Learner output, hypothesis testing, and internalizing linguistic knowledge

Ali Shehadeh
DOI: https://doi.org/10.1016/s0346-251x(03)00018-6
IF: 4.518
2003-06-01
System
Abstract:This study investigates how output can be a process by which second language (L2) learners test out hypotheses about the L2 and the extent to which learner hypothesis testing attempts that result in non-target like (NTL) output are challenged by interlocutors. A picture-description task was used to collect data from 16 participants, eight native speakers (NS) and eight nonnative speakers (NNS) of English, forming eight NS–NNS dyads. All interactions were audio-taped. The data were analyzed and examined specifically for hypothesis testing episodes (HTEs) by NNSs. The results showed that NNSs tested out one hypothesis about the target language (TL) every 1.8 min. The results also revealed that those HTEs that resulted in NTL output and constituted over a third of all HTEs found went completely unchallenged by interlocutors. These results were interpreted to mean that failing to provide corrective feedback or negative evidence to learner output that exhibits NTL utterances or rules may constitute a signal for the confirmation of these utterances or rules, albeit non-target like, from the perspective of the internal processing systems of the learner, which, in turn, constitutes a step toward internalizing linguistic knowledge.
education & educational research,linguistics
What problem does this paper attempt to address?