ResearchDetail1

Can young children learn about language from on-screen teachers?

Babies learn about language better from a real person than from video. Why is this the case?
One important difference between learning from a real person and from a video is the absence or presence of interaction.
So what happens if a screen is interactive?

In one of our studies, a virtual agent taught Japanese 12-month-old babies new object names. In one condition, the agent was interactive, for instance following the baby’s gaze direction. In the other condition, this was not the case.
We found that babies only learned new words from the interactive teacher, indicating that on-screen interactivity can enhance language learning.

However, in a study with 17-month-old French toddlers, we found different results. Toddlers here did not successfully learn novel words from the interactive virtual agent. They did, however, learn from a female teacher that interacted with them over video chat (though not quite as well as from a female teacher that interacted with them in the same room).
At this point we do not know whether this has to do with age or culture, and we are exploring this question in collaboration with a large network of researchers in the JEWEL project.

In sum, interactivity might be an important mechanism through which infants identify a relevant social agent, which might ultimately help them learn language.

You can read more about the topic of screens and language learning in our FAQ (#ScreenTime)

Right now, we are recruiting participants for a follow-up study on learning from an on-screen teacher (in Japanese). 

References

Tsuji, S.,Jincho, N., Mazuka, R., & Cristia, A. (2020). Communicative cues in the absence of a human interaction partner enhance 12-month-old infants’ word learning. Journal of Experimental Child Psychology, 191, 104740.

Tsuji, S., Fiévét, A.-C., & Cristia, A. (2021). Toddler word learning from contingent screens with and without human presence. Infant Behavior and Development, 63, 101553. doi: doi.org/10.1016/j.infbeh.2021.101553