Lee B B, Pokorny J, Smith V C, Kremers J
Department of Neurobiology, Max Planck Institute for Biophysical Chemistry, Göttingen, Germany.
Vision Res. 1994 Dec;34(23):3081-96. doi: 10.1016/0042-6989(94)90074-4.
The goal of the study was to compare pulse responses with sinusoidal temporal responsivity. The response of macaque ganglion cells was measured to brief luminance and chromatic pulses and to luminance or chromatic sinusoidal modulation. To make both positive and negative lobes of the pulse response visible, responses to pulses of opposite polarity were combined to yield a linearized pulse response. Tests of superposition were used to evaluate the linearized pulse response to different combinations of pulse duration and Weber contrast. A prediction of the pulse response was derived using sinusoidal responsivity functions and Fourier synthesis. For ganglion cells of the parvocellular (PC) pathway, shape and absolute amplitude of linearized pulse responses corresponded well to the predicted responses over a range of pulse durations at 0.5 and 1.0 Weber contrast for both luminance and chromatic modulation. For ganglion cells of the magnocellular (MC) pathway, shape and amplitude of the linearized pulse responses and the predicted responses corresponded when the contrast-duration product was low. This correspondence held for luminance modulation over a thousand-fold range of retinal illuminance. For contrast-duration combinations that produced a more vigorous response, over 100 imp/sec, the linearized pulse responses of MC-pathway cells became larger and time-advanced relative to the linear prediction until saturation became apparent. Incorporation of high Michelson contrast responses in the Fourier synthesis captured the timing but not the amplitude of the linearized pulse response. The data suggest that a mechanism similar to a contrast gain control acts upon MC- but not PC-pathway-cells. The data confirm that use of linear modelling to describe temporal behaviour of retinal ganglion cells is appropriate for small signals.
该研究的目的是比较脉冲响应与正弦时间响应性。测量了猕猴神经节细胞对短暂亮度和色度脉冲以及亮度或色度正弦调制的响应。为了使脉冲响应的正负波瓣都可见,将对相反极性脉冲的响应进行合并,以产生线性化的脉冲响应。使用叠加测试来评估线性化脉冲响应对不同脉冲持续时间和韦伯对比度组合的情况。利用正弦响应函数和傅里叶合成得出脉冲响应的预测值。对于小细胞(PC)通路的神经节细胞,在亮度和色度调制下,线性化脉冲响应的形状和绝对幅度在0.5和1.0韦伯对比度的一系列脉冲持续时间范围内与预测响应非常吻合。对于大细胞(MC)通路的神经节细胞,当对比度-持续时间乘积较低时,线性化脉冲响应的形状和幅度与预测响应相对应。这种对应关系在视网膜照度的千倍范围内的亮度调制中都成立。对于产生更强烈响应(超过100次脉冲/秒)的对比度-持续时间组合,MC通路细胞的线性化脉冲响应相对于线性预测变得更大且时间提前,直到饱和变得明显。在傅里叶合成中纳入高迈克尔逊对比度响应能够捕捉到线性化脉冲响应的时间,但不能捕捉到其幅度。数据表明,一种类似于对比度增益控制的机制作用于MC通路细胞而非PC通路细胞。数据证实,使用线性模型来描述视网膜神经节细胞的时间行为适用于小信号。