Understanding the experiences that shape the development of spatial reasoning is important, as it is related to academic achievement and participation in STEM fields (Newcombe, 2010). Play with puzzles, a common experience for many children, contributes to the development of spatial concepts, through hands-on experience manipulating spatially relevant objects and through the use of spatial language (Levine et al., 2012). However, these types of experiences may be changing as touchscreen devices become more commonly used for play, and there is very limited research that investigates this relationship.

This dissertation addresses the gap by comparing 6-year-olds’ play with tangram puzzles using tangible or digital pieces. Specifically, I investigated whether playing with a tangible tangram set or one of two digital sets with different mechanics (e.g., tapping vs. twisting) would differentially affect children’s spatial reasoning. In addition, I investigated whether parental use of spatial language would differ between tangram sets. Sixty parent-child dyads participated in the study and were randomly assigned to play tangram puzzles using one of three different forms: a traditional, tangible set (Tangible condition), a digital set where they rotated pieces using two-fingers and a twisting motion (Digital-Rotate condition), or a digital set where they rotated pieces by double tapping them (Digital-Tap condition). Children’s spatial reasoning was measured before and after play. In addition, two independent raters coded parents’ use of spatial language during the play session.

The results revealed that children in the Tangible condition showed greater improvement than those in the Digital-Tap condition on items that required diagonal rotation. In addition, parents in the Digital-Tap condition used a higher proportion of deictic words than parents in the Digital-Rotate condition. Additional exploratory analyses examined children’s everyday use of touchscreen devices and revealed that children predominantly use devices by themselves, without the presence of their parents. Together, these findings add to the growing body of research that examines the impact of touchscreen use on child development and bears practical implications for the design and use of touchscreen games in early childhood.

Detecting others' preferences for a relational category of objects (e.g., taller objects) may be difficult, because doing so requires comparing within and between object pairs to extract the common relation. Extending from Duh and Wang (2014), the present research examined the conditions under which infants detect such preferences. In Experiment 1, an experimenter chose the taller of two objects in three examples and the shorter object in one example; the order of this last example was manipulated. In test, the experimenter chose either the taller or the shorter of two novel objects. Fourteen-month-olds detected the experimenter's overall preference for taller objects only when the inconsistent example occurred after three consecutive consistent examples. In Experiment 2, 11-month-olds detected the preference with three consistent examples, but only if they had made a choice prior to the experiment. Together, the results highlight the role of mixed data and action experience in infants' preference reasoning.

Relational reasoning lies at the core of math and spatial learning. Like other cognitive abilities, relational reasoning is intertwined with one’s own cultural experiences. Growing evidence has shown cross-cultural differences in attention to relation and object in East Asian and North American participants (Kuwabara & Smith, 2012; Masuda & Nisbett, 2001). However, our understanding of possible mechanisms beyond the cultural variations is limited. The present study aimed to fill this gap by examining how mothers from first-generation Chinese immigrant and European-American cultural backgrounds use relational language with their 3-year-olds in math-related activities in everyday contexts (e.g., home). Specifically, we examined parental use of three types of relational language (e.g., structural relation, object similarity, self-association) in three activities (e.g., 3-D puzzle play, sorting activity, book reading). Additionally, we examined cultural variations in parental language that directs children’s attention to labeling objects or events. As a secondary question, we investigated the correlation between parental relational language and children’s math performance in four tasks (e.g., modified TOSA, matching dots, counting, give-n-task). Twenty European-American (English-speaking) and 16 first generation Chinese immigrant (Chinese-speaking) families were observed mostly at home. Results showed that the Chinese immigrant mothers used a significantly higher proportion of structural relation in the puzzle and the sorting activities than European-American mothers. Chinese immigrant mothers also attended more to labeling objects or events than their counterparts. Generally, the mothers cross two groups used more structural relation in the puzzle and the sorting activities whereas they used more self-association in the book reading activity. No significant correlations were found between parental use of relational language and children’s math performance in the four tasks. Together, the findings suggest a culturally specific way of engaging young children in relational learning by Chinese immigrant mothers, and offer important insights into how parental language practice may serve as a potential mechanism of early childhood math and spatial learning.

Robotic toys are more and more present in the lives of today’s children. The present research sought to address two research questions: (1) How preschool-age children conceptualize a robotic toy and (2) how parents talk with children about the robotic toy. Parent-child dyads were randomly assigned to watch videos of a contingent robot that followed directions (Contingent condition) or a non-contingent robot that did not follow directions (Non-Contingent condition). After watching the videos, children answered questions about the robotic toy (Interview 1), dyads participated in a short conversation, and children answered more questions about the robotic toy (Interview 2). Results suggest that children's judgment of the toy in the first interview was not related to the condition to which they were assigned. However, in the second interview, a significant difference was observed in children's judgment of the communicative ability of the toy. Children in the Contingent condition were more likely to say the toy could hear. During the parent-child conversation, more dyads in the Non-Contingent condition discussed the toy’s contingent behavior. This exploratory work provides insight into how parents help their young children make meaning out of their experience with a novel robotic toy.