Category - Specific semantic deficits
Category - Specific semantic deficits
复制标题
DOI:
--
复制
发表时间:
2008-07
期刊:
影响因子:
--
通讯作者:
George S. Cree;K. McRae;Larry Barsalou;Mike Dixon;Jeff Elman;Albert Katz
中科院分区:
文献类型:
--
作者:
George S. Cree;K. McRae;Larry Barsalou;Mike Dixon;Jeff Elman;Albert Katz
Seven trends regarding the categories that tend to be impaired/preserved in category-specific semantic deficits were identified. The authors hypothesized that these trends arise despite the multiple sources of variation in patient testing because numerous factors that structure semantic memory probabilistically converge to make some categories of knowledge more susceptible to damage than others. Analysis of semantic feature norms and corpus data for 541 concepts revealed that differences in the distribution of knowledge types across categories are sufficient to explain 6 of the trends and are necessary to explain loss of knowledge about nonliving things. Feature informativeness, concept confusability, visual complexity , familiarity, and name frequency contributed to this patterning and provide insight into why knowledge about living things is most often impaired. From infancy on, human beings acquire vast amounts of knowledge about living and nonliving things from watching them, using them, observing others use them, and talking and reading about them. In a typical day, people watch animals move about; manipulate tools and utensils; prepare, smell, and taste food; and hear musical instruments. From these interactions, people develop internal representations that allow them to discuss and reason about these objects. This knowledge is often referred to as semantic knowledge, and the memory systems involved in the representation and processing of such knowledge are called semantic memory. People are amazed by the development of semantic knowledge in infants and troubled by its loss in debilitating conditions like dementia of the Alzheimer type, but for most of their adult lives, they simply take semantic knowledge for granted. In this article, we provide insight into how information about entities and objects is represented, organized, and computed. Our approach involved combining data generated from a large set of semantic feature production norms with constraints imposed by the patterns of impairments observed in patients with category-specific semantic deficits. The goal was to evaluate the efficacy of current models of semantic memory by assessing the major factors thought to structure semantic memory and influence semantic computations. Cases of category-specific semantic deficits are of particular interest to those studying semantic memory because regularities in patterns of impairment can be used to derive constraints on possible theories of semantic memory. Over 100 case studies have been reported, with deficits generally manifesting themselves as the loss of knowledge about items in some semantic domains (living things) relative to others (nonliving things). 1 Regularities within and across patients in …