The Role of Theory of Mind, Cognitive Flexibility, and Age in Children’s Developing Cross-classification Abilities

• In Event: Poster Session 14
  In Poster Session: PS 14 Topic - Cognitive Processes

Abstract

Children are constantly organizing information in their environment through the process of categorization (Markman, 1989). From a young age, children are able to categorize the same item into both taxonomic and script categories (i.e., cross-classification; Nguyen & Murphy, 2003). Cross-classification is a necessary skill as almost every item belongs to multiple categories. Previous research has indicated that children’s ability to separately and simultaneously cross-classify an item improves over the early elementary school years (Nguyen & Murphy, 2003; Nguyen, 2007). The current study aims to further our understanding of children’s cross-classification behaviors, by investigating two mechanisms, theory of mind and cognitive flexibility, that may underlie children’s abilities in this domain.

In the present study, we focus on 3- to 5-year-olds (N=75, Mage=4.52, females=35). We measured children’s cross-classification behaviors by asking children to first classify eight items into script categories and then to classify the same eight items into taxonomic categories. For example, children were first asked, “Are pajamas the same kind of thing as a blanket (script match) or candles (unrelated)?” In the second half of the measure, children were asked, “Are pajamas the same kind of thing as a sweater (taxonomic match) or a dog (unrelated)?” A composite cross-classification score was created based on accurate classification of the eight items into both categories (see Nguyen, 2007). We utilized the Theory of Mind Task Battery to measure children’s ToM (Hutchins, Prelock, & Chace, 2008) and the 3-Dimensional Change Card Sort to measure children’s cognitive flexibility (Deák & Wisehart, 2015). Scores on all three measures were converted into a proportional score out of one (see Table 1).

We employed multiple regression to analyze these data, using ToM, cognitive flexibility, and age as predictors of children’s cross-classification behaviors. The regression equation for this model was significant, F(3, 70)=9.244, p<.001, with an R2 of .284 (see Table 2). Age and ToM accounted for a significant proportion of the variance in children’s cross-classification, p’s=.008 and .035, respectively, but cognitive flexibility did not significantly contribute to the model beyond the effects of the other two predictors, p=.499.

In line with previous work, these data suggest that children’s cross-classification abilities improve between ages 3- and 5-years. We build on this previous work by finding that ToM is a significant predictor of children’s cross-classification behaviors. One explanation for this finding could be that some of the perspective shifting abilities used in ToM tasks are also involved in children’s cross-classification abilities. Interestingly, cognitive flexibility was not a significant predictor of cross-classification behaviors. However, this could be because children’s cognitive flexibility scores in the current study were approaching ceiling (M=.91 out of 1). In the future, it will be beneficial to include cognitive flexibility measures that tap into more advanced abilities (e.g., DCCS – Border Version). The present work furthers our understanding of what mechanisms contribute to children’s developing cross-classification behaviors and highlights the need to explore other cognitive mechanisms (e.g., broader executive functioning skills) that may account for additional variance in children’s cross-classification behaviors.

Authors

• Catherine H McDermott, Victoria University of Wellington
  Presenting Author

• Simone Nguyen, University of North Carolina at Wilmington
  Non-Presenting Author