Levi D M, Polat U, Hu Y S
College of Optometry, University of Houston, Texas 77204-6052, USA.
Invest Ophthalmol Vis Sci. 1997 Jul;38(8):1493-510.
To determine the nature and limits of visual improvement through repetitive practice in human adults with naturally occurring amblyopia.
A key measure the authors used was a psychophysical estimate of Vernier acuity; persons with amblyopia have marked deficits in Vernier acuity that are highly correlated with their loss of Snellen acuity. The experiment consisted of three phases: pretraining measurements of Vernier acuity and a second task (either line-detection thresholds or Snellen acuity) in each eye with the lines at two orientations; a training phase in which observers repetitively trained on the Vernier task at a specific line orientation until each had completed 4000 to 5000 trials; and posttraining measurements (identical to those in the first phase). Two groups of amblyopic observers were tested: novice observers (n = 6), who had no experience in making psychophysical judgments with their amblyopic eyes, and experienced observers (n = 5), who had previous experience in making Vernier judgments with their amblyopic eyes (with the lines at a different orientation) using the signal-detection methodology.
The authors found that strong and significant improvement in Vernier acuity occurs in the trained orientation in all observers. Learning was generally strongest at the trained orientation but may partially have been transferred to other orientations (n = 4). Significant learning was transferred partially to the other eye (at the trained orientation) in two observers with anisometropic amblyopia. Improvement in Vernier acuity did not transfer to an untrained detection task. In two observers, the improvement in Vernier acuity was accompanied by a commensurate improvement in Snellen acuity.
Some adults with amblyopia retain a significant degree of neural plasticity. Although several observers (primarily novices) showed evidence of generalized learning, several amblyopic patients showed evidence for improvement that was orientation and task specific. In this latter group of observers, the improvement appeared to reflect alterations that were, at least in part, in early neural processes that were orientation specific and were localized beyond the site of convergence of the two eyes.
通过对患有自然性弱视的成年人进行重复训练，确定视力改善的性质和限度。
作者使用的一项关键指标是对游标视力的心理物理学评估；弱视患者在游标视力方面存在明显缺陷，这与他们的斯内伦视力丧失高度相关。实验包括三个阶段：对每只眼睛在两种方向的线条下进行游标视力和第二项任务（线条检测阈值或斯内伦视力）的预训练测量；训练阶段，观察者在特定线条方向上对游标任务进行重复训练，直到每人完成4000至5000次试验；以及训练后测量（与第一阶段相同）。对两组弱视观察者进行了测试：新手观察者（n = 6），他们没有用弱视眼进行心理物理学判断的经验；经验丰富的观察者（n = 5），他们之前使用信号检测方法用弱视眼（线条为不同方向）进行过游标判断。
作者发现，所有观察者在训练方向上的游标视力都有显著且明显的改善。学习通常在训练方向上最强，但可能部分转移到了其他方向（n = 4）。在两名屈光参差性弱视观察者中，显著的学习部分转移到了另一只眼睛（在训练方向上）。游标视力的改善没有转移到未训练的检测任务中。在两名观察者中，游标视力的改善伴随着斯内伦视力的相应改善。
一些患有弱视的成年人仍保留着显著程度的神经可塑性。虽然有几名观察者（主要是新手）表现出了广义学习的证据，但有几名弱视患者表现出了针对方向和任务的改善证据。在后者这组观察者中，改善似乎反映了至少部分在早期神经过程中的改变，这些改变是方向特异性的，并且局限于双眼汇聚点之外的部位。
