| As a figure of speech,transferred epithet is commonly used in literary works as well as daily life.The studies on transferred epithet have so far concentrated on its definition,classification,aesthetic function,mental basis,etc.Previous studies have laid a solid basis for the understanding of transferred epithet.However,further efforts are still needed to obtain a deep understanding of the construal mechanism of transferred epithet.In addition,no empirical study on transferred epithet has been found.Liao Qiaoyun(2015)puts forward Annotation-Denotation Relevance-Inheritance Model(ADRIM)to explain the construal mechanism of Chinese semantic rhetoric,and holds that a process of possible feature derivation exists in the construal of Chinese transferred epithet.Therefore,a new approach to study the construal of Chinese transferred epithet has been provided.Besides,with the development of scientific techniques,ERPs(Event-Related Potentials)technique offers a new empirical method to study language comprehension.Therefore,this study will adopt ERPs technique to testify the Annotation-Denotation Relevance-Inheritance Model proposed by Liao Qiaoyun(2015).The research question is: Is there any electrophysiological evidence to support the psychological reality of possible feature derivation in the construal of Chinese transferred epithet?The experiment is conducted in the Key Lab of Cognitive Neuroscience and Foreign Language Learning in SISU,Chongqing,with 20 postgraduate students from SISU participating in the experiment.It is a one-factor within-subjects design with three levels(types of critical sentences: transferred epithet sentences,literal sentences and nonsensical sentences).Reaction time and ERPs data are used as indexes in the present study.Materials in the experiment are Chinese and are presented in randomized order.First,subjects are required to read a discourse made up of a context sentence and a critical sentence carefully and thoroughly.Next,subjects are presented with a probe word and are asked to judge whether the probe word is a real word or a pseudo word.At last,subjects are presented with a declarative sentence and need to judge whether the meaning of the declarative sentence is consistent with the previous discourse.E-prime 2.0 is used to program the experiment.When subjects perform the lexical decision task,their behavioral data and ERPs data are recorded and then off-line analyzed by using the software Neuroscan4.5,and at last the data are further analyzed by statistic software SPSS16.0.The research obtains the following results:(1)According to the behavioral data,the reaction time of the probe words is909.48 ms in transferred epithet sentences,946.49 ms in literal sentences and1123.35 ms in nonsensical sentences.There is a significant difference between reaction time of probe words in transferred epithet sentences and that in nonsensical sentences [t(18)=-2.903,p=.013<.05];The reaction time of probe words in literal sentences and that in nonsensical sentences shows marginal difference [t(18)=-2.151,p=.053 ≈.05].However,the difference between reaction time of probe words in transferred epithet sentences and that in literal sentences is not obvious [t(18)=-.983,p=.345>.05].The accuracy rates of probe words in three sentence types are high and show no significant difference.(2)According to the ERPs data,significant difference exists between the amplitude of N400 elicited by the probe words in transferred epithet sentences and that in literal sentences [F(1,17)=5.557,p=.036<.05],and N400 amplitude elicited by probe words in transferred epithet sentences(-2.221±5.675)is larger than that elicited by probe words in literal sentences(-1.211±5.086).The difference lies in the middle frontal region [t(18)=-2.704,p=.019<.05],left frontal region [t(18)=-2.305,p=.040<.05],middle central region [t(18)=-2.349,p=.037<.05],left central region[t(18)=-3.151,p=.008<.01],and right central region [t(18)=-3.109,p=.009<.01].Besides,there is a significant difference between the N400 amplitude elicited by probe words in nonsensical sentences and that elicited by probe words in literalsentences [F(1,17)=6.204,p=.028<.05],and N400 amplitude elicited by probe words in nonsensical sentences(-2.992±5.933)is larger than that elicited by probe words in literal sentences(-1.211±5.086).The difference lies in the middle frontal region[t(18)=2.942,p=.012<.05],left frontal region [t(18)=2.549,p=.026<.05],right frontal region [t(18)=2.334,p=.038<.05],middle central region [t(18)=3.011,p=.011<.05],left central region [t(18)=3.150,p=.008<.01],and right central region[t(18)=2.448,p=.031<.05].(3)The N400 effect is different in the left and right hemispheres.The t-test results demonstrate that the N400 effect is significant both in the left hemisphere and the right hemisphere,but the effect is more evident in the right hemisphere[t(18)=-3.510,p=.001<.01].Both left and right hemispheres are involved in the construal of Chinese transferred epithet sentences,and to be more specific,the right hemisphere plays a more important role.Based on the above results,the present study draws the following conclusion:The behavioral data show that reaction time of probe words in transferred epithet sentences is shorter than that in literal sentences,which proves that subjects successfully derive the possible feature in processing transferred epithet sentences,and there exists psychological reality of possible feature derivation in the comprehension of transferred epithet sentences.The ERPs data demonstrate that the differences of processing probe words in transferred epithet sentences and literal sentences lie in frontal and central region and in the left and right hemisphere.The results of present study provide the electrophysiological evidence for the possible feature derivation in processing Chinese transferred epithet sentences and support the Annotation-Denotation Relevance-Inheritance Model proposed by Liao Qiaoyun(2015). |
PDF Full Text Request