Making sense of it all : mapping the current to the past
Dennis, John Lawrence, 1973-
MetadataShow full item record
What are the representational differences between situations that do and do not map well onto previous experiences? This research offers some answers to this question by having participants compare two narratives that were either reality or fantasy-based. Fantasy-based narratives, with their deviations from reality, were considered similar to situations that do not map well onto previous experience. The concept of systematicity, where high-order relations constrain low-order relations was used to describe such situations (Bowdle & Gentner, 1997). Compared to a reality-based narrative, extra processing is required to maintain a systematic representation of a fantasy-based narrative. One can reduce the amount of processing needed by grounding that fantasy-based narrative in a reality-based or another fantasy-based narrative. Comparative judgments were used to measure processing differences. In three studies, participants read two narratives and then performed a series of comparative judgments derived from retrospective duration judgment (Block, 1992), event-structure perception (Zacks & Tversky, 2001), and structure-mapping theory (Gentner, 1983) research. For example, one of the comparative judgments adopted from structure-mapping theory was the rating of directional similarity, or the similarity rating of the second-read narrative relative to the first-read narrative. Directional similarity was proposed to increase as the amount of processing associated with maintaining a systematic representation of the first and second-read narrative decreased. For Studies 1A-E, the directional similarity was higher for the RealityFirst condition (reality read first) than the FantasyFirst condition (fantasy read first). These results are interpreted as indicated that the increase in directional similarity for the RealityFirst conditions was due to structure lending from the first-read reality-based narrative and that the decrease in directional similarity for the FantasyFirst conditions was due to representational disruption from the first-read fantasy-based narrative. Results also indicated that comparing two reality-based narratives (Studies 2A-B) was similar to comparing two fantasy-based narratives (Studies 3A-B) for the directional similarity and directional duration judgments, but differed for the listing of commonalities and differences and the segmentation of the narrative event structure. According to the systematicity principle (Gentner, 1989), people prefer mappings between two representations that form coherent and highly interconnected structures. The results from Studies 1A-E demonstrate a clear directional preference for the RealityFirst conditions. The results, therefore, indicate that it was more difficult to utilize the inherent structure of the narratives for the FantasyFirst conditions then the RealityFirst conditions. Comparing the results across the final set of studies, the increase in segmentation and increase in word count for the commonalities and differences were clear indications that participants still had difficulties in utilizing the structure of the narratives when both narratives being compared were fantasy-based (Studies 3A-B). In operationalizing systematicity with fantasy and reality-based narratives, I have been able to extend our understanding of how structure-lending can occur between these two narrative types. The results, therefore, extend our understanding of the structural alignment approach to narrative comparisons. But, since this research also involves the theoretical integration of the structure alignment approach (directional similarity and listing of commonalities and differences) with theories of time estimation (directional duration), event structure representation (segmentation), the basic findings herein should be applicable to comparisons ranging from auditory narrative structures to simple lexical units (e.g., unicorns vs. horses) to visual depicted objects (e.g., aliens vs. humans), even if the current set of studies described in this article involved only the comparison of written narrative structure.