Word Frequency Effect And Polysemy Effect In Chinese Spoken Bisyllable Word Recognition

Spoken word recognition is one of most important area in language speech. Since the 1960s and 70s, people started the language cognition study and developed a series of hearing spoken word model. Most auditory recognition models are divided into two stages. The first stage is called the prelexical stage, speech input signals were transformed into phonemes or characteristics and began activating a number of candidate entries in the mental lexicon. The second stage is lexical stage; a number of candidate words were activated. Cohort model deem that, in the activation stage, top-down information was not given feedback, vocabulary recognition is modular and entirely dependent on a bottom-up source of information; TRACE model think that top-down information was processing in the beginning.However, Chinese are different from Western language. The morpheme in Chinese can provide more semantic information. So it is more interested in Chinese spoken word recognition to discuss whether a source involved in top-down information processed in different stages. The authors believe that the top-down information sources processed in the prelexical the thing that is depending on conditions. The study consists of five experiments, which is discuss whether the frequency and semantic information in a variety are processing in the prelexical stage.Experimental 1 discusses the time course of the Chinese two-word for the heating process, and found when the subjects can know the two-word. Experimental 2a proved that word frequency effects are related with the time course of word recognition. 2b experimental results show that in the hearing naming experimental, frequency information is processing in the prelexical, and through feedback to guide sensory input. 2 Experimental 2 results show that the former word frequency information will available in the processing. Experimental 3a shows that in the semantic judgment tasks, and only if lexical access, Polysemous semantic information that can affect the auditory word-processing. Experimental 3b results show that in the naming experiment, there was no polysemous effect in auditory processing. These results proved the idea which researchers expected, that in the prelexical stage of spoken word recognition, top-down information sources involved in the processing conditions, by the types and sources of information processing extent. The experimental results do not fully comply with TRACE and other interactive activation points or conform to the modular model points. Researchers believe spoken word recognition model of the theory should be integrated modules and interactive perspective, and then will suggest a more dynamic integration of the language processing model.