Mediators and Moderators of Perceptual Learning

这一建议解决了感知环境中培训的特殊性和普遍性的基本问题。 学习（PL）。PL广泛地包括一套机制，通过这些机制， 环境引起感知处理的变化。PL潜在的更广泛的影响是巨大的。 在这一领域的仔细研究可以大大提高我们对感知系统的基本理解， 这些系统的可塑性。此外，以PL基础科学为基础的翻译方法 正变得越来越突出。这包括许多新兴的翻译方法， 知觉缺陷和认知训练的康复，这被认为是共享皮层可塑性 机制等然而，虽然现有的研究提供了证据，PL方法可以提高知觉， 技能，我们制定有效干预措施的能力受到缺乏对行为的理解的限制。 与不同PL方法相关的结果。在这里，我们建议，要了解和最大限度地利用 PL，有必要知道不同任务和不同个体的训练如何产生不同的 成果。成功翻译PL的一个主要障碍是该领域迄今为止一直受到强烈的推动 通过小N研究证明的“新颖”和“挑衅性”的发现，很少有项目深入挖掘， 实现稳健和可靠的结果。反过来，毫不奇怪，PL领域，像许多其他心理学领域一样， 面临着许多复制方面的挑战。此外，也许是因为在直接的脚步 与上述趋势相反，不同的研究小组使用相同的方法是令人惊讶的罕见的。 训练任务或结果测试。这是有问题的，因为研究表明，任务的微小变化- 不同的程序会导致学习结果的巨大差异。在这里，我们克服这些限制， 在一个大的主题中，使用共同的结果测量来比较大量不同的训练任务 人口这些任务中的每一项都涉及到特定研究提出的不同的学习“关键特征” 组然而，这些任务从未被直接比较或对比。建议的结果 本文的研究对于理解PL的基本含义以及如何翻译PL以帮助 有视觉需求的人。我们将通过培训大量的参与者样本来获得稳健可靠的结果 对PL任务，并通过一套共同的措施评估结果。我们还将收集广泛的评估 个体差异，这将提供一个独特的数据集，可以解决文献中的争议， 导致新的理解。我们提出的分析方法测试了该领域的中心关键假设， 探讨了不同的培训方法在多大程度上导致了系统性的不同学习概况， 并研究这些如何根据正在接受培训的个人而有所不同。进一步通过释放我们的训练 和测试工具以及收集的数据，我们将使其他团体能够模拟结果，复制我们的 研究，并作出明确的修改训练任务与已知的结果，以指导未来的研究。

This proposal addresses fundamental issue of specificity and generality of training in the context of Perceptual Learning (PL). PL broadly encompasses the set of mechanisms through which experience with the environment gives rise to changes in perceptual processing. The potential broader impacts of PL are immense. Careful research in this domain can greatly enhance our basic understanding of the perceptual systems and the plasticity of these systems. Furthermore, translational approaches underpinned by the basic science of PL are becoming increasingly prominent. This includes a host of emerging translational approaches for the rehabilitation of both perceptual deficits and for cognitive training, which are believed to share cortical plasticity mechanisms. However, while existing research provides evidence that PL approaches can improve perceptual skills, our ability to develop effective interventions is limited by a lack of understanding of the behavioral outcomes associated with different PL approaches. Here we suggest that to understand and maximally exploit PL, it is necessary to know how training with different tasks and in different individuals gives rise to different outcomes. One major obstacle to successful translation of PL is that the field to-date has been strongly driven by “novel” and “provocative” findings demonstrated via small N studies with very few projects digging deep to achieve robust and reliable results. In turn, not surprisingly, the field of PL, like many others in psychology, has suffered from numerous replication challenges. Furthermore, perhaps because following in direct footsteps runs counter to the tendencies noted above, it is surprisingly rare for different research groups to use identical training tasks or outcome tests. This is problematic given research showing that small changes in task- procedures can give rise to large differences in learning outcome. Here we overcome these limitations by comparing a large number of different training tasks using common outcome measures and in a large subject population. Each of these tasks involves a different “critical feature” for learning proposed by a given research group. However, these tasks have never been directly compared or contrasted. The outcome of the proposed research will be of tremendous value to both basic understanding of PL as well as how to translate PL to help those with visual needs. We will achieve robust and reliable results by training a large sample of participants on PL tasks and assess the outcomes via a common set of measures. We will also collect a broad assessment of individual differences, which will provide a unique dataset that can resolve controversies in the literature and lead to new understandings. Our proposed analytical approach tests central key hypothesis in the field, explores the extent to which different training approaches leads to systematically different profiles of learning, and examines how these can differ based upon the individuals being trained. Further by releasing our training and testing tools as well as the data collected, we will enable other groups to model results, replicate our studies, and make well specified modifications of training tasks with known outcomes to guide future research.