| The metaphor is an important phenomenon of human beings, but modern cognitive science regards the metaphor as the ways of exploring the external world. In this study, we control the properties of the material itself by metaphor (including familiarity, typicality and conventional) and rank the stimulus by salient with levels of strong and weak. Through four event related potential experiments to investigate the cognitive processing of Chinese metaphor.Experiment 1 with double-word paradigm, through the analysis of strong and weak salient of metaphors and literal language, we found no difference between literal language and metaphor. While the level of salient play important role in metaphor comprehension that weak stimulus evoked more negative N400, and smaller P600. The results support the graded salience hypothesis (GSH) proposed by Giora that factors on degree of salient influence the metaphor processing. Even more, in the latter part of the semantic integration phase, there may be factors other than salient influence the metaphor processing.Experiment 2, based on the study of strong and weak degree of salient metaphor sentences, we compared the literal language, metaphor and error sentence by influence of semantic network distance represent by two domain:ontology and vehicle, which is correspond with target domain and source domain. Behavioral results suggest that salient influence language comprehension, and reaction time of the strong-literal, strong-metaphor and strong-error with a line increase trend suggested that not only the salient but sa also the semantic network distance influence the language comprehension. In addition, the ERP results also proved that level of salient and semantic network distance affect language materials processing. The mean amplitude of N200 once again proved dominant factor in language accessible early on the impact of language processing, low explicit material on an error sentence, metaphor sentences, straight linear decrease of semantic, once again illustrates the domain factors. With the results of N400 and P600 did not find the significant difference of sentence types and a significant main effect of degree of dominance, but with the ERP amplitude from right and left hemispheres were investigated, the high volatility of the strong-errorr sentence N400 was significantly higher than metaphor was straight and high justice, a large N400 effect that an error sentence of the source domain and target domain semantic distance away, and the high was in the middle sentence metaphors and right hemisphere brain regions induced the greatest P600, illustrates the semantic integration at a later stage, there has been difference between metaphor and processing differences. This difference may caused from semantic distance between source domain and target domain. Although the P600 amplitude of strong-metaphor was found larger in the right hemisphere, but combination the N400 and P600 results of the left and right hemispheres, we indicating that metaphor processing requires the two hemispheres involving, that in different time points a brain region may be dominant, but initiation in language processing, semantic accessibility, semantic integration of post-collaborative the whole brain is required to participate.In experiment 3, between the source domain and target domain, we import a modificability structure? and aim to investigate whether the ingredients of this modification will affect the nature of the source domain and target domain mapping, and then affect the metaphor of semantic integration. The result is that the modification of the composition makes the source domain and target domain mapping become more profit. When the source domain is active, may activate the domain associated with the intention of the source schema, the modification of the active ingredient is further directed to this target domain, making the source domain and target domain mappings automatically and quickly, this modification masks metaphorical meaning of the domain in the semantic network distance. Although in the N200 time window there some difference between metaphor and literal language, but the distance both in the N400 and P600 time window is not significant. In Experiment 3, although we have no adequacy reason to regard the modification structure as "context", but prompt us to focus on the the role of context. Parallel hypothesis assumption that metaphor and literal language processing are parallel, that with automatic activation and no need to refuse its metaphorical meaning. Even more, Gibbs suggested that metaphor could activate automatic with the situation of suited context. Although we can not confirm the same activation of ERP between literal language and metaphor is due to modification of the composition before vehicles which is the so called "context", the results of experiment 3 could confirm the role of context that impacted the semantic integration. The lager P600 which evoked by conflict from context suggested that although the metaphor of processing will affect by the dominance, but while the "context" appropriate, the source domain and target domain network distance may be masked by context, which with automatic and rapid completion of mapping, and direct access to the metaphorical meaning and integration.For experiment 4, taking into account of role of salient impacted domains between target and source, we use the visual half-field paradigm to examine the right and left hemispheres in the role of metaphor processing. By the end-word paradigm, subjects read the sentence of "A is B" form, one half of strong and weak stimulus presented in the left half of vision, and one half of stimulus presented in the right vision. The subjects read both the right and left vision sentence and make the wrong and right judgment. Behavioral results showed that reaction time of metaphor longer than literal language and weak stimulus longer than strong sentence. The view presentation impacted the brain activation, for N1 component, we found lager N1 in contralateral brain region. When use left visual field/right hemisphere presentation, the N400 amplitude is much larger than the right visual field/left hemisphere presenting. Those results illustrate the left hemisphere plays an important role in language comprehension. Also, the presentation paradigm impacted the role the salient. Those results suggested the two hemispheres in language processing there are some differences, but have not yet reached the degree of lateralization. For the processing of metaphor and literal language in terms of the main straight, straight in the left hemisphere on the meaning of metaphor and did not show a difference, but in the right hemisphere, the metaphor evoked greater N400 than literal language, which suggested that metaphor processing may require more processing than the literal language with the right hemisphere activation. We also found that the influence of salient, that strong-literal with fine processing with left hemisphere, and metaphor with coarse coding of right hemisphere, which support the hypothesis of Beeman's coarse semantic coding. The results only partially support the right brain hemisphere dominant of metaphor processing, that compared to literal language, the metaphor comprehension may require more activation in the right hemisphere, but for the metaphor itself, we found not difference between left and right hemisphere of metaphor processing. From this, we can infer that for language processing, the left hemisphere does have an important role, but the right hemisphere is also participated in language processing. The metaphor processing requires not only the right hemisphere activation and participation but also the left hemisphere, that required cooperation of two hemisphere activation.
|
PDF Full Text Request