Development of the Taiwanese Mandarin Main Concept Analysis (TM-MCA): Normative and Preliminary Aphasic Data

• ¹ University of Central Florida, Department of Communication Sciences and Disorders, United States

Introduction
Compared to the English literature, there is a great paucity of studies focusing on quantitative analysis of aphasic oral discourse in Chinese, especially for native speakers of Taiwanese-Mandarin residing in Taiwan. Several studies that investigated deficits of aphasic discourse have been conducted in Cantonese Chinese, including the Cantonese Linguistic Communication Measure (CLCM, Kong & Law, 2004), Main Concept Analysis (MCA; Kong, 2009), and Cantonese version of the Quantitative Production Analysis (Law, 2001; Yiu & Worrall, 1996). Clinical or research application of these systems to the Taiwanese population is limited by the differences of phonology, semantics, and grammar between Taiwanese Mandarin and Cantonese. The current study, therefore, aimed to develop a culturally appropriate Taiwan Mandarin Main Concept Analysis (TM-MCA), by translating and adapting the MCA in Cantonese (Kong, 2009, 2011), and then to validate the tool on the basis of normal speakers and people with aphasia in Taiwan.

Method
Picture stimuli and main concepts for the TM-MCA: Eight certified Taiwanese SLPs participated and listed all main concepts upon presentation of the sequential pictures in Kong (2009). A main concept referred to a statement consisting of only one main verb with two or more pieces of essential information accurately related to the pictures and informative about the content. All main concepts mentioned by six out of eight SLPs were included in a preliminary main concept list. Further validation of the list involved recruiting eight normal speakers to orally describe the same picture sets. Only those scored as Accurate and Complete (AC) or Accurate but Incomplete (AI) by at least five speakers were put in the final list.
Normative and aphasic data collection and analysis: Thirty six native Taiwanese Mandarin speakers and ten individuals with aphasia orally perform the same picture description tasks. Six measures, including Number of Accurate and Complete (AC), Accurate but Incomplete (AI), Inaccurate (IN), and Absent (AB) concepts, overall main concept score (MC), as well as AC per minute, were used to analyze each language samples. A pair-sample t-test was conducted to evaluate the different performance between the aphasic and control groups. External validity of the TM-MCA was established by conducting Pearson product-moment correlations between the TM-MCA measures and the overall aphasia score in the Concise Chinese Aphasia Test (CCAT; Chung, Li, & Chang, 2003) and Linguistic Communication Measure adapted for Taiwanese Mandarin (TM-LCM).

Results and Discussion
The results suggested that both age and education affected the oral discourse performance in our normal subjects. Significant differences on the measures in TM-MCA were noted between the normal and aphasic groups. It was also found that overall aphasia severity, as reflected by the CCAT, affected the picture description performances. Finally, significant correlations between most TM-MCA measures and related TM-LCM indices were noted. In conclusion, the TM-MCA is culturally appropriate to the Taiwanese Mandarin population as a supplement to existing standardized aphasia batteries, which typically do not include comprehensive and objective evaluation of oral discourse.