SPEECH MODIFICATION TO NON-NATIVE SPEAKERS AND CONTENT DILUTION: IMPLICATIONS FOR ENGLISH AS A MEDIUM OF INSTRUCTION (EMI)
Al Thowaini, Assma Mohammad
Long, Michael H
MetadataShow full item record
With the rapid growth of language education programs, such as Content-and-Language-Integrated Learning (CLIL) and English as a medium of instruction (EMI), research on input modification shifted perspectives. The current study investigates L2 input modification by comparing the speech of non-native speakers to that of native speakers towards low-proficiency learners of English using quantitative methods. Furthermore, the study explores the effects of these modifications on learners’ content comprehension and the possible content dilution (e.g., loss of essential information) triggered by linguistic simplification. In this experiment, two types of participants were recruited: speakers and listeners. Twenty native and advanced non-native speakers of English participated (ten of each). The speaker participants were divided into two subgroups: those with language teaching experience and those without. For the listeners, three groups were recruited: 20 native speaker controls, 20 high-proficiency, and 20 low-proficiency English learners (listeners). Each speaker narrated stories to three assigned listeners (one from each condition) in one-on-one sessions. Each session included an introduction, two warm-up stories, and three main stories. Speech was audio-recorded to examine the types of modification employed with high- and low-proficiency listeners, as opposed to native listener controls, and the effects of those modifications on story content and listener comprehension. After each story, the listeners took a content comprehension assessment. The transcripts were coded for lexical complexity (diversity and sophistication), syntactic complexity, and content dilution. The results showed a significant difference between native and non-native speakers in their speech to the three listener conditions in terms of lexical sophistication and syntactic complexity, as well as a significant difference between speakers with language teaching experience and speakers without in terms of lexical diversity. Furthermore, all speaker conditions exhibited significant linguistic modification (lexical diversity, lexical sophistication, and syntactic complexity) in their speech towards low- and high-proficiency listeners compared to their speech towards the native controls. In addition, only native speakers showed significant content dilution (measured by the count of mentioned information bits) in their speech towards high- and low-proficiency listeners. Finally, the high- and low-proficiency listeners’ scores on the content comprehension assessment were significantly lower than the scores of the native controls.