The Anatomy Of Categorization: A Review Of The FMRI And ERP Study

Categorization is “classification”. ERP and fMRI technologies are regarded as “the window on advanced functions of the brain”. Few linguists apply the ERP/fMRI technology into the research on categorization.This thesis mainly focuses on three research questions: 1) from the perspective of fMRI, what specific brain regions are activated when the semantic knowledge of a specific category(animal/tool) is extracted? 2) From the perspective of ERP, when extracting the semantic knowledge of one category(animal/tool), which category and which level occurs faster than the other category during categorization? 3) From an interdisciplinary perspective, what empirical results can support linguistic theories and what empirical results can support neurological theories on categorization?This thesis, first of all, collected lots of literature on categorization in terms of ERP and fMRI, selected the most representative research as references and selected the categorization of animal/artifact, categorization on three levels and social categorization as the research objects. Afterwards, the spatial relationship through fMRI and the temporal relationship through ERP between the brain and categorization are thoroughly explored. Finally, the fMRI and ERP empirical results are combined with the theories and hypotheses in the field of linguistics and neuroscience.Through induction and summary, the findings are: 1) there indeed exist the specific brain regions activated by specific categories. Such as, animals mainly activate the superior temporal sulcus(STS), the inferior occipital lobe and inferior occipital gyrus; tools, mainly activate the posterior middle temporal gyri, medial posterior fusiform gyri, ventral premotor and the intraparietal solos; 2) there indeed exists a different time course during categorization, such as ERP response to animals is faster than to artifacts; “the base-level categorization” is not the earliest, however, as to which level of categorization is the earliest, there still exist disagreements; 3) there indeed exist the ERP and fMRI evidences to support the theories on categorization in both fields. Applying the “rule” often activates the frontal and temporal regions; “prototype”, parietal regions; thus, this thesis holds that evidences can be found through comparing the activated brain regions in ERP and fMRI research with the brain regions activated by the rule, prototype. However, multiple regions are activated during categorization, which also demonstrates that categorization is a complex cognitive activity, sometimes not only applying the rule or prototype.To sum up, this thesis is of significance for the future research on categorization in the field of linguistics. Meanwhile, researchers in the field of linguistics should shift their focus from the classical theory and the prototype theory to the exemplar-based theory appropriately.