Audiology & Deafness Research Group, School of Psychological Sciences, The University of Manchester, Ellen Wilkinson Building, Manchester, M13 9PL, UK,
J Assoc Res Otolaryngol. 2013 Dec;14(6):879-90. doi: 10.1007/s10162-013-0417-9. Epub 2013 Sep 19.
An experiment was conducted with eight cochlear implant subjects to investigate the feasibility of using electrically evoked compound action potential (ECAP) measures other than ECAP thresholds to predict the way that behavioral thresholds change with rate of stimulation, and hence, whether they can be used without combination with behavioral measures to determine program stimulus levels for cochlear implants. Loudness models indicate that two peripheral neural response characteristics contribute to the slope of the threshold versus rate function: the way that neural activity to each stimulus pulse decreases as rate increases and the slope of the neural response versus stimulus current function. ECAP measures related to these two characteristics were measured: the way that ECAP amplitude decreases with stimulus rate and the ECAP amplitude growth function, respectively. A loudness model (incorporating temporal integration and the two neural response characteristics) and regression analyses were used to evaluate whether the ECAP measures could predict the average slope of the behavioral threshold versus current function and whether individual variation in the measures could predict individual variation in the slope of the threshold function. The average change of behavioral threshold with increasing rate was well predicted by the model when using the average ECAP data. However, the individual variations in the slope of the thresholds versus rate functions were not well predicted by individual variations in ECAP data. It was concluded that these ECAP measures are not useful for fully objective programming, possibly because they do not accurately reflect the neural response characteristics assumed by the model, or are measured at current levels much higher than threshold currents.
一项针对 8 名人工耳蜗植入受试者的实验旨在研究使用电诱发复合动作电位 (ECAP) 测量值（除 ECAP 阈值外）来预测行为阈值随刺激率变化的方式的可行性，以及是否可以在不与行为测量值相结合的情况下使用这些测量值来确定人工耳蜗植入的刺激程序水平。响度模型表明，两个外围神经反应特性对阈值与速率函数的斜率有贡献：随着刺激率的增加，每个刺激脉冲的神经活动减少的方式以及神经反应与刺激电流函数的斜率。分别测量了与这两个特性相关的 ECAP 测量值：ECAP 幅度随刺激率降低的方式和 ECAP 幅度增长函数。使用响度模型（包含时间整合和两个神经反应特性）和回归分析来评估 ECAP 测量值是否可以预测行为阈值与电流函数的平均斜率，以及测量值的个体变化是否可以预测阈值函数斜率的个体变化。当使用平均 ECAP 数据时，模型很好地预测了行为阈值随刺激率增加的平均变化。然而，阈值与刺激率函数斜率的个体变化并不能很好地通过 ECAP 数据的个体变化来预测。得出的结论是，这些 ECAP 测量值不适用于完全客观的编程，可能是因为它们不能准确反映模型所假设的神经反应特性，或者是在比阈值电流高得多的电流水平下进行测量。
