The Influence Of Input Distribution On The Acquisition Of The Covariational Conditional Construction Under Different Learning Conditions

The usage-based theory is now a leading theoretical approach in linguistic studies.Guided by this approach,the usage-based language acquisition theory hypothesizes that the properties of input distribution have an impact on the language acquisition.Generally,there are two types of input distribution: skewed input and balanced input.In skewed input,more exemplars are created from one lexical item(i.e.,high token frequency)while the other lexical items occur less frequently(i.e.,low token frequency);in balanced input,the distribution of exemplars across the lexical verb or noun has equal token frequency.Studies on native language acquisition have demonstrated that the acquisition of a construction may be facilitated when the input is skewed input.However,the related research on second language acquisition found no advantage on the part of skewed input over balanced input.The effects of input distribution on language acquisition are still controversial,and one of the possible reasons is that the effect of input frequency distribution may be influenced by other factors such as explicit or implicit instructions.Moreover,the existing studies almost focused exclusively on verb argument constructions,leaving those peripheral constructions out of consideration.Thus,the present study attempts to explore the role of input distribution in the acquisition of covariational conditional construction under different learning instruction and see whether such effects can last a period of time.In order to achieve the above research objectives,the author adopted quantitative research method and designed a 2(skewed/balanced input distribution)×2(explicit/implicit instruction)×3(pretest,posttest,delayed posttest)teaching experiment on the acquisition of covariational conditional construction.A total of 150 Chinese middle school students in grade1 without relevant knowledge were randomly assigned to four experimental groups and a control group,with thirty people in each group.In the skewed input groups,the prototypical exemplar was presented 6 times with the other three example sentences presented twice.While in the balanced input groups,all four examples were each presented 3 times.In the explicit instruction groups,students were told to learn a new sentence pattern.In addition,the detailed grammar rules of this construction were explained to them before learning.Students in the implicit instruction groups learned the target construction only by repeating the sentence describing the picture without the relevant interpretation of grammar rules and semantic features.Before and after the intervention,the students were required to complete a Chinese-English translation task to test the output knowledge of the target construction,and to take a delayed posttest of the same type of task one week after the intervention.The results revealed 1)there was no significant difference between the influence of the balanced input distribution and the skewed input distribution on target construction acquisition in the posttest.2)the influence of the balanced input distribution on the acquisition of the target construction lasted for one week,yet that of skewed input distribution failed.3)explicit instruction seemed to better affect the acquisition of the target construction than the implicit instruction in posttest.4)explicit instruction did not show a lasting advantage in the maintenance of the acquisition effect in the delayed posttest.All in all,the study not only confirmed the influence of input distribution and the types of instruction on the acquisition of covariational conditional constructions by Chinese EFL students,but also provided us with useful inspiration for the acquisition of peripheral construction and the testing of usage-based language acquisition theory.