A code in the node: The use of a story schema in retrieval∗

Some characteristics of story schemata and their role in encoding and retrieval of stories are briefly described. Story schemata are described in terms of the story grammar outlined in Mandler and Johnson (1977). Using the rules of the grammar, canonical two‐episode stories were generated as well as versions whose surface structure violated the posited underlying structure by interleaving the events of the two episodes. Predictions were made concerning the quantity, quality, and temporal sequencing of recall for the standard and interleaved versions. Recall by second‐, fourth‐, sixth‐grade, and adult subjects was studied. Quantity of recall was less for the interleaved stories but more marked differences were found in the quality of recall; many more distortions and repetitions occurred in recall of interleaved stories. The most pronounced effects were found in sequencing of recall. Subjects hearing interleaved stories showed a strong tendency to recall stories in their canonical form rather than in the correct input order. This tendency was more pronounced for children. It was suggested that children are more dependent on familiar schemata for retrieval than are adults.