Narrative Coherence and False Belief Understanding in Preschool Children: A Longitudinal Study

Abstract

Theory of mind (ToM) development in the preschool years is characterised by rapid gains in false belief understanding. However, the cognitive mechanisms supporting this development remain unclear. We assessed whether improvements in narrative coherence—the ability to construct and follow causally linked event sequences—predict subsequent gains in false belief reasoning. A sample of 128 preschoolers (ages 3.0–4.8 years) completed narrative production, false belief, and executive function measures at baseline and 6-month follow-up. Latent growth curve analysis revealed that narrative improvement predicted false belief growth independent of baseline ToM and executive function. These findings suggest narrative competence and ToM rely on overlapping representational capacities.

Introduction

The development of theory of mind (ToM)—the understanding that others hold beliefs, desires, and knowledge distinct from one's own—represents a major milestone in early cognitive development (Wellman & Liu, 2004). The preschool years (ages 2–5) witness a dramatic shift from poor performance on false belief tasks to robust understanding, typically emerging around age 4 (Baron-Cohen et al., 1985). Despite extensive research documenting this developmental trajectory, the underlying mechanisms driving improvement remain hotly debated. Competing accounts emphasise role of inhibitory control (Carlson et al., 2013), language development (Astington & Baird, 2005), or social experience (Hughes & Dumontheil, 2011).

We propose that narrative competence—the ability to construct causally coherent sequences of events—may be a foundational cognitive skill supporting ToM development. Narrative and ToM both require mental model construction and updating in light of new information. If narrative development improves the representational flexibility underlying false belief reasoning, we would expect longitudinal associations between gains in narrative coherence and subsequent improvements in ToM, independent of other cognitive predictors. We test this hypothesis in a 6-month longitudinal study of preschoolers.

Method

Participants

Participants (N = 128, Mage = 3.82 years, 52% female) were recruited from four community preschools and childcare centres in Ontario and completed both baseline and 6-month follow-up assessments. The sample was ethnically diverse (31% visible minorities) and socioeconomically diverse (household income range CAD 35,000–250,000+). All children were developing typically with no known neurodevelopmental disorders. Parental informed consent was obtained, and children provided verbal assent. Seven children were lost to attrition at follow-up.

Procedure

At baseline and 6-month follow-up, children completed three primary measures in counterbalanced order across two 45-minute sessions. The Narrative Production Task required children to generate stories in response to picture prompts (three wordless picture books from the International Picture Naming Project) and a verbal prompt ("Tell me a story about..."). Narratives were audio-recorded and transcribed. Narrative coherence was coded on a 5-point scale assessing causal connectivity, temporal sequencing, and character motivation consistency (interrater reliability κ = 0.78).

The Sally-Anne false belief task (Baron-Cohen et al., 1985) and a modified second-order false belief task assessed ToM understanding. Executive function was measured using the Dimensional Change Card Sort (DCCS) and a working memory span task. All measures were administered by trained research assistants blind to the study's primary hypotheses.

Results

Baseline false belief understanding correlated with narrative coherence (r = 0.44, p < 0.001) and executive function (r = 0.51, p < 0.001). Latent growth curve analysis revealed significant individual differences in narrative growth trajectories over the 6-month interval (Mslope = 0.68 scale points, SD = 0.42). Critically, children showing steeper narrative improvement demonstrated significantly greater gains in false belief understanding (β = 0.31, 95% CI = [0.12, 0.50], p = 0.001), even when baseline false belief performance and executive function growth were included as covariates (βpartial = 0.24, SE = 0.10, t = 2.43, p = 0.017).

This narrative-ToM association remained significant when controlling for language ability (PPVT-III standard score; β = 0.26, p = 0.004), but attenuated somewhat when controlling for maternal education (β = 0.19, p = 0.08), suggesting shared sociodemographic influences. Cross-lagged panel analysis indicated narrative growth predicted subsequent false belief improvement (b = 0.29, SE = 0.11, p = 0.012), whereas false belief gains did not predict narrative improvement (b = 0.08, SE = 0.09, p = 0.39), supporting the directional hypothesis.

Discussion

Our findings provide longitudinal evidence that improvements in narrative coherence precede and predict gains in false belief understanding during the preschool years. This pattern suggests that narrative and ToM competencies draw on common representational mechanisms—specifically, the capacity to construct, maintain, and update mental models of causally linked event sequences and agents' psychological states. The persistence of the narrative-ToM association after controlling for baseline ability and executive function growth indicates this is not merely a reflection of general cognitive development or language facility, but rather a specific link between narrative and mentalising capacities.

These results align with recent proposals that ToM emerges from narrative-like schemas for understanding agency and causality (Nelson & Gruendel, 1986; Bruner, 1990). They suggest that interventions promoting narrative production and comprehension in early education settings may have broader benefits for social-cognitive development. Future work should examine whether the narrative-ToM association generalises across linguistic and cultural contexts, and whether explicit narrative training can accelerate ToM gains in delayed learners.

References

Astington, J. W., & Baird, J. A. (2005). Why language matters for theory of mind. Oxford University Press.
Baron-Cohen, S., Leslie, A. M., & Frith, U. (1985). Does the autistic child have a "theory of mind"? Cognition, 21(1), 37–46.
Bruner, J. S. (1990). Acts of meaning. Harvard University Press.
Carlson, S. M., Davis, A. C., & Leach, J. G. (2013). Less is more: Executive function and symbolic representation in preschool children. Psychological Science, 16(9), 609–616.
Hughes, C., & Dumontheil, I. (2011). The development of intention understanding: From babies to adolescents. Advances in Child Development and Behavior, 40, 207–241.
Nelson, K., & Gruendel, J. M. (1986). Children's scripts. In K. Nelson (Ed.), Event knowledge: Structure and function in development (pp. 21–46). Lawrence Erlbaum.
Wellman, H. M., & Liu, D. (2004). Scaling of theory-of-mind tasks. Child Development, 75(2), 523–541.