Department of Psychology, Queen's University, Humphrey Hall, 62 Arch Street, Kingston, Ontario K7L 3N6, Canada.
J Acoust Soc Am. 2010 Feb;127(2):1059-68. doi: 10.1121/1.3278606.
Previous auditory perturbation studies have demonstrated that talkers spontaneously compensate for real-time formant-shifts by altering formant production in a manner opposite to the perturbation. Here, two experiments were conducted to examine the effect of amplitude of perturbation on the compensatory behavior for the vowel /epsilon/. In the first experiment, 20 male talkers received three step-changes in acoustic feedback: F1 was increased by 50, 100, and 200 Hz, while F2 was simultaneously decreased by 75, 125, and 250 Hz. In the second experiment, 21 male talkers received acoustic feedback in which the shifts in F1 and F2 were incremented by +4 and -5 Hz on each utterance to a maximum of +350 and -450 Hz, respectively. In both experiments, talkers altered production of F1 and F2 in a manner opposite to that of the formant-shift perturbation. Compensation was approximately 25%-30% of the perturbation magnitude for shifts in F1 and F2 up to 200 and 250 Hz, respectively. As larger shifts were applied, compensation reached a plateau and then decreased. The similarity of results across experiments suggests that the compensatory response is dependent on the perturbation magnitude but not on the rate at which the perturbation is introduced.
先前的听觉干扰研究表明，说话者会通过与干扰相反的方式改变共振峰产生来自发补偿实时共振峰变化。在这里，进行了两项实验来检验干扰幅度对元音 /epsilon/ 的补偿行为的影响。在第一个实验中，20 名男性说话者接受了三种声学反馈的阶跃变化：F1 增加了 50、100 和 200 Hz，而 F2 同时减少了 75、125 和 250 Hz。在第二个实验中，21 名男性说话者接受了声学反馈，其中 F1 和 F2 的变化在每个发音时分别增加了 +4 和 -5 Hz，最大可达 +350 和 -450 Hz。在这两个实验中，说话者以与共振峰变化相反的方式改变了 F1 和 F2 的产生。对于 F1 和 F2 的变化，补偿约为干扰幅度的 25%-30%，最高可达 200 和 250 Hz。随着较大的变化应用，补偿达到一个平台，然后下降。实验结果的相似性表明，补偿反应取决于干扰幅度，而不取决于干扰引入的速度。
