| dc.description.abstract | The current project combines two very different lines of research to investigate word finding difficulties of non-fluent aphasic patients in spontaneous speech. The first line is that of lexical retrieval in aphasic speech production. A literature review shows that the model by Levelt, Roelofs & Meyer (1999) can be used to accurately describe aphasic word finding difficulties. These difficulties likely arise due to reduced processing capacities. The amount of effort that is necessary to process words can be influenced by several variables, such as frequency, age of acquisition and entropy. The variable frequency provides the link to the second line of research, which is the distribution of word frequencies. The frequency distributions of natural language texts follow a power law called Zipf’s law. Deviations from this law have been found for different groups of patients. In the current study, Zipf’s law was investigated in the spontaneous speech of four non-fluent aphasic patients. Four speakers from the Corpus Gesproken Nederlands (Corpus Spoken Dutch) served as control group. Results show that speech from all patients conforms to Zipf’s law. A difference between the two groups was found in the slope of this distribution, which is due to the fact that aphasic speech shows a less varied vocabulary and larger groups of high frequency words. This finding is explained as indicative of an unimpaired lexicon and an adaptation to reduced processing capacities. A detailed suggestion for future research is provided, in which the disruptions found in speech from aphasic patients is hypothesized to be reflected in the numbers of lexical connections of the words that occur in their spontaneous speech. | |
