UBC Theses and Dissertations
Effects on prose recall of learner’s prior knowledge and name versus attribute organization Richter, Joan Harriet Hedge
This study examined the effects on prose recall of two variables. The first variable was the reader's prior knowledge of the dimensional variation of the multidimensional stimuli described in the text materials. The variable was defined in terms of familiarity and unfamiliarity. Descriptions of chess pieces were unfamiliar because subjects were not knowledgeable about the game of chess; descriptions of fish were familiar because subjects were adults and assumed to be cognizant of dimensions used to describe living organisms such as habitat, diet, etc.. The second variable was the type of organization of the text materials, name organization or attribute organization. Name organization listed all the attribute values for each instance in a separate paragraph. Attribute organization listed all attribute values for each attribute in a separate paragraph. There were five attributes and six instances for both the chess and the fish decriptions. It was predicted that familiar materials would be recalled better under attribute organization than under name organization because the multidimensional discrimination space was presumably established for these materials and could be easily accessed. Conversely, unfamiliar materials would be recalled better under name organization than under attribute organization because presumably the multidimensional space was not established; it was hypothesized that "bundles" of attributes associated with a name would lead to better recall under these conditions. The results showed a significant interaction between prior knowledge and text organization, but in the opposite direction: Attribute organization leading to better recall than name organization with unfamiliar materials and name organization leading to better recall than attribute organization with familiar materials. It was suggested that the observed interaction was due to the two sub-processes in the multidimensional discrimination space model. The first sub-process establishes the dimensional space; this is facilitated by attribute organized text. The second sub-process embeds instances in the dimensional space; this is facilitated by name organized text. This would account for the overall superiority of familiar materials over unfamiliar materials. However, the number of attributes involved may also be a variable which interacts with prior knowledge and text organization.