Saccadic plasticity: parametric adaptive control by retinal feedback.
Saccadic plasticity: parametric adaptive control by retinal feedback.
复制标题
扫视可塑性：通过视网膜反馈进行参数自适应控制。
DOI：
10.1037//0096-1523.7.2.356
复制
发表时间：
1981
期刊：
影响因子：
--
通讯作者：
W. B. Templeton
中科院分区：
文献类型：
--
作者：
Joel M. Miller;Tania Anstis;W. B. Templeton
扫视增益（扫视幅度与目标偏心率的比率）被实验性地改变，因为正常人类观察者向右进行再注视扫视，这导致目标水平位置的阶跃变化。眼运动通过漫射红外利姆布斯反射监测。我们发现，扫视增益的下降达到60%的完整性，而增加只有25%完成。适应能力的这种不对称性可能反映了扫视系统需要避免超过目标。对于单个目标，适应是快速的（时间常数= 6次扫视）;如果训练分布在六个不同的目标上，适应是相当慢的（时间常数= 57次扫视）。由给定目标的训练产生的增益变化不会强烈地转移到相同水平方向上的其他目标，并且可能根本不会转移到相反方向上的目标。对一个目标的扫视的增益可以减小，并且同时对在相同方向上的不同距离处的另一目标的扫视的增益增加。这些结果表明，每一个元件的感觉运动结构的基础扫视可塑性与特定的视网膜或空间感觉位点，可以改变其运动反应，而不会影响相邻元素的反应。这与分布式训练减缓适应的发现一致。
Saccadic gain (the ratio of saccadic amplitude to target eccentricity) was experimentally altered as normal human observers made refixation saccades to the right, which caused step changes in the horizontal position of the target. Eye movements were monitored by diffuse infrared limbus reflection. We found that decreases in saccadic gain reached 60% of completeness, whereas increases were only 25% complete. This asymmetry in adaptive capacity may reflect the saccadic system's need to avoid overshooting a target. With a single target, adaptation is rapid (time constant = 6 saccades); if training is distributed over six different targets, adaptation is considerably slower (time constant = 57 saccades). Gain changes that result from training with a given target do not transfer strongly to other targets in the same horizontal direction and may not transfer at all to targets in the opposite direction. The gain of saccades to one target may be decreased, and simultaneously the gain of saccades to another target at a different distance in the same direction is increased. These results suggest that each element of a sensory-motor structure underlying saccadic plasticity is associated with a particular retinal or spatial sensory locus and can alter its motor response without much affecting the response of neighboring elements. This is consistent with the finding that distributed training slows adaptation.