Scientists have long been fascinated by how babies interact with adults. A new research, conducted by a team of roboticists, computer engineers and developmental psychologists at the University of California, San Diego, shows that babies smile intentionally to make the person, they are interacting with, smile in return. Often, this act of smiling is so perfectly timed that infants achieve their goal while doing very little themselves.
The NSF-funded research, which was recently published in the PLOS ONE journal, makes use of a baby-faced robot to study the behavioral patterns of non-verbal children and adults, especially people with autism. It is based on an earlier study, in which the facial expressions and interactions between 13 pairs of mothers and babies, aged less than four months, were carefully recorded and analyzed. Of them, 11 infants exhibited distinct signs of deliberate smiling. Speaking about the findings of the particular project, Paul Ruvolo, of UCSD’s Jacobs School of Engineering, said:
We thought either the babies had no goal or it was about mutual smiling. We are not claiming that a particular cognitive mechanism, for instance conscious deliberation, is responsible for the observed behaviors. Our methods are agnostic to this question.
For the last few years, Javier Movellan, a scientist at UCSD’s Machine Perception Laboratory, and his team have been working to build a robot that can help researchers to study human behavioral development in greater detail. The result is an incredibly realistic humanoid robot, named Diego San, which not only has an adorable baby-like face, but is also capable of accurately mimicking the gestures and expressions that are typical of infants. The team was reported saying:
If you’ve ever interacted with babies, you suspect that they’re up to something when they’re smiling. They’re not just smiling randomly. But proving this is difficult… Our goal was to have human development inform the development of social robots.
In the new research, Diego San was made to interact with 32 of the university’s undergraduate students, for a total of three minutes each. It was programmed to smile every time the undergraduates smiled. Upon observing the behavior of both parties, the scientists reached a similar conclusion: the toddler-like robot got the students to smile as much as possible, although it did very little on its own. Here, the students behaved like a mother would while interacting with her child. To discern the babies’s ulterior motive in making other people smile, the researchers used a special robotics tool known as optimal control theory. Using this, the team was able to reverse-engineer the infants’ goal from their behavior. Dan Messinger, of the University of Miami, explained:
What makes our study unique is that previous approaches to studying infant-parent interaction essentially describe patterns. But we couldn’t say what the mother or infant is trying to obtain in the interaction. Here we find that infants have their own goals in the interaction, even before four months of age.
The study’s findings will be presented at the upcoming Contextual Robotics Forum, at UCSD, on October 30.