Representations in Memory for Spatial Location

71 p.Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 2006.While spatial memory is crucial to our ability to function in the environment, we make systematic errors when locating a stimulus in relation to the region to which the stimulus belongs. For example, people incorrectly infer that San Diego, California is to the west of Reno, Nevada presumably because they apply their knowledge that California is generally to the west of Nevada (e.g., Stevens & Coupe, 1978). This dissertation aims to examine when and how these category-based distortions occur, focusing on the question of whether memories are truly distorted or whether distorted reports result from biasing processes operating on intact memories. It is clear that the bias is due to a blend between an exact coding of the target's location with the target's categorical coding but it is unclear if the blending occurs at encoding, during the delay, or at retrieval. The results of Experiments 1, 2, and 3 reveal that people, even after making errors in reproduction/recall, retain a veridical memory of the original location and are able to select that location after a short delay. Experiment 4 tested the locus of the blending by examining the impact of the originally encoded categorical information on judgments of location with the introduction of new categorical information at the time of retrieval. The results show that reports of location do not reflect the use of the originally encoded category, thus indicating that the memory for the target location is intact at the time of response. Results are interpreted as evidence for the unbiased nature of memories. Models of category-effect are discussed. Experiments 5 and 6 focus on the function of category information on the process of estimation of locations, and it is to increase overall accuracy when memory is imprecise.U of I OnlyRestricted to the U of I community idenfinitely during batch ingest of legacy ETD