In the Extension Test of Verb Learning (a), children trained via video chat and
live interaction looked significantly longer toward the matching action at test,
p < .01, whereas children trained through yoked
video performed at chance; their looking times to either side of the screen were
not different, p > .05. In the Stringent Test of Verb
Learning (b), conducted only for the Video Chat and Live Interaction conditions,
children looked longer toward the matching action during the Extension Test,
looked away during the new verb trial and then resumed looking to the matching
action during the recovery trial p < .01. Error bars
represent the standard error of the mean. ** p
< .01.