A new dissimilarity measure for finding semantic structure in category fluency data with implications for understanding memory organization in schizophrenia

摘要

The ordering of words in category fluency lists is indicative of the semantic distance between\uditems in conceptual memory. Several studies have concluded from structural analyses of\udsuch data, using cluster analysis or multidimensional scaling, that the semantic memory of\udpatients with schizophrenia is more disorganized than that of controls. Previous studies have\udbased their analyses on a measure of average inter-item dissimilarity devised by Chan et al.\ud(1993). Here we derive a new and improved method of determining dissimilarity and show\udthat when this measure is applied to the fluency lists of patients with schizophrenia the\udaverage pattern of organization for the animal category has similar structure to that of\udcontrols, but with greater variability between individuals.