Understanding the Influence of a Teachable Robot on STEM Skills and Attitudes

Arizona State University, Scottsdale AZ

Investigators

Abstract

This National Robotics Initiative project will develop a robotic learning environment for middle school geometry students where students who are novices in geometry will learn new concepts by tutoring a humanoid robot to manipulate its gestures and spoken prompts in response to student utterances and problem-solving actions. The project is based on the principle that the act of tutoring can lead to motivational benefits such as student engagement, positive attitudes toward the subject being studied, and increased confidence. In this application, the robot is simultaneously a tool that students can program and a social actor that intelligently responds. This research project will engage a broadly diverse population and is aimed at increasing the participation and retention of underrepresented groups in fields of science, technology, engineering, and mathematics (STEM). There are three components to the broader impacts of this project: 1) Scientific understanding of how robotic learning companions affect STEM attitudes and confidence, 2) Learning among students from traditionally underrepresented groups in the research process, and 3) Creation of a human-robot interaction platform for education and experimentation. The principles discovered through this project are expected to promote increased participation of women and other underrepresented populations in STEM educational activities and STEM-related careers. The goals of this research are to link robot behaviors to mediating motivational factors and STEM outcomes with the ultimate goal of understanding how to manipulate robot behaviors to improve a learning interaction. The proposed research will make contributions to understanding of how robotic learning environments can be designed to have a transformative impact on STEM learning. The research is guided by two questions: 1) How do robot behaviors influence students' mediating motivational factors and affect STEM skills and attitudes? And 2) What is the impact of adapting robot behaviors to student behaviors on mediating variables and STEM skills and outcomes? To find answers to these questions, the project will undertake three initiatives: 1) Further development of NaoTAG (Nao Tangible Activities for Geometry) to serve as a platform for experimenting with how different aspects of robot behaviors influence student-robot interactions within the teachable agent context; 2) Understand how student and robot behaviors influence the mediating factors that have identified, and STEM skills and outcomes; and 3) Modeling student-robot interactions.