INM, Inserm, Univ Montpellier, 34295 Montpellier, France.
INM, Inserm, Univ Montpellier, 34295 Montpellier, France
J Neurosci. 2018 Jun 20;38(25):5727-5738. doi: 10.1523/JNEUROSCI.3103-17.2018. Epub 2018 May 23.
Auditory nerve fibers (ANFs) encode pure tones through two modes of coding, spike time and spike rate, depending on the tone frequency. In response to a low-frequency tone, ANF firing is phase locked to the sinusoidal waveform. Because time coding vanishes with an increase in the tone frequency, high-frequency tone coding relies on the spike rate of the ANFs. Adding a continuous broadband noise to a tone compresses the rate intensity function of ANFs and shifts its dynamic range toward higher intensities. Therefore, the ANFs with high-threshold/low-spontaneous rate (SR) are thought to contribute to behavioral tone detection in noise. However, this theory relies on the discharge rate of the ANFs. The direct comparison with the masking threshold through spike timing, irrespective of the spontaneous rate, has not so far been investigated. Taking advantage of a unique proxy to quantify the spike synchrony (i.e., the shuffle autocorrelogram), we show in female gerbils that high-SR ANFs are more adapted to encode low-frequency thresholds through temporal code, giving them a strong robustness in noise. By comparing behavioral thresholds measured using prepulse inhibition of the acoustical startle reflex with population thresholds calculated from ANFs pooled per octave band, we show that threshold-based spike timing provides a better estimate of behavioral thresholds in the low-frequency range, whereas the high-frequency behavioral thresholds rely on the spiking rate, particularly in noise. This emphasizes the complementarity of temporal and rate modes to code tone-in-noise thresholds over a large range of frequencies. There is a general agreement that high-threshold/low-spontaneous rate (SR) auditory nerve fibers (ANFs) are of prime importance for tone detection in noise. However, this theory is based on the discharge rate of the fibers. Comparing the behavioral thresholds and single ANF thresholds shows that this is only true in the high-frequency range of tone stimulations. In the low-frequency range of tones (up to 2.7 kHz in the gerbil), the most sensitive ANFs (high-SR fibers) carry neural information through a spike-timing mode, even for noise in which tones do not induce a noticeable increment in the spike rate. This emphasizes the interplay between spike-time and spike-rate modes in the auditory nerve to encode tone-in-noise threshold over a large range of tone frequencies.
听神经纤维（ANF）通过两种编码模式（即尖峰时间和尖峰率）对纯音进行编码，这取决于音调频率。对于低频音调，ANF 的放电与正弦波的相位锁定。由于随着音调频率的增加，时间编码消失，高频音调编码依赖于 ANF 的尖峰率。在音调上添加连续宽带噪声会压缩 ANF 的率强度函数，并将其动态范围向更高强度转移。因此，具有高阈值/低自发率（SR）的 ANF 被认为有助于在噪声中进行行为音调检测。然而，这种理论依赖于 ANF 的放电率。直接通过尖峰时间而不是自发率与掩蔽阈值进行比较，到目前为止尚未进行研究。利用一种独特的代理来量化尖峰同步性（即，随机自相关图），我们在雌性沙鼠中表明，高 SR ANF 更适应通过时间编码对低频阈值进行编码，从而在噪声中具有很强的鲁棒性。通过将使用声激发反射的前脉冲抑制测量的行为阈值与从每个倍频带中汇集的 ANF 计算得出的群体阈值进行比较，我们表明基于阈值的尖峰时间为低频范围内的行为阈值提供了更好的估计，而高频行为阈值则依赖于尖峰率，尤其是在噪声中。这强调了在较大频率范围内通过时间和率模式来编码音调噪声阈值的互补性。人们普遍认为，高阈值/低自发率（SR）听觉神经纤维（ANF）对噪声中的音调检测至关重要。然而，这种理论是基于纤维的放电率。比较行为阈值和单个 ANF 阈值表明，这种情况仅在音调刺激的高频范围内是正确的。在音调的低频范围内（在沙鼠中高达 2.7 kHz），最敏感的 ANF（高 SR 纤维）通过尖峰时间模式传递神经信息，即使在噪声中，音调也不会引起尖峰率的明显增加。这强调了听觉神经中尖峰时间和尖峰率模式在较大音调频率范围内编码音调噪声阈值的相互作用。
