This research program aims to further our understanding of processes that mediate language production and the breakdown of those processing aphasia. These broad aims will be addressed in the context of a more central questions in language processing theory that concerns the relation between semantic and phonological representations in lexical retrieval. At issue is whether the flow of information from conceptual to output systems proceeds serially and independently through semantic lexical and phonological retrieval stages, or whether feedback connections between levels of linguistic representation serve to modulate the output of the production system. A second theme of this proposal concerns the relationships between aphasic deficits and normal retrieval operations. One principle objective of the proposed research is to define explicitly the relationship between normal and impaired language function. Thus, to address the issue of interaction vs. independence in lexical retrieval, converging evidence will be sought in both normal and aphasic populations. In Part I experimental paradigms that manipulate the influence of semantic and phonological variables on naming errors and naming time will be used to test assumptions of the interactive model concerning the dynamics of lexical activation including the temporal course of semantic and phonological activation, effects of lexical similarity on substitution errors and word movement errors. In Part II, the model will be tested by examining its ability to account for error patterns in aphasics, including changes in naming and word repetition patterns over the course of recovery. This aspect of the project will involve assembly of a data base of generated errors that will be used in conjunction with computational data to test assumptions of the interactive model developed by Dell. In addition, we will explore relationships between naming, repetition of single words and repetition of strings of words in an attempt to apply the interactive framework to a broader range of phenomena that involve word retrieval. The data from the longitudinal case studies of aphasic patients will be interpreted with respect to their implications for models of normal and impaired language.
|Hula, William D; Fergadiotis, Gerasimos; Martin, Nadine (2012) Model choice and sample size in item response theory analysis of aphasia tests. Am J Speech Lang Pathol 21:S38-50|