Information in content equivalent movie and text stories∗

New methods of data analysis are introduced to assess similarities and differences in responses of junior high and college students to the “same” story presented as movie vs. text. First, ways to determine to what degree a movie and text are the same in content are specified. Then, using a response task of ordering 23 photos or sentences from the story, an out‐of‐order score is introduced to quantify the amount of difference between the actual and recalled story order. The score is converted to bits of information, and a model of the response task, with four encoding and four problem solving parameters, is developed. Different amounts of information can be assigned to different parameters in the model. Children encode movie better than text, but adults encode them equally well. Adults are better problem solvers than children, but children have a transient text memory which helps them solve the problem of ordering sentences. An analysis using linguistic vs. pictorial cohesion explains why people make the errors they do in ordering.