Individual Submission Summary
Share...

Direct link:

Poster #111 - Pretend Play Predicting Children’s Language Competence

Thu, March 21, 2:15 to 3:30pm, Baltimore Convention Center, Floor: Level 1, Exhibit Hall B

Integrative Statement

During pretend play, an object can be used to represent a different object, inanimate beings can be treated as if they are animate, and imaginary objects and/or animate beings can be constructed without any tangible referents (Nielsen & Dissanayake, 2000). The relationship between language abilities and pretense has been proposed and mostly supported by examining language (e.g., vocabulary size, syntax) with standardized language tests and different types of pretense abilities including object substitution and play sequences (for a review see Lillard et al., 2013). In this longitudinal study, we ask whether (1) children’s pretense abilities predict their language competence (i.e., linguistic complexity) beyond their vocabulary knowledge when they are 4-years-old at Time-1 (T1) and 5-years-old at Time-2 (T2), and (2) Time-1 language abilities and Time-2 pretense abilities predict Time-2 language competence.

One hundred-twenty Turkish-reared children (Time-1: Mage=45.82-months, SD=6.01, Time-2: Mage=.57.68-months, SD = 5.88) participated in the study. Both at Time-1 and Time-2, they completed the telephone task (Mottweiler & Taylor, 2014), narrative creation tasks (Tahiroğlu, Mannering, & Taylor, 2011), and a vocabulary measure (Kazak Berument & Güven, 2013). In the narrative creation tasks, children’s use of complex clauses (i.e., adverbials, relative clauses, conjunctions, “if-then” statements, indirect speeches) and simple clauses were coded. The percentage of complex clauses with respect to the total number of clauses indicated linguistic complexity. Children’s pretense abilities were measured through the telephone task where children pretended as if they were calling their best friend and talking with them on a toy telephone.

Two hierarchical linear regression equations were conducted for Time-1 and Time-2 based on the same predictors and outcome variables (Model-1) (see Table 1). The Model-1 included T1-overall vocabulary and T1-age in the first step to predict T1-linguistic complexity as an outcome variable. The regression model was significant (R2=.116),F(2,94)=6.16,p=.003. Then in the second step, the addition of T1-telephone score increased the explained variance from 12% to 21% (R2=.206),F(3,93)=8.06,p<.001. For Time-2 measures, a similar regression analysis was performed. The results were in line with Time 1 measures, (R2=.194),F(3,90)=7.22,p<.001. When children were 4-years-old, pretense abilities predicted linguistic complexity over and above vocabulary competence, yet overall vocabulary knowledge was a significant predictor only at the age of five. Then, for the Model-2, in the first step, we added T2-age and T1-linguistic complexity to predict T2-linguistic complexity and the regression model was significant (R2=.094),F(2,83)=4.30,p=.017. The final Model-2 had T2-telephone score added as a predictor and the model was significant (R2=.203),F(3,82)=6.96,p<.001. When T2-telephone score was included in the model, T1-linguistic complexity no longer explained the variance, however T2-telephone score was a significant predictor (see Table 1).

The present study demonstrates how pretense engagement positively predicts language competence of children in a Turkish sample through linguistic complexity measures, which are detailed indicators of language abilities. Children who engaged in pretense more used advanced language while creating a narrative, producing more temporal, locative, causal information, logically related statements, and indirect speeches. These findings suggest the important role of pretense in children’s language development, particularly for complex linguistic components.

Authors