The Cognitive Mechanisms Of Chinese Natives And L2 Learners:Exploring The Implicature Processing Models

There is a hot debate on meaning processing model in pragmatics.Therefore,scholars have attempted to conduct experiments to assess which model better explains the real cognitive processing mechanism of utterance meanings.The purpose of this study is to participate in the debate on the accounts of the processing model among three models,the Context-driven model,the Default model,and the Standardization model.The Context-driven model predicts that the literal meaning is just a stimulus and the meaning processing totally depends on the context when people process implicature.The latter two models predict that people first process the literal meaning of an utterance,and then process what is implicated if there is a context.However,the difference of the prediction of the latter two models lies in the situational context.The Default model predicts that scalar implicature is inferred and then cancelled in the lower-bound context;thus,the reaction time of this context is significantly slower than that in the upper-bound context;while,the Standardization model predicts that scalar implicature is inferred in the upper-bound and lower-bound context and there exists no significant difference between the reaction time for the two types of contexts.All the participants are randomly chosen from the same population of Chinese native speakers(22-26 year old graduates at Lanzhou University,half from social science and half from natural science departments,half male and half female).They are divided into two groups,a Chinese native speaker group and a Chinese learner of English group.Each of them participates in only one experiment.This study consists of three experiments,which are as follows.Experiment 1 aimed to explore whether people processed the literal meaning of an utterance through a contrast of scalar implicature processing of the two groups without a context.The experiment referred to the materials of Experiment 1 of Slabakova(2010)without a context.The test materials for Chinese learners of English used his English materials,and those for Chinese native speakers used Chinese translation of the English materials.The distinction of this study from Slabakova’s(2010)lay in that his was an off-line experiment while mine was an on-line experiment.This on-line experiment presented test materials without a context through a computer system,DMDX.The participants were asked to judge right or wrong and then the system automatically recorded their accuracy rate and their reaction time.After the participants finished the on-line experiment,questionnaires and interviews followed.The results showed that,(1)the two groups could infer scalar implicature with higher accuracy rate;(2)there was a significant difference between the reaction time for the two groups and the Chinese native speaker group was faster than the Chinese learner of English group;(3)the results of questionnaires and interviews reported that the participants claimed no doubt about the experiment and made judgments properly.The result(1)indicated that the two groups inferred scalar implicature without a context and processed the literal meaning.The result(2)further demonstrated that the participants processed the literal meaning because the mental contexts of the two groups were the same.If the literal meaning was not processed and the participants depended on the context to infer implicature,the reaction time for the two groups should be the same.However,this result showed there was a significant difference between the reaction time for the two groups.The result(3)confirmed that the results of the experiment were reliable.From these indications,this study could make a preliminary conclusion that the Default model and the Standardization model might better explain the implicature processing of Chinese native speakers than the Context-driven model.Experiment 2 was intended to verify which model was a better account of the meaning processing of the two groups with the situational context,the Default model or the Standardization model.The prediction of these two models was similar in that people all processed the literal meaning,so this study predicted that the difference lay in the processing model with the situational context.Therefore,the test materials of Experiment 2 were presented with contexts.The contexts were demonstrated in pictures including 8 upper-bound and 8 lower-bound contexts(referring to Breheny et al.,2006),8 neutral contexts(referring to Katsos et al.,2005),8 control contexts and 8 fillers.The questionnaires and interviews followed after the on-line experiment.The results showed that,(1)there was no significant difference between scalar implicature accuracy rate and reaction time for the two groups;(2)the results of questionnaires and interviews displayed that it was hard for participants in the two groups to make judgment because they were interfered with test materials in the upper-bound context that were contrary to the reality.From these findings,the author became aware that there were two problems in test materials.(1)The sentence structures in Experiment 1("you" structure)were simple and consistent.Therefore,when designing context scenarios with complex structures in Experiment 2,this study just took the word number into consideration,while ignored sentence structures.(2)In the upper-bound context,logically true but pragmatically infelicitous "some" materials(for example,“Some elephants have trunks" is pragmatically wrong since the fact is that all elephants have trunks)influenced the participants’ judgment and needed to be revised.Therefore,Experiment 2 was redone.The test materials were revised in two important aspects.That is,the syntactic structures were unified and the logical meaning of materials was consistent with the reality.Experiment 3 was a repetition of Experiment 2 with the revision of the test materials.Before a formal experiment,a pilot study was conducted applying the revised materials.The formal test in Experiment 3 found that,(1)the reaction time for the upper-bound and lower-bound context of the two groups showed no significant difference and were both faster than that for the neutral context;(2)there was a difference between the accuracy rate and reaction time for the upper-bound and lower-bound context in the two groups while there was no significant difference in the neutral context;(3)the results of questionnaires and interviews revealed that the participants claimed no doubt about the experiment and made judgment properly.The result(1)indicated that the processing model of the two groups was in line with the Standardization model.The result(2)demonstrated that the Chinese learner of English group,due to the second language environment,could not process English literal meaning and linguistic form in the same way as the Chinese native speaker group with Chinese materials,which further illustrated that the participants needed to process the literal meaning.The result(3)confirmed that the results of this experiment were reliable.In sum,this study demonstrates that the Standardization model is a better model to explain the cognitive processing model of Chinese natives in interpreting Chinese and English meaning.