Training to Perceive English Lexical Stress in Rising Intonation: Generalizability and Retainability

Abstract

This chapter reports the results from the generalization test and the retention test. The former assessed whether the training effects observed in the post-test extended to stimuli not included in the training; the latter examined whether these effects were still maintained three months after the completion of the training. It was found that the trainees’ improved ability to perceive stress in rising intonation did generalize to novel stimuli, but only partially. However, such improvement was robustly retained, as there was no significant difference between the retention test and post-test in identification accuracy of stress in rising intonation. When compared with the pre-test, the trainees did not show decreased performance for the stimuli with trochaic stress in falling intonation in the generalization and retention tests. Correlation analyses revealed that in these two tests, the trainees no longer exploited relative vowel durations to identify stress patterns. Taken together, the findings lead to the inference that the only cue to stress in rising intonation that was robustly learned by the trainees was the timing of pitch elbow. The bias toward using relative vowel durations seems to be temporary and stimulus-specific.