Polat Uri
Goldschleger Eye Research Institute, Tel Aviv University, Sheba Medical Center, Ramat Gan, 52621 Tel Hashomer, Israel.
Vision Res. 2009 Oct;49(21):2566-73. doi: 10.1016/j.visres.2009.06.005. Epub 2009 Jun 9.
Task-specific improvement in performance after training is well established. The finding that learning is stimulus-specific and does not transfer well between different stimuli, between stimulus locations in the visual field, or between the two eyes has been used to support the notion that neurons or assemblies of neurons are modified at the earliest stage of cortical processing. However, a debate regarding the proposed mechanism underlying perceptual learning is an ongoing issue. Nevertheless, generalization of a trained task to other functions is an important key, for both understanding the neural mechanisms and the practical value of the training. This manuscript describes a structured perceptual learning method that previously used (amblyopia, myopia) and a novel technique and results that were applied for presbyopia. In general, subjects were trained for contrast detection of Gabor targets under lateral masking conditions. Training improved contrast sensitivity and diminished the lateral suppression when it existed (amblyopia). The improvement was transferred to unrelated functions such as visual acuity. The new results of presbyopia show substantial improvement of the spatial and temporal contrast sensitivity, leading to improved processing speed of target detection as well as reaction time. Consequently, the subjects, who were able to eliminate the need for reading glasses, benefited. Thus, here we show that the transfer of functions indicates that the specificity of improvement in the trained task can be generalized by repetitive practice of target detection, covering a sufficient range of spatial frequencies and orientations, leading to an improvement in unrelated visual functions. Thus, perceptual learning can be a practical method to improve visual functions in people with impaired or blurred vision.
训练后特定任务的表现改善已得到充分证实。学习具有刺激特异性，在不同刺激之间、视野中的刺激位置之间或双眼之间不能很好地迁移，这一发现被用来支持神经元或神经元集合在皮层处理的最早阶段被改变的观点。然而，关于感知学习潜在机制的争论仍是一个持续存在的问题。尽管如此，将训练任务推广到其他功能对于理解神经机制和训练的实际价值而言都是一个重要关键。本手稿描述了一种先前用于（弱视、近视）的结构化感知学习方法以及一种应用于老花眼的新技术和结果。一般来说，受试者在侧向掩蔽条件下接受Gabor目标的对比度检测训练。训练提高了对比度敏感度，并在存在侧向抑制时（弱视）减弱了侧向抑制。这种改善转移到了诸如视力等不相关的功能上。老花眼的新结果显示空间和时间对比度敏感度有显著提高，从而提高了目标检测的处理速度以及反应时间。因此，能够不再需要老花镜的受试者从中受益。所以，我们在此表明，功能的转移表明通过重复进行目标检测练习，覆盖足够范围的空间频率和方向，可以使训练任务中改善的特异性得以推广，从而改善不相关的视觉功能。因此，感知学习可以成为改善视力受损或模糊人群视觉功能的一种实用方法。
