Child using screen technology
News Story

Study Finds Children’s Math Learning Improves by Interacting with Intelligent Characters

November 20, 2019 – Children in the United States are internationally ranked below children from other countries in science, technology, engineering and math (STEM). In an effort to improve education in STEM fields, Georgetown Department of Psychology Professor Sandra Calvert and her colleagues conducted research, published in Child Development, that examined whether children could learn math more easily if they interacted with a responsive media character. They found that children more quickly and accurately answered math questions when children thought of the character as their friend, known as a parasocial relationship, and when the character provided verbal feedback to the children about their math performance, a type of parasocial interaction.

 “Our study suggests that children’s relationships and interactions with intelligent characters can provide new pathways for 21st-century education, with popular media characters bridging traditional boundaries between home and school settings,” says Sandra L. Calvert.

Learning from Intelligent Characters

Professor Sandra Calvert

Professor Sandra Calvert

Researchers studied 217 children who were ages 3 to 6 years. They examined the children’s math learning from a game featuring a prototype of an intelligent character based on the media character Dora from the animated series, Dora the Explorer, who was programmed to respond to children with spoken language. In three studies, researchers began by asking if children could learn from intelligent character. They then examined the role of children’s parasocial, or one-sided, relationships by including or not including a character in the virtual game. Finally, they examined the role of social contingency, with some children’s talk about math receiving corrective feedback from the character and other children’s talk not receiving the feedback.

Children were taught the add-1 rule—that adding 1 to a number increases the total sum by a single unit—which is one of the most basic and earliest math concepts children learn. Researchers examined whether the children could learn this rule from an intelligent character in a virtual game, and how that learning was influenced by the children’s feelings for the character and their talk with the character. They also examined whether the children’s learning in a screen-based context would transfer to learning with physical objects, such as stickers.

Study Findings

Children who had stronger emotional feelings for the character and who talked more to the character about math had quicker, more accurate math responses during their virtual gameplay, the study found. Children also transferred what they had learned from the virtual game to physical objects more successfully when the game included an embodied virtual character (as opposed to a noncharacter female voiceover of what was said) and when the character used socially contingent replies to children’s talk about math. The findings suggest that children’s emotionally tinged parasocial relationships and parasocial talk about math with virtual characters increased their mastery of early math skills.

“Our work sheds light on how children’s connection to a character and interactions with them through math talk can improve learning of basic early math skills, a lesson that may be extended to other academic and social areas,” explains Evan Barba, associate professor of communication, culture, and technology at Georgetown University, who co-authored the study.

“The implication of our findings is that media characters that are children’s friends and playmates can also be children’s trusted peers and teachers in math and other subjects,” concludes Calvert.

The study was supported by the National Science Foundation. The authors are Sandra Calvert, Marisa Putnam, Naomi Aguiar, Rebecca Ryan, Charlotte Wright, Yi Hui Angella Liu, and Evan Barba.

-by Shelby Roller (G’19)

Tagged
Faculty
Faculty Research
Psychology
Research