The Roles Of Segmental And Supra-segmental Information On Mandarin Spoken Word Recognition In 4 Years Old Children

We used eye tracking to examine the roles of segmental and supra-segmental information on Mandarin spoken word recognition in 4 years old children.Participants listened to target words(Experiment 1)or sentences(Experiment 2)while looking at displays of two pictures.The displays contained a target word(eg.Shu1,book)and four types of contrast words,which were cohort(eg.Shua1,brush),consonant(eg.Wu1,room),tonal(eg.Shu4,tree)contrast words,and one phonologically unrelated distracter(eg.Tao2,peach).These words are presented in the form of pictures to 4 years old children.The experiment has ensured that all images can be correctly named.In Experiment 1,target words(eg.Shu1,book)were presented to subjects through headphones.Experiment 2 is similar to experiment 1,only the Materials which the subjects heard were high constrain sentences(such as " The story book mom bought is very good ").A few of eye movements measures such as First Fixation Duration,Total Fixation Duration,Fixation Count and Percentage Fixated were analyzed.The results showed that:(1)In Experiment 1,we found familiarity effect and a main effect of the type of contrast words on the index of First Fixation Duration.We found there was no familiarity effect in Experiment 2.Only found a rhyme effect when the target words were low familiarity,means that the high constrain context plays an important role during early Mandarin spoken words recognition in 4 years old children;(2)On the index of Total Fixation Duration,we found that Experiment 1 and Experiment 2 had a similar result.No main effects were found here which means the high constrain context could barely influence the advanced processing of lexical access;(3)Tonal mismatch could restrict lexical activation in 4 years old children;(4)Word familiarity effect appeared in Experiment 1,there was no word familiarity effect in high constrain context;(5)Based on the above two experiments,draw the following conclusions: the current study supports for the TRACE model(It believes that auditory processing is more dependent on the overall consistency between the auditory input and the mental lexicon).