Abstract
Learning from news stories was greater when visuals and script were redundant than when they were not redundant. Adding continuous redundant captions, however, appeared to impede learning.