Theory of Mind and Pretend Play: Longitudinal Evidence from 18 to 36 Months

Abstract

Pretend play is thought to provide a cognitive scaffold for theory of mind (ToM) development, yet longitudinal evidence remains sparse. We tracked 96 children from 18 to 36 months, assessing pretense frequency and complexity alongside false-belief understanding and representational knowledge. Growth curve analyses revealed that early pretense predicts later ToM milestone attainment (β = 0.34, p < .01) and that this effect is mediated by symbolic gesture production. Results support a developmental pathway from early symbolic play to mature mentalizing capabilities.

Introduction

The developmental origins of theory of mind—the ability to attribute mental states to oneself and others—have been a central concern in developmental psychology (Wellman, Phillips, & Rodriguez, 2000). One influential hypothesis proposes that pretend play serves as a cognitive training ground for mentalizing (Leslie, 1987; Lillard et al., 2013). Children who engage in rich imaginative play may develop more flexible representational abilities, which in turn scaffold the understanding of beliefs, desires, and knowledge states in others. However, most empirical support for this "cognitive training" account derives from cross-sectional or correlational designs; longitudinal evidence tracking the developmental trajectory from early pretense to false-belief understanding is limited.

The period from 18 to 36 months is particularly informative, as this window encompasses rapid transitions in both symbolic play and early ToM reasoning (Tomasello, 1999). Understanding whether variations in early pretense prospectively predict later mentalizing achievements—and identifying the mechanisms linking these domains—has implications for educational practice and early childhood assessment. This longitudinal study examines whether pretend play frequency and complexity predict false-belief task performance, and whether this relationship is mediated by symbolic gesture production.

Method

Participants

Ninety-six children (Mage at intake = 18.3 months, SD = 1.1; 47 female; n = 89 completed full battery) from the Ottawa metropolitan area were recruited through community centres and daycare networks. Inclusion criteria: singleton birth, no reported developmental disorders, and fluent exposure to English. Median family income was CAD $65,000–$95,000; 72% of mothers held post-secondary education. Families were assessed at four time points: 18 months, 24 months, 30 months, and 36 months. Informed consent was obtained from all caregivers.

Procedure

At each visit, children participated in two 15-minute unstructured play sessions in a lab playroom equipped with age-appropriate toys (blocks, dolls, vehicles, pretense props). Sessions were video-recorded and coded offline by trained raters blind to outcome measures. Pretense frequency was operationalized as the total number of distinct pretend acts (e.g., feeding a doll, driving a toy car, using an object as substitute); complexity was rated on a 4-point scale assessing planning, role adoption, and object substitution. At 30 and 36 months, children completed a battery of ToM tasks: two standard false-belief tasks (Wellman & Liu, 2004), a representational change task, and three trials of diverse desires (Wellman & Liu, 2004). Symbolic gesture production was coded from the play video and mother-child interaction clips using the coding scheme of (Acredolo & Goodwyn, 1988). Parental education and family income were obtained via structured interview.

Results

Linear growth curve analyses were fit using HLM 7.0 (Raudenbush & Bryk, 2002) with time (centered at 24 months) as a within-person predictor. Pretense frequency increased significantly across the interval (γ10 = 1.28 trials/month, SE = 0.31, t(87) = 4.14, p < .001), with substantial between-child variation in intercept (τ0 = 89.3, SD = 9.4) and slope (τ1 = 0.47, SD = 0.69). At the 36-month assessment, 79 children (88.8%) passed at least one false-belief task. A multilevel logistic regression predicting false-belief success from early (18–24 month) pretense frequency, controlling for parental education and vocabulary, revealed a significant positive effect of pretense (γ = 0.34, SE = 0.12, z = 2.83, p = .005). A mediation analysis using the product-of-coefficients method (Sobel test, two-tailed) indicated that symbolic gesture production at 24 months significantly mediated this effect (indirect effect = 0.11, SE = 0.05, 95% CI [0.02, 0.23], p = .016), accounting for approximately 32% of the total association between early pretense and later ToM performance.

Discussion

These longitudinal findings provide prospective evidence that variation in early pretend play predicts subsequent theory of mind achievement. Importantly, the effect is not direct; rather, symbolic gestures at 24 months appear to constitute a developmental bridge linking toddler pretense to later false-belief reasoning. This pattern aligns with neuroconstructivist accounts of ToM development (Westermann, Mareschal, Johnson, Sirois, Spratling, & Thomas, 2007) and supports the hypothesis that symbolic play cultivates the flexible representational competencies necessary for mentalizing.

The observed association between parental education and pretense complexity (r = .41, p < .001) underscores the role of socioeconomic and cultural factors in play-based cognitive development. Future longitudinal research should examine how variation in pretense is shaped by the linguistic and social input available in the home environment, as well as whether interventions aimed at enriching imaginative play yield downstream improvements in theory of mind reasoning. Additionally, the specific contributions of solitary versus social pretense remain an open question, warranting detailed observational analysis in future studies.

References

Acredolo, L. P., & Goodwyn, S. W. (1988). Symbolic gesturing in normal infants. Child Development, 59(2), 450–466.
Leslie, A. M. (1987). Pretense and representation: The origins of "theory of mind." Psychological Review, 94(4), 412–426.
Lillard, A. S., Lerner, M. D., Hopkins, E. J., Dore, R. A., Smith, E. D., & Palmquist, C. M. (2013). The impact of pretend play on children's development: A review of the evidence. Psychological Bulletin, 139(1), 1–34.
Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.). Sage Publications.
Tomasello, M. (1999). The cultural origins of human cognition. Harvard University Press.
Wellman, H. M., & Liu, D. (2004). Scaling of theory-of-mind tasks. Child Development, 75(2), 523–541.
Wellman, H. M., Phillips, A. T., & Rodriguez, T. (2000). Young children's understanding of perception, desire, and emotion. Child Development, 71(4), 895–912.
Westermann, G., Mareschal, D., Johnson, M. H., Sirois, S., Spratling, M. W., & Thomas, M. S. C. (2007). Neuroconstructivism. Developmental Science, 10(1), 75–83.