Researchers Apply Observed Infant Social Skills to Baby Robot

DailyTech - UM, UC Researchers Apply Observed Infant Social Skills to Baby Robot


Baby robot, Diego-San, could help researchers better understand the development of human learning

University of Miami developmental psychologists and University of California (UC) San Diego computer scientists have studied infant-mother relationships in an effort to better understand human cognitive development, and to eventually create a realistic baby robot.

Daniel Messinger, lead author of the study and associate professor of Psychology in the UM College of Arts and Sciences, and Paul Ruvolo, co-author of the study and graduate student in the Computer Science Department at UC San Diego, have observed infant-mother relationships in hopes of applying early communication skills to a robotic baby. This sort of design could help researchers further understand regarding the development of human learning.

The study consisted of 13 mothers and 13 babies between the ages of one and six months old. The mothers and their babies were observed and recorded while playing together for five minute intervals weekly. There were about 14 sessions per day.

These observations of the infant-mother relationship concluded that infants learn turn-taking skills during their first six months of life. Also, researchers found that patterns form during the mother-infant playtime, which become more stable patterns over the years as the child grows.

"As babies get older, they develop a pattern with their moms," said Messinger. "When the baby smiles, the mom smiles; then the baby stops smiling and the mom stops smiling, and the babies learn to expect that someone will respond to them in a particular manner. Eventually the baby also learns to respond to the mom."

Now researchers are taking what they've observed from human infant-mother relationships and applying it to a baby robot. The idea is to give the baby robot, whose name is Diego-San, social skills and the ability to learn. Researchers also want Diego-San to mimic human infant movements, such as shifting their gaze from people to objects.

"One important finding here is that infants are most likely to shift their gaze, if they are the last ones to do so during the interaction," said Messinger. "What matters most is how long a baby looks at something, not what they are looking at."

Diego-San is 1.3 meters tall and was modeled after a one-year-old child. This baby robot was built by Kokoro Dreams and researchers at the Machine Perception Laboratory at UC San Diego.

"A unique aspect of this project is that we have state-of-the-art tools to study development on both the robotics side and the developmental psychology side," said Ruvolo. "On the robotics side, we have a robot that mechanically approximates the complexity of the human motor system, and on the developmental psychology side, we have a fine-grained motion capture and video recording that shows the mother-infant action in great detail.

"It is the interplay of these two methods for studying the process of development that has us so excited."

This study was published in Neural Networks.