Comparison and Distinctiveness
In two experiments, we investigated the role of dimensional distinctiveness on the generalization of novel names for unfamiliar objects. In a comparison versus no-comparison design, we also manipulated within- and between-category comparisons with 4 and 6-year-olds. In the high-distinctiveness case, both age groups benefited from within comparisons in contrast to no comparison conditions. In the low-distinctiveness case, however, only older children benefited from comparison. Experiment 2 decomposed sources of distinctiveness along the two dimensions in order to determine which dimension must be distinctive in comparison format for better generalization. We interpret these findings in terms of within and between category similarities and differential costs of comparison for varying levels of distinctiveness. In Experiment 1, we used a forced-choice categorization task in which children had to decide which of two simultaneously presented objects would have the same name as the standard(s). Each child performed two practice trials followed by five experimental trials presented in random order. The design was a 4x2x2 with Comparison (within-and between-category comparison, 2-1, vs. within category-comparison, 2-0; vs. between-category comparison 1-1, vs. no comparison, 1-0), Age (4- or 5-year-old), and Distinctiveness (low distinctiveness vs. high distinctiveness) as between-subjects factors. The eight experimental conditions defining the materials were named according to the following logic combining condition of distinctiveness (High or Low), and condition of comparison (1-0; 1-1; 2-0; 2-1). For example, High-1-0 means high distinctive stimuli, one standard (no within-category comparison) and no contrast (no between-category comparison), or Low-2-1 means low distinctive stimuli, two standards (within-category comparison), and one contrast (see Table 1). We also used a forced-choice categorization task in Experiment 2. Each child performed two practice trials followed by five experimental trials presented in random order. The design was a 4x2 with Distinctiveness (High-Shape/High-Texture, HS/HT, High-Shape/Low-Texture, HS/LT, Low-Shape/High-Texture, LS/HT, Low-Shape/Low-Texture, LS/LT) and Age (4- or 5-year-old) as between-subjects factor. The four experimental conditions defining the materials were named according to each dimension’s distinctiveness: - In the High-Shape/High-Texture condition both shape and texture were distinctive - In the High-Shape/Low-Texture condition, the shape was distinctive but not the texture - In the Low-Shape/High-Texture condition, the texture was distinctive but not the shape - In the Low-Shape/Low-Texture condition, both shape and texture were not distinctive In the Data files, 1 corresponds to a response based on texture, and 0 corresponds to a response based on shape.