Hall J W, Grose J H, Hatch D R
Division of Otolaryngology/Head and Neck Surgery, University of North Carolina Medical School, Chapel Hill 27599-7070, USA.
J Acoust Soc Am. 1996 Oct;100(4 Pt 1):2365-72. doi: 10.1121/1.417946.
Thresholds for a 400-ms 1000-Hz pure-tone signal were obtained as a function of masking noise bandwidth for unmodulated and square wave modulated masking noise. Rates of modulation were 10 and 40 Hz. Noise bandwidths were 128 Hz, 387 Hz, 921 Hz, and 1505 Hz. The masking noise was either continuous or gated on and off with the signal. In general, signal thresholds were relatively constant as a function of noise bandwidth in unmodulated noise, and improved as a function of increasing noise bandwidth in modulated noise. Noise gating had little or no effect on signal threshold in unmodulated noise. At the 10-Hz modulation rate, signal thresholds were somewhat higher in gated than in continuous noise at relatively narrow noise bandwidths, but thresholds were similar in gated and continuous noise for relatively wide noise bandwidths. At bandwidths of 387, 921, and 1505 Hz, comodulation masking release (CMR) was calculated as the unmodulated noise threshold minus the modulated noise threshold, corrected by the difference between the unmodulated noise threshold and the modulated noise threshold at the 128-Hz bandwidths. At wide masker bandwidth, CMRs were higher for gated noise than for continuous noise. This was due almost entirely to the threshold gating effect found in the 128-Hz bandwidth condition. These results suggested that there was a within-channel effect for gated noise thresholds to be higher than continuous noise thresholds, but essentially no across-channel effect of gating. At the 40-Hz modulation rate, signal thresholds were similar for gated and continuous noise at all noise bandwidths. There was a very small but significant' effect for the gated noise threshold to be lower than the continuous noise threshold at the widest noise bandwidth. It was speculated that this effect may be related to a decrease in sensitivity to modulation with continuous stimulation. In general, effects of gating appear to be small or absent for across-channel masking release in broadband modulated masking noise.
针对未调制和方波调制的掩蔽噪声,获取了400毫秒1000赫兹纯音信号的阈值,该阈值是掩蔽噪声带宽的函数。调制速率为10赫兹和40赫兹。噪声带宽分别为128赫兹、387赫兹、921赫兹和1505赫兹。掩蔽噪声要么是连续的,要么与信号同步开关。一般来说,在未调制噪声中,信号阈值相对恒定,是噪声带宽的函数;而在调制噪声中,信号阈值随着噪声带宽的增加而改善。噪声门控对未调制噪声中的信号阈值几乎没有影响。在10赫兹调制速率下,在相对窄的噪声带宽时,门控噪声中的信号阈值略高于连续噪声中的信号阈值,但在相对宽的噪声带宽时,门控噪声和连续噪声中的阈值相似。在387赫兹、921赫兹和1505赫兹带宽下,共调制掩蔽释放(CMR)计算为未调制噪声阈值减去调制噪声阈值,并通过128赫兹带宽下未调制噪声阈值与调制噪声阈值之间的差异进行校正。在宽掩蔽带宽下,门控噪声的CMR高于连续噪声。这几乎完全归因于在128赫兹带宽条件下发现的阈值门控效应。这些结果表明,存在通道内效应,即门控噪声阈值高于连续噪声阈值,但门控基本上没有跨通道效应。在40赫兹调制速率下,在所有噪声带宽下,门控噪声和连续噪声的信号阈值相似。在最宽的噪声带宽下,门控噪声阈值略低于连续噪声阈值,存在非常小但显著的效应。据推测,这种效应可能与连续刺激下调制敏感性的降低有关。一般来说,对于宽带调制掩蔽噪声中的跨通道掩蔽释放,门控效应似乎很小或不存在。