Unveiling the Relationship Between L2 Utterance Fluency and Perceived Fluency in Monologic and Dialogic Speaking

Area of Science:

• Applied Linguistics
• Second Language Acquisition
• Speech Communication

Background:

• Understanding L2 utterance fluency and its relationship with perceived fluency is crucial for language pedagogy.
• Previous research often treats speaking contexts monolithically, potentially overlooking task-specific nuances.
• The distinction between monologic and dialogic speech in L2 fluency perception requires further investigation.

Purpose of the Study:

• To investigate the relationship between objective L2 utterance fluency measures and subjective perceived fluency in both monologic and dialogic speaking tasks.
• To identify which specific fluency dimensions (speed, breakdown, repair) most significantly predict perceived fluency in different speaking contexts.
• To explore qualitative factors influencing L2 speaking fluency perceptions beyond quantitative measures.

Main Methods:

• A mixed-methods approach was employed, combining quantitative regression analysis with qualitative stimulated recalls.
• 136 Chinese university English learners and three experienced raters participated.
• Participants completed monologic and dialogic speaking tasks, with fluency measures analyzed against rater perceptions.

Main Results:

• In monologic tasks, most utterance fluency dimensions predicted perceived fluency, with breakdown fluency (especially silent pauses) being most impactful.
• In dialogic tasks, only breakdown fluency significantly predicted perceived fluency, surpassing speed and repair measures.
• Turn-taking temporal measures did not significantly influence perceived fluency.

Conclusions:

• The relationship between L2 utterance fluency and perceived fluency is context-dependent, varying between monologic and dialogic speech.
• Breakdown fluency, particularly silent pauses, plays a critical role in perceived fluency across both speaking modes.
• L2 speaking proficiency assessment tools should be adapted to differentiate between monologic and dialogic speaking contexts for accurate evaluation.