Differences In Lexical Processing Between Generation And Recognition Of Nouns In Chinese

The brain mechanism underlying language production and comprehension is a heat question in cognitive science. In what order does the human brain process various linguistic information and what are the corresponding activated brain regions? As the development of cognitive neuroscience and formal linguistics, the study of language has deepened into the level of the intact brain. Our present study's main purpose is to investigate on the differences in lexical processing between generation and comprehension of concrete nouns in Chinese by means of event-related potentials and provide evidence for the non-linear processing of lexical information hypothesis. Question-answer task and lexical decision task were used in the experiment. We selected 70 concrete nouns in Chinese and designed 70 simple questions. Subjects were required to give a rapid answer only in their mind after they saw the questions. In lexical decision task, subjects were required to decide as quickly and as accurately as possible whether the word presented on the screen is a real word or a pseudoword. Results show that within 100-300ms post-stimuli, the N200-like component and the P300 were both elicited by generation and recognition of concrete nouns. In 300-400ms after the onset of stimulus, the ERP component elicited by generation of concrete nouns is the P350 while that elicited by recognition of nouns is the N362. Within 400-500ms, the amplitude of the N400 elicited by recognition of concrete nouns is more negative than that of the N400 elicited by generation of concrete nouns in parietal and parietal-occipital regions. The early N200-like component and the large N362 components observed in the current study indicates that semantic activation occurs before phonological activation in lexical processing and interaction between the two stages exists. The ERP differences are on broadly distributed cortical areas. Activation of the cortexes begins from the visual cortex in prefrontal and occipital lobes, transmits to the posterior temporal lobe of Wernicke's area, the frontal lobe of Broca's area and to the motor cortex in central-frontal lobe, later the activation in post-lexical semantic integration stage returns back to the posterior-occipital region. Furthermore, Broca's area and Wernicke's area demonstrate different functions in lexical processing. The results indicate that information processing of concrete nouns in Chinese is a non-linear continuous process, thus provide evidence for neurobiological theories of lexical processing, but the relationship between dynamic activation of corresponding brain regions in language processing and the time course of specific syntactic information processing is not clarified yet. The significant question in neurolinguistics calls for deeper investigation and studies in the future.