Here at OnFiction.ca we’ve often discussed how empathy helps us to understand the minds of fictional characters. But readers don’t just work to understand the momentary thoughts of feelings of characters, they also form rich models or representations of what characters are like. Just as we can think of a close friend and predict how he or she might respond in a certain situation (e.g., a loud party) based on his or her personality, we similarly form expectations for story characters based on past behavior (either earlier in the story, or in a previous book). David Rapp and Richard Gerrig conducted research on this topic and published it in a paper that is now a classic of the field (Rapp & Gerrig, 2001). In their first study, readers read a series of stories in which characters behaved in ways consistent with one trait or another and then judged the plausibility of different story endings. An example story follows below:
Suzy planned on meeting her family at the park. They were having a huge family reunion that had been in the planning for months. The park was almost 40 minutes away. She knew she needed to leave soon if she wanted to get there on time. Along the way, Suzy came to a stop to let a baby squirrel cross the road. When she arrived at the barbecue Suzy saw a few family members she was looking forward to talking to. Pretty soon they were laughing and reminiscing about things they did when they were younger. Suzy was having a wonderful time talking about old memories, and she didn’t want the day to end. Her 8-year-old nephew William ran over and interrupted the conversation by tugging on Suzy’s dress. William asked Suzy if she would take him to play on the park’s swings.
After reading this story, they judged one of two possible outcomes:
Suzy took William to play on the swings. [trait-consistent outcome]
Suzy didn’t take William to play on the swings. [trait-inconsistent outcome]
Reader who read about Suzy being caring responded quicker to the story ending that was consistent with this model or representation of Suzy’s personality. A follow up, Study 2, found that readers applied trait-based models of characters in a specific rather than general way, indicating that they were judging individual characters based on particular traits rather than simply bad or good. In two additional studies, Studies 3 and 4, the researchers demonstrated that readers apply these models of characters during normal reading (as evidenced by their reading times) and that they are able to quickly update their models of characters as they learn new information. These studies demonstrate once again that many of the social processes that take place in the real world can also be applied to the fictional one.
Rapp, D. N. & Gerrig, R. J. (2001). Readers’ trait-based models of characters in narrative comprehension. Journal of Memory and Language, 45, 737–750.
* For a copy of this article, please contact Raymond Mar (e-mail address in profile). Apologies for the late posting.