the WPPSI-IV as a control for IQ. For parent-child activities, several analyses violated the assumptions of parallelism; however, when we reran the analyses using multinomial regression, we observed similar results, so we present results from the ordinal regression analyses for ease of interpretation. We report the individual effects of levels of the group variables only if they had a significant effect (i.e., p < .05) on the frequency ratings. Tables 1 and 2 summarize the reported frequencies by gender and SES.

Table 1.

Reported Frequencies for Each Type of Play by Gender and Socioeconomic Status

	Puzzles, blocks, board games			Drawing materials			Sound-producing toys			Dolls, balls, cars, trucks			Bicycles, skateboards, scooters, swing sets
Group	Never/rarely	Sometimes	Often	Never/rarely	Sometimes	Often	Never/rarely	Sometimes	Often	Never/rarely	Sometimes	Often	Never/rarely	Sometimes	Often
Gender
Girls	26.5	50.7	22.8	4.6	23.6	71.8	52.2	32.3	15.5	11.0	31.1	57.9	26.4	44.7	28.9
Boys	22.3	48.6	29.1	14.7	43.1	42.2	51.4	34.9	13.6	11.6	29.7	58.7	21.4	40.5	38.1
SES
Low	26.9	50.6	22.4	10.1	36.0	53.8	43.3	35.9	20.8	10.6	25.6	63.8	27.5	36.8	35.6
Middle	24.8	49.6	25.6	8.9	32.2	58.9	50.6	36.6	12.8	9.3	28.8	61.9	21.7	47.7	30.6
High	22.1	48.2	29.7	10.3	30.0	59.7	58.5	29.7	11.8	13.5	32.9	53.5	22.1	42.6	35.3

Note: For each group and type of play, the table reports the percentage of responses in each category. SES = socioeconomic status.

Table 2.

Reported Frequencies of Each Parent-Child Activity by Gender and Socioeconomic Status

	Teaching number skills			Teaching shapes			Playing math games			Playing word and spelling games			Telling stories
Group	Never/rarely	Sometimes	Often	Never/rarely	Sometimes	Often	Never/rarely	Sometimes	Often	Never/rarely	Sometimes	Often	Never/rarely	Sometimes	Often
Gender
Girls	15.7	49.0	35.3	31.5	33.3	35.3	44.9	43.4	11.7	33.0	44.3	22.7	28.9	44.8	26.3
Boys	16.6	44.9	38.5	24.1	39.9	36.0	45.2	39.7	15.1	35.7	46.1	18.2	27.8	47.3	24.9
SES
Low	15.9	40.2	43.9	21.1	30.5	48.4	40.2	41.9	17.9	35.4	47.2	17.5	24.8	46.3	28.9
Middle	19.7	44.8	35.5	25.5	41.3	33.2	49.6	41.5	8.9	38.8	42.6	18.6	30.1	45.2	24.7
High	14.4	53.2	32.4	34.9	36.4	28.7	46.3	39.9	13.8	30.2	44.6	25.2	29.9	45.5	24.6

Note: For each group and activity, the table reports the percentage of responses in each category. SES = socioeconomic status.

Play with toys

Spatial play—play with puzzles, blocks, and board games—showed a gender difference. Boys were reported to engage in such play more than girls, b = −0.317, 95% confidence interval (CI) = [−0.57, −0.06], Wald χ²(1, N = 843) = 5.83, p = .02. This result remained significant when we controlled for Block Design score rather than FSIQ. Thus, the gender difference in frequency of spatial play does not seem to have been driven by a difference in spatial ability. Compared with boys, girls were reported to play more with drawing materials, b = 1.28, 95% CI = [1.00, 1.57], Wald χ²(1, N = 845) = 79.70, p < .001, and to play less with bicycles, skateboards, scooters, and swing sets, b = −0.35, 95% CI = [−0.60, −0.10], Wald χ²(1, N = 845) = 7.45, p < .01.

Children in the low-SES group had higher reported frequency of play with sound-producing toys than did children in the high-SES group, b = 0.648, 95% CI = [0.305, 0.992], Wald χ²(1, N = 842) = 13.69, p < .001. Using both PLUM and univariate ANCOVAs, we found no difference in average reported frequency of all play types across the SES groups (ps = .19 and .74, respectively) or between the genders (ps = .32 and .66, respectively).

Parent-child activities

Children in the low-SES group, compared with those in the high-SES group, had higher frequencies reported for teaching number skills, b = 0.46, 95% CI = [0.12, 0.80], Wald χ²(1, N = 845) = 7.07, p < .01. They also had higher frequencies reported for teaching shapes, b = 0.80, 95% CI = [0.47, 1.14], Wald χ²(1, N = 843) = 22.09, p < .001; compared with the high-SES group, the middle-SES group also had higher reported frequencies for teaching shapes, b = 0.31, 95% CI = [0.01, 0.62], Wald χ²(1, N = 843) = 4.13, p = .04. Reported frequency of playing math games was higher for children in the low-SES group than for high-SES children, b = 0.57, 95% CI = [0.23, 0.91], Wald χ²(1, N = 845) = 10.58, p = .001.

Ordinal regression (PLUM) showed that the average reported frequency of all parent-child activities was higher in the low-SES group than in the high-SES group (p < .01). An ANCOVA showed a similar main effect of SES on average reported frequencies (p < .001). We coded reported frequency for these activities on a 4-point scale (0 = never, 1 = rarely, 2 = sometimes, 3 = often). On this scale, the average score was 2.0 (SE = 0.04) for low-SES children but 1.8 (SEs = 0.04 and 0.03) for both middle- and high-SES children (ANCOVA: ps < .01). Average reported frequencies of parent-child activities did not differ between boys and girls (PLUM: p = .90; ANCOVA: p = .53).

Discussion

In this study, we investigated the link between spatial-play experiences and spatial performance using a large, diverse sample and controlling for general cognitive ability. With this control, boys outperformed girls, and children in the higher-SES groups outperformed those in the low-SES group. Gender differences in Block Design performance, which are not typically observed (Voyer et al., 1995), emerged only when we controlled for verbal comprehension, processing speed, working memory, and fluid reasoning. We confirmed that spatial play (i.e., playing with puzzles, blocks, and board games) is positively associated with spatial skill (i.e., Block Design performance) while controlling for other aspects of ability. This relation was highly specific: Other types of play and interactive parent-child activities were not related to Block Design score.

Gender differences emerged for reported frequency of play with puzzles, blocks, and board games. Reported frequency was higher for boys than for girls, which is consistent with previous findings with smaller samples (Farrell, 1957; Saracho, 1994; Tracy, 1987). The fact that our analyses were correlational limited our ability to determine whether both this gender difference in play frequency and the relationship between spatial play and spatial performance were causal; it is quite likely that spatial ability itself influences choice of play activities. However, interventions including spatial experience, and even just repeated spatial testing on its own, have led to improved spatial performance (Uttal et al., 2013), as have interventions with playful spatial activities (Casey, Erkut, Ceder, & Young, 2008; Casey, Andrews, et al., 2008; Grissmer et al., 2013). Thus, repeated spatial-play experiences, such as experiences with puzzles, blocks, and board games, may also lead to improvement, especially when those experiences provide feedback (e.g., when pieces of a puzzle do not fit together). Uttal et al. (2013) also presented evidence that the effects of spatial interventions are larger for people with lower initial performance. Though we again emphasize that further research is needed to assess whether playing with puzzles, blocks, and board games has a causal relationship with spatial performance, their finding does suggest that more frequent engagement in such play could have a positive effect on those children who are getting the least exposure—in this case, girls.

In contrast to play with puzzles, blocks, and board games, play with drawing materials was reported to be more frequent among girls than boys. Play with drawing materials might also foster spatial growth (see Caldera et al., 1999; Grissmer, Grimm, Aiyer, Murrah, & Steele, 2010; Tzuriel & Egozi, 2010; Wai et al., 2009), perhaps because of the demands on fine motor skills (Cameron et al., 2012). In our study, however, frequency of play with drawing materials did not relate to Block Design performance when we controlled for general cognitive ability. An important distinction between these activities—play with puzzles, blocks, and board games versus drawing materials—is that drawing is a two-dimensional activity, whereas building with blocks and playing some types of puzzles and board games involve three-dimensional objects and space. McGuinness and Morley (1991) found that gender differences in performance were not significant for a two-dimensional spatial task, but were present for a three-dimensional task, and the Block Design test included in this study was a three-dimensional task.

There were no SES differences in reported spatial play, although other researchers have seen such differences in actual observations of play (Levine et al., 2012). In fact, compared with children in the higher-SES groups, children in the low-SES group had higher reported frequencies of being taught number skills and shapes and playing math games with a parent—activities that could enhance spatial skills. However, that possibility was not supported in our analyses, perhaps because of the observed tendency of low-SES parents to report higher overall frequencies of parent-child activities across the five types of activity. Note that we observed no significant interaction effects of gender and SES on Block Design performance.

Although an advantage of the present study is its use of data from a large representative sample, working with large samples often entails using questionnaire data, which has associated limitations. The questionnaire items combined categories of play (e.g., blocks, puzzles, and board games were included in a single item) that might be important to differentiate. Further, parents’ responses could have been biased by their perception of the importance of each type of activity. Most important, we could not evaluate the quality of the materials used in spatial play or the quality of interaction with other people during spatial play. Levine et al. (2012) suggested that children might learn best from optimally challenging spatial toys. In that study, when children played with more difficult puzzles, their parents were more engaged and used more spatial language, which in turn improved the quality of the play, and the quality of play was related to spatial performance for girls. Children’s spatial skills are related to their own spatial language, which is influenced by the spatial language they hear (Pruden, Levine, & Huttenlocher, 2011).

Our results provide clear support for a relation between spatial play and Block Design scores. They also suggest that spatial play specifically is important for development of spatial skill, as nonspatial play had no relationship with Block Design scores. The relationship between spatial play and spatial performance was observed across SES levels and genders, though more spatial play was reported for boys than for girls. Although more research is needed to determine the mechanisms responsible for this relationship, these findings add strength to the hypothesis that spatial play could be an effective intervention for improving children’s spatial ability (Verdine et al., 2014). Both early education and spatial learning are important considerations for preparing children for success in STEM domains and for increasing the presence of underrepresented groups in these fields (e.g., women and minority populations), and spatial play is one potential method of addressing these goals.

Footnotes

Acknowledgements

We are grateful to Pearson for providing the data for this study.

Declaration of Conflicting Interests

The authors declared that they had no conflicts of interest with respect to their authorship or the publication of this article.

Funding

This work was supported by the Spatial Intelligence and Learning Center (National Science Foundation Grant SBE1041707).

Notes

References

Brosnan

M. J.

(1998). Spatial ability in children’s play with Lego blocks. Perceptual & Motor Skills, 87, 19–28.

Caldera

Y. M.

Culp

A. M.

O’Brien

Truglio

R. T.

Alvarez

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.

Cameron

C. E.

Brock

L. L.

Murrah

W. M.

Bell

L. H.

Worzalla

S. L.

Grissmer

Morrison

F. J.

(2012). Fine motor skills and executive function both contribute to kindergarten achievement. Child Development, 83, 1229–1244.

Casey

Erkut

Ceder

Young

J. M.

(2008). Use of a storytelling context to improve girls’ and boys’ geometry skills in kindergarten. Journal of Applied Developmental Psychology, 29, 29–48.

Casey

B. M.

Andrews

Schindler

Kersh

J. E.

Samper

Copley

(2008). The development of spatial skills through interventions involving block building activities. Cognition and Instruction, 26, 269–309.

Connor

J. M.

Serbin

L. A.

(1977). Behaviorally based masculine- and feminine-activity-preference scales for preschoolers: Correlates with other classroom behaviors and cognitive tests. Child Development, 48, 1411–1416.

Dehaene

Bossini

Giraux

(1993). The mental representation of parity and number magnitude. Journal of Experimental Psychology: General, 122, 371–396.

Estes

Felker

(2012). Confidence mediates the sex difference in mental rotation performance. Archives of Sexual Behavior, 41, 557–570.

Farrell

(1957). Gender differences in block play in early childhood education. Journal of Educational Research, 51, 279–284.

10.

Galton

(1880a). Statistics of mental imagery. Mind, 19, 301–318.

11.

Galton

(1880b). Visualised numerals. Nature, 21, 252–256.

12.

Grissmer

Grimm

K. J.

Aiyer

S. M.

Murrah

W. M.

Steele

J. S.

(2010). Fine motor skills and early comprehension of the world: Two new school readiness indicators. Developmental Psychology, 46, 1008–1017.

13.

Grissmer

Mashburn

Cottone

Brock

Murrah

Blodgett

Cameron

(2013, September). The efficacy of minds in motion on children’s development of executive function, visuo-spatial and math skills. Paper presented at the Society for Research in Educational Effectiveness Conference, Washington, DC.

14.

Groth-Marnat

Teal

(2000). Block design as a measure of everyday spatial ability: A study of ecological validity. Perceptual & Motor Skills, 90, 522–526.

15.

Jirout

Newcombe

N. S.

(2014). Mazes and maps: Can young children find their way? Mind, Brain, and Education, 8, 89–96.

16.

Levine

S. C.

Ratliff

K. R.

Huttenlocher

Cannon

(2012). Early puzzle play: A predictor of preschoolers’ spatial transformation skill. Developmental Psychology, 48, 530–542.

17.

Levine

S. C.

Vasilyeva

Lourenco

S. F.

Newcombe

N. S.

Huttenlocher

(2005). Socioeconomic status modifies the sex difference in spatial skill. Psychological Science, 16, 841–845. doi:10.1111/j.1467-9280.2005.01623.x

18.

Lippa

R. A.

Collaer

M. L.

Peters

(2010). Sex differences in mental rotation and line angle judgments are positively associated with gender equality and economic development across 53 nations. Archives of Sexual Behavior, 39, 990–997.

19.

McGuinness

Morley

(1991). Sex differences in the development of visuo-spatial ability in pre-school children. Journal of Mental Imagery, 15, 143–150.

20.

Miller

D. I.

Halpern

D. F.

(2014). The new science of cognitive sex differences. Trends in Cognitive Sciences, 18, 37–45.

21.

National Research Council. (2006). Learning to think spatially: GIS as a support system in the K-12 curriculum. Washington, DC: National Academies Press.

22.

Newcombe

N. S.

(2007). Taking science seriously: Straight thinking about sex differences. In Ceci

Williams

(Eds.), Why aren’t more women in science? Top gender researchers debate the evidence (pp. 69–77). Washington, DC: American Psychological Association.

23.

Newcombe

N. S.

Stieff

(2012). Six myths about spatial thinking. International Journal of Science Education, 34, 955–971.

24.

Pruden

Levine

S. C.

Huttenlocher

(2011). Children’s spatial thinking: Does talk about the spatial world matter? Developmental Science, 14, 1417–1430. doi:10.1111/j.1467-7687.2011.01088.x

25.

Ramani

G. B.

Siegler

R. S.

(2008). Promoting broad and stable improvements in low-income children’s numerical knowledge through playing number board games. Child Development, 79, 375–394. doi:10.1111/j.1467-8624.2007.01131.x

26.

Saracho

O. N.

(1994). The relationship of preschool children’s cognitive style to their play preferences. Early Child Development and Care, 97, 21–33.

27.

Serbin

L. A.

Zelkowitz

Doyle

Gold

Wheaton

(1990). The socialization of sex-differentiated skills and academic performance: A mediational model. Sex Roles, 23, 613–628.

28.

Siegler

R. S.

Ramani

G. B.

(2009). Playing linear number board games—but not circular ones—improves low-income preschoolers’ numerical understanding. Journal of Educational Psychology, 101, 545–560.

29.

Smith

I. M.

(1964). Spatial ability: Its educational and social significance. San Diego, CA: R. R. Knapp.

30.

Tracy

D. M.

(1987). Toys, spatial ability, and science and mathematics achievement: Are they related? Sex Roles, 17, 115–138.

31.

Tzuriel

Egozi

(2010). Gender differences in spatial ability of young children: The effects of training and processing strategies. Child Development, 81, 1417–1430.

32.

Uttal

D. H.

Cohen

C. A.

(2012). Spatial thinking and STEM education: When, why and how? In Ross

B. H.

(Ed.), The psychology of learning and motivation (Vol. 57, pp. 147–181). San Diego, CA: Academic Press.

33.

Uttal

D. H.

Meadow

N. G.

Tipton

Hand

L. L.

Alden

A. R.

Warren

Newcombe

N. S.

(2013). The malleability of spatial skills: A meta-analysis of training studies. Psychological Bulletin, 139, 352–402.

34.

Verdine

B. N.

Golinkoff

R. M.

Hirsh-Pasek

Newcombe

N. S.

(2014). Finding the missing piece: Blocks, puzzles, and shapes fuel school readiness. Trends in Neuroscience & Education, 3, 7–13. doi:10.1016/j.tine.2014.02.005

35.

Voyer

Bryden

M. P.

(1995). Magnitude of sex differences in spatial abilities: A meta-analysis and consideration of critical variables. Psychological Bulletin, 117, 250–270. doi:10.1037/0033-2909.117.2.250

36.

Wai

Lubinski

Benbow

C. P.

(2009). Spatial ability for STEM domains: Aligning over 50 years of cumulative psychological knowledge solidifies its importance. Journal of Educational Psychology, 101, 817–835.

37.

Wechsler

(2012). Wechsler Preschool and Primary Scale of Intelligence – Fourth Edition technical and interpretative manual. San Antonio, TX: Psychological Corp.

38.

Yang

J. C.

Chen

S. Y.

(2010). Effects of gender differences and spatial abilities within a digital pentominoes game. Computers & Education, 55, 1220–1233.

39.

Yang

Chen

Zhou

Dong

Chen

(2014). Development of spatial representation of numbers: A study of the SNARC effect in Chinese children. Journal of Experimental Child Psychology, 117, 1–11.