Cohesion in oral discourse of Mandarin-speaking adults with traumatic brain injury : report of pilot data on story telling

Abstract

Background The literature reporting the use of cohesive markers in discourse production among English-speakers with traumatic brain injury (TBI) have focused on the amount and types of cohesive devices produced. Apart from reduced amount of cohesive tie production (e.g., Carlomagno et al., 2011; Davis & Coelho, 2004), TBI narratives have been reported to contain a higher percentage of ellipsis and incomplete cohesive ties (e.g., Mentis & Prutting, 1987). According to a recent review by McDonald et al. (2013), the degree of cohesion in TBI discourse production can be affected by a number of factors, such as discourse task, severity of TBI, and cognitive integrity of TBI patients. Owing to typological features that are different from English, Chinese (as a pro-drop language in which sentential meaning is frequently recovered from context instead of overt linguistic forms) has unique characteristics of cohesive devices to achieve inter-sentential cohesion (Halliday & Webster, 2009). This paper aimed to explore how well existing measures of impaired discourse cohesion can be applied to Mandarin Chinese. Methods Eighteen subjects with a single closed-head TBI with a post onset time of at least four months were recruited. Language samples were collected and orthographically transcribed using the Chinese AphasiaBank protocol (see Kong et al., 2015) modified for TBI. Each sample was segmented into elementary discourse units (EDUs), which is the minimal semantic building blocks of a discourse (Mann & Thompson, 1988). Degree of cohesion on the story telling task was subsequently quantified on three aspects: (1) cohesive adequacy ratio, obtained based on the distribution of EDUs with cohesive ties that were “Complete,” “Incomplete,” or “Error/ Ambiguous”; (2) accuracy and error rate of cohesive marker usage, and (3) distribution of types of accurately used cohesive markers, including reference markers (i.e., identity of the thing or class of things being referred to in the preceding or following text; Liles et al., 1989), synonyms (or identical lexicons), conjunctions (such as those indicating additive, temporal, causal, or adversive relationships) and , ellipsis for topic-chain, etc. Results and Discussion Preliminary results based on six TBI subjects suggested that the average number of EDUs was 13. Percentage of “Complete,” “Incomplete,” and “Error/ Ambiguous” cohesive ties was 42.5%, 32.5%, and 25%, respectively. Ratio of accurate cohesive markers usage (per EDU) was found to be 0.40, which was close to the erroneous ratio of 0.43. In addition, the most commonly used cohesive marker was conjunction (51.61%), followed by ellipsis (29.03%), and reference (e.g., pronouns, demonstratives, and classifiers) and lexical (e.g., synonyms or antonyms) makers (19.35%). Further analyses involving additional narrative tasks (including sequential picture descriptions, procedurals, and monologues) as well as comparison between TBI and controls are in progress. Details in the modifications necessary to accurately and adequately highlight the cohesive markings in Chinese will also be discussed. For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.