| Chinese characters consist of semantic and phonological cues, with greater visual complexity than alphabetic scripts. Hence, this script serves as an interesting test area for questions regarding the linguistic factors involved in reading and lexical recognition. Previous reaction time studies of character and lexical recognition have examined general linguistic factors, particularly regarding the relative salience of phonology versus semantics for character recognition. These studies have found disparate results in comparing semantic and phonological processing, which might be due to insufficient controls, as relatively simple designs were used, without investigation of or controls for various orthographic, phonological, and semantic factors. This dissertation thus examines semantic and phonological information encoded in traditional Chinese characters, using survey data and masked priming experiments. The results indicate that previously uncontrolled linguistic aspects of characters are indeed accessed very early by readers, and thus should be taken into consideration in studies of visual character recognition.;First, a series of surveys were used to create semantic indices for over 3000 characters. These indices included judgments of the relatedness of semantic radicals to their visual forms (designated as semantic transparency); the semantic relatedness of radicals to character meanings (radical regularity); the variability of radical-character relatedness across characters with the same radical (radical consistency); and finally, the semantic concreteness of radicals and characters, i.e., semantic abstractness or tangibility. The data were normed to create semantic indices, and multivariate analytical techniques were applied to establish the contribution of various factors to the relative informational content of individual characters. Several new phonological indices were also created for a more detailed examination of phonological regularity, consistency, and frequency effects.;Next, a series of masked priming experiments was conducted, with the first goal being to determine whether the semantic indices proved to be significant factors contributing to character recognition. The second goal was to more closely examine the relative contribution of various semantic and phonological factors in early character recognition. In the first priming experiment, semantic concreteness, semantic consistency, and other factors were examined, and the second experiment examined a subclass of characters consisting of two semantic components. The third priming experiment examined the effects of phonological regularity and consistency of phonograms (phonetic components) for onsets and rimes, and phonological frequency effects. The final set of experiments tested these various factors with different prime durations, for a more detailed view of the various factors involved, and thus, a partial view of the various factors involved in the time course of character recognition.;The results of the priming experiments demonstrated the important of controlling for a number of factors, including the semantic indices and phonological variables, and the relevance of these factors that had not been previously considered or adequately controlled for in previous reaction time studies. These studies also indicate the complexity of Chinese character recognition, with various factors being operative simultaneously, and these studies served as an informative means of exploring such complexities at a newer level of detail. Implications are discussed regarding various models of visual word recognition and the stage theory of literacy development. |
