首页> 外文会议>International Congress on Sound and Vibration >PERCEPTUAL LEARNING OF FUNDAMENTAL FREQUENCY DISCRIMINATION: EFFECTS OF HARMONIC NUMBER
【24h】

PERCEPTUAL LEARNING OF FUNDAMENTAL FREQUENCY DISCRIMINATION: EFFECTS OF HARMONIC NUMBER

机译:感知基本频率歧视的感知学习:谐波数的影响

获取原文

摘要

Perceptual learning was used to examine mechanisms of pitch perception. Thresholds (FODLs) were measured for discrimination of the fundamental frequency (F0) of a complex tone with a nominal F0 of 100 Hz and cosine-phase harmonics, bandpass filtered into one of three frequency regions, LOW (harmonics 1-5), MID (harmonics 11-15), and HIGH (harmonics 28-32). Tone LOW contained resolved harmonics, conveying both place and temporal fine structure (TFS) information. Tone HIGH contained only unresolved harmonics and conveyed only temporal envelope information. Tone MID probably conveyed TFS information but not information about individual harmonics. All tones were presented in threshold equalizing noise. Fifteen subjects were tested over a period of 10 days, 2 for measurement of pre-training thresholds, 6 for training, and 2 for measurement of post-training thresholds. In the pre-training and post-training sessions, all subjects were tested using all three tones. Subjects were divided into three groups of five subjects for the training sessions: each group was trained with one stimulus type. The group trained with LOW stimuli showed a large training effect, with transfer to the MID stimuli, but no transfer to the HIGH stimuli. The group trained with MID stimuli showed a large training effect, with transfer to the LOW stimuli and no transfer to the HIGH stimuli. The group trained with HIGH stimuli showed no training effect for any stimuli. The results suggest that similar pitch mechanisms were used for F0 discrimination of the LOW and MID stimuli, and that a different mechanism was used for the HIGH stimuli. It is proposed that the LOW and MID stimuli were both discriminated on the basis of TFS information, in the former case TFS information about individual resolved harmonics, and in the latter case TFS information about the periodicity of the waveform evoked by interfering harmonics.
机译:感知学习被用来检查音调感知的机制。测量阈值(FODL)以辨别复杂音调的基本频率(F0),标称F0为100Hz和余弦相次谐波,带通滤波到三个频率区域中的一个,低(谐波1-5)中,中间(谐波11-15)和高(谐波28-32)。音调低包含已解决的谐波,传达地点和时间细结构(TFS)信息。音调高包含未解决的谐波并仅传达时间信封信息。音调中期可能传达了TFS信息,而不是有关个人谐波的信息。所有音调都以阈值均衡噪声呈现。在10天内测试了十五个受试者,用于测量预训练阈值,6用于训练,2个用于测量训练后阈值。在预培训和培训后会议中,所有受试者都使用所有三个音调进行测试。受试者分为三组培训课程的五个科目:每组都有一个刺激型训练。用低刺激训练的小组显示出大的训练效果,转移到中期刺激,但没有转移到高刺激。培训中刺激的小组显示出大的培训效果,转移到低刺激,没有转移到高刺激。高刺激培训的小组对任何刺激都没有培训效果。结果表明,类似的俯仰机制用于低和中间刺激的F0辨别,并且使用不同的机制用于高刺激。建议在前案例中,在TFS信息的基础上,在前案例中,在TFS信息的基础上,关于个别解决的谐波的信息,以及后一种情况,关于干扰谐波引起的波形周期性的TFS信息。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号