Investigating the Effectiveness of a Signal-based Approach inImproving Learners’ Decoding of Connected Speech The Case Study of Second-year Students, University of M’sila

This study examines the effectiveness of connected speech (CS) instruction in improving learners’ decoding of spoken English. It follows the trend of a balanced listening session that focuses on both top-down and bottom-up modes. It attempts to integrate CS instruction into the listening comprehension lessons following a diagnostic approach that uncovers CS features which may cause comprehension breakdown. An experimental group (N= 19) received listening lessons with an extended post-listening phase to address the CS features diagnosed as problematic during listening. A control group (N= 19) received similar lessons without specific focus on CS. The pre-test results revealed that CS aspects pose serious obstacles in decoding and segmenting speech for the participants, with a possible negative interference from top-down processing. The experimental group has shown a statistically significant improvement after the treatment. A post-treatment interview revealed positive reactions to the instruction, and a reported relative improvement in listening comprehension.