Norcia A M, Tyler C W, Hamer R D, Wesemann W
Smith-Kettlewell Eye Research Institute, San Francisco, CA 94115.
Vision Res. 1989;29(5):627-37. doi: 10.1016/0042-6989(89)90048-5.
Contrast response functions (CRFs) for the VEP were obtained with a Discrete Fourier Transform (DFT) technique employing swept contrast gratings. VEP CRFs in infants were found to have a form similar to those observed in adults, being linear functions of log contrast over a range of near-threshold contrasts. CRFs with low and high contrast lobes were present in infants, as they are in adults. Contrast thresholds were estimated by extrapolation of the CRF to zero microvolts. The effects of additive EEG noise and of the DFT data window on the shape of the measured CRF are considered. For large signals, the measured CRF is nearly independent of the additive noise, but at small signal values additive noise introduces a small bias towards larger amplitudes. The VEP signal-plus-noise distribution was modeled as a family of Rice distributions in order to evaluate the effects of bias on the estimates of threshold. The amount of bias depends inversely upon the slope of the CRF. The amount of bias introduced by a smoothing window also depends upon slope of the CRF as well as the sweep rate. The combined effects of additive noise and window bias were such that the total bias was nearly independent of CRF slope. Sweep VEP contrast thresholds were shown empirically to be unaffected by changes in the range of contrast swept.
采用扫频对比度光栅的离散傅里叶变换（DFT）技术获得了视觉诱发电位（VEP）的对比度响应函数（CRF）。研究发现，婴儿的VEP CRF形式与成人相似，在接近阈值的对比度范围内是对数对比度的线性函数。与成人一样，婴儿也存在具有低对比度和高对比度波瓣的CRF。通过将CRF外推至零微伏来估计对比度阈值。考虑了叠加脑电图噪声和DFT数据窗口对测量的CRF形状的影响。对于大信号，测量的CRF几乎与叠加噪声无关，但在小信号值时，叠加噪声会导致幅度向更大方向出现小的偏差。将VEP信号加噪声分布建模为一族莱斯分布，以评估偏差对阈值估计的影响。偏差量与CRF的斜率成反比。平滑窗口引入的偏差量还取决于CRF的斜率以及扫频速率。叠加噪声和窗口偏差的综合影响使得总偏差几乎与CRF斜率无关。根据经验表明，扫频VEP对比度阈值不受扫频对比度范围变化的影响。
