Mediators and Moderators of Perceptual Learning

这一建议解决了在知觉背景下训练的特殊性和一般性的根本问题 学习(PL)。PL广泛地包含了一套机制，通过这些机制，可以获得 环境引起了知觉加工的变化。物流潜在的更广泛的影响是巨大的。 对这一领域的仔细研究可以极大地提高我们对知觉系统和 这些系统的可塑性。此外，以公共图书馆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，用于 知觉缺陷和认知训练的康复，据信这两种训练具有皮质可塑性 机械装置。然而，尽管现有的研究提供了证据表明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 与不同的PL方法相关的结果。在这里，我们建议理解并最大限度地利用 有必要知道不同任务和不同个体的训练是如何产生不同的 结果。成功翻译PL的一个主要障碍是，到目前为止，这一领域一直受到强烈的推动 通过很少的项目深入挖掘的小规模N研究，展示了“新颖”和“具有挑衅性”的发现 取得稳健可靠的结果。反过来，并不令人惊讶的是，与心理学中的许多其他领域一样，PL领域 遭遇了众多复制挑战。此外，可能是因为追随直接的脚步 与上述趋势背道而驰的是，不同的研究小组使用相同的 培训任务或结果测试。这是有问题的，因为研究表明，任务中的微小变化- 程序可能会导致学习结果的巨大差异。在这里，我们通过以下方式克服这些限制 在大型受试者中使用共同结果衡量标准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学习的不同“关键特征” 一群人。然而，这些任务从未被直接比较或对比过。建议的结果是 无论是对公共语言的基本认识，还是对如何翻译公共语言都有很大的参考价值 那些有视觉需求的人。我们将通过培训大样本的参与者来取得稳健和可靠的结果 并通过一套通用的衡量标准来评估结果。我们还将收集广泛的评估 这将提供一个独特的数据集，可以解决文献中的争议和 带来新的理解。我们提出的分析方法测试了该领域的中心关键假设， 探索不同的培训方法在多大程度上导致系统地不同的学习概况， 并研究了这些如何根据正在接受培训的个人而有所不同。进一步通过发布我们的培训 和测试工具以及收集的数据，我们将使其他小组能够模拟结果，复制我们的 研究，并对已知结果的培训任务进行明确的修改，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.