The Mental Representation Of Modern Chinese Verb-noun Heterosemous Lexemes From The Perspective Of The Two-level Word Class Categorization Theory

Word class categorization is considered as the study of “God particles” in language.The treatment of multifunctionality or heterosemy of lexemes is a hot and knotty issue among various linguistic theoretical camps,which makes studies on Modern Chinese word classes a conundrum.In the past few decades,psycholinguistic and neurolinguistic research has shown that verbs and nouns are processed differently in terms of the comprehension and recognition mechanism in Modern Chinese.However,whether the conventionalized verb-noun heterosemy at the lexical level in a communal language(i.e.langue)is represented in mental lexicon is poorly understood.Besides,existing studies on the factors affecting word recognition of verb-noun heterosemous lexemes have not been fully explained.Some scholars held the view that parts of speech or categorizes were not regulations of the lexicon but were derivatives of the syntactic structure under the framework of Distributed Morphology.Therefore,from the perspective of the Two-level Word Class Categorization Theory,this thesis aims to investigate what factors affect the recognition of verb-noun heterosemous lexemes with the help of behavioral techniques.The following questions are addressed:(1)Are native speakers able to identify the verb-noun heterosemous lexemes and non-heterosemous lexemes in their mental lexicon;(2)Does frequency affect the recognition of verb-noun heterosemous lexemes and non-heterosemous lexemes in native speakers;(3)Does context affect the recognition of verb-noun heterosemous lexemes and non-heterosemous lexemes in native speakers? In order to answer these questions,two reaction time experiments were carried out.Experiment One was an experiment without any contexts.Stimuli were composed of 40 typical verb word types(e.g.得到),40 typical noun word types(e.g.政府)and 40 verb-noun heterosemous lexemes(e.g.领导).35 postgraduates from SISU were chosen as participants and they were required to make a lexical decision task.Reaction times and accurate rates were recorded for data analysis.Experiment Two was a sentential context experiment.Stimuli were the same as those used in Experiment One and it added 240 sentences before the stimuli.Participants and tasks were the same with Experiment One.All the experiments were conducted in the Dictionary Research Center in Sichuan International Studies University.The experimental procedure was programmed by the software E-prime 2.0 and all the data were edited and analyzed by the statistic software SPSS 20.0.The statistical methods of repeated-measures ANOVA,post hoc comparison and t-test were employed to accomplish off-line data analysis.The results are as follows:(1)The results of Experiment One showed that the accurate rate of identifying verb-noun heterosemous lexemes,typical verb word types and typical noun word types had significant difference.Besides,the reaction times between verb-noun heterosemous lexemes and typical noun word types and between typical verb word types and typical noun word types had significant difference,whereas there was no significant difference between verb-noun heterosemous lexemes and typical verb word types.This result indicated that parts of speech are specified in the mental lexicon and participants were able to correctly identify verb-noun heterosemous lexemes and non-heterosemous lexemes.Besides,there was also significant difference respectively between the reaction time of high-frequency and low-frequency verb-noun heterosemous lexemes and between the accurate rate of high-frequency and low-frequency pairs,too.It can be referred from the results that word frequency can influence the recognition of Chinese verb-noun heterosemous lexemes.(2)The results of Experiment Two showed that under the sentential contexts,which included two contexts(in Context One target words were used as predicate and in Context Two target words were used as subject or object),participants were prone to identify the two disyllabic lexemes as typical verbs or typical nouns instead of verb-noun heterosemous lexemes.It indicated that participants ascertained a lexeme’s category according to its propositional speech act functions in syntax at parole.In addition,the reaction time of high-frequency and low-frequency verb-noun heterosemous lexemes revealed significant difference while the accurate rate of the high-frequency and low-frequency pairs revealed no significant difference,which suggested that word frequency can affect the recognition of Modern Chinese verb-noun heterosemous lexemes.(3)Compared Experiment One with Experiment Two,there was significant difference separately of reaction time as well as accuracy of verb-noun heterosemous lexemes under non-context and sentential context conditions,indicating that sentential contexts can influence the recognition of verb-noun heterosemous lexemes.Based on the results of the above mentioned experiments,the present study may carefully draw the conclusion that class membership of Modern Chinese word types including verb-noun heterosemous lexemes in lexicon in the communal language are specified in native speakers’ mental lexicon.Frequency of word types and context of word tokens can be regarded as important factors affecting the recognition of Chinese disyllabic words.This study not only helps to deepen the studies of word classes in Modern Chinese and other analytic languages,but also helps to solve the problem of word class labeling in Chinese and Chinese-foreign language dictionaries as well as the problem of POS tagging in Modern Chinese corpora.