DeeChee is an iCub robot who can learn language through interactions with humans. (Photo: PLoS One: Lyon et al.)

You might think that 'teaching' a robot really just comes down to programming it to do what we'd like. But new research, published in PLoS One, shows that humanoids learn a lot like, well, humans.   

The study showed that a small, child-like robot dubbed DeeChee, which is similar to a child between 6 and 14 months old, can learn rudimentary language skills that are not pre-programmed into the robot.

By having non-scientist volunteers spend a few minutes using blocks with colored sides to try to teach the artificial youngster words as if it were a toddler, the robot began to expand its vocabulary from "random syllabic babble to producing some salient wordforms, the names of simple shapes and colors," the authors said.

Like Us on Facebook

When DeeChee was able to pick out a word and repeat it back, participants reinforced that behavior with approving comments like "well done" or "clever," which provoked the robot to save the words it had just said as part of its vocabulary. You can see in this video how, with encouragement from a participant, the iCub robot learns the word "green."

"Learning needs interaction with a human, and robot embodiment evokes appropriate reactions in a human teacher, which disembodied software does not," computer scientist and study leader Caroline Lyon told Wired.

Basically, it's the back and forth with a human that produced this type of learning in the mechanized being. The outcome wouldn't have been the same if the robot were interacting with, say, a human face on a computer.

DeeChee's case is a study of embodied cognition, where one learns the building blocks of language through sensitivity to the frequency of sounds, reports GMA News Online. Human babies learn as their carers speak to them and they repeat sounds. That's exactly how DeeChee learned simple words, as well.

"Since our work concerns the acquisition of a human language by a robot we are inspired by the process in humans. Thus the basis of our experimental work is a real-time interactive situation where a human participant talks to a robot, using his or her own spontaneous words," the authors said.

Though the 8-minute experiments cannot compare to a child's immersion in a language-rich environment, Lyon said these results may help us understand how humans learn to speak.

"It is known that infants are sensitive to the frequency of sounds in speech, and these experiments show how this sensitivity can be modelled and contribute to the learning of word forms by a robot," she said, according to The Huffington Post UK.