Effects of auditory context on nonnative recognition of reduced speech: Does meaning explain it all?

Previous studies have shown that acoustic cues in the sentential context improve native (L1) listeners’ recognition of reduced speech (Janse & Ernestus, 2011). That is, when hearing a reduced target in casual speech (e.g., too into /twntə/), L1 listeners would be more accurate in recognizing it in its surrounding speech (e.g., she’s too into computers) than if they merely see the text of its context (e.g., she’s ____ computers), even though semantic/syntactic information is equivalent in both conditions. It is unclear, however, whether nonnative (L2) listeners make use of the acoustic cues in the context to facilitate reduced speech recognition to the same extent as L1 listeners do. This study investigates this question by comparing L1 and L2 listeners of English in their recognition of reduced words that were aurally presented in three conditions—Isolation, Textual (with surrounding text given), and Auditory (with both text and auditory forms of context)—in a repetition task. Results showed that L2 listeners were significantly more accurate in the Auditory than Textual condition, and such difference was not distinguishable from L1 listeners. These findings lend support to the hypothesis that the facilitative effects of context acoustic cues are language general.