Cognitive predictors of second language learning success

学习区分非母语语音是很有挑战性的，尤其是当这些语音的区别不是母语的一部分，或者依赖于特定的语音线索，而不是母语的一部分。虽然随着学习时间的推移，这些问题可能会在数月或数年的沉浸式学习环境中持续存在。有假说认为，接触多个说话者的语音自然变化有助于学习，但目前尚不清楚这种变化有多重要，也不清楚这种作用是否取决于学习者的能力。在个体层面上，即使在受控的实验室训练模式下，学习的数量和最终达到的程度也显示出巨大的个体差异。虽然有迹象表明，最初的天赋可以预测学习能力，但目前还不清楚是什么预测了最初的天赋。一些研究发现，长期结果与认知因素（如非言语听觉处理或注意力）之间存在关联。然而，目前尚不清楚这些因素是否可以预测学习，或者接触第二语言是否会影响认知技能。本研究计划的长期目标是了解影响第二语言语音学习的认知因素。这包括个人特质水平的因素和训练水平的因素。该建议有三个具体的短期目标：目标1：了解接触不同的语音是否以及在多大程度上对稳健的学习（训练水平）至关重要。目的2：测试哪些个体差异因素可以预测初始阶段的学习能力（特质水平）目的3：揭示识别因素与双语对认知能力的影响的预测作用（两个水平）学习新声音的可变性的重要性影响了言语感知和婴儿语言发展的模型。因此，更深入地了解多语者差异在学习中的作用，将对这些模型以及我们学习新挑战技能的模型产生影响。识别预测初始表现和学习的认知因素告诉我们，学习新的复杂听觉类别的机制和技能是什么。它还指出了哪些个人在学习环境中需要额外的支持和可能的额外输入。最后，研究结果将对第二语言学习产生影响，无论是在教育环境中，还是在基于应用程序和计算机的培训项目中。

Learning to distinguish between non-native speech sounds can be challenging, especially when those sounds make distinctions that are not part of the native language or rely on particular acoustic-phonetic cues in a way that is not part of the native language. While there is generally learning over time, these problems can persist even after months or years of learning in an immersion environment. Exposure to natural variation in speech sounds from multiple talkers has been hypothesized to facilitate learning, however it's not clear how critical this variation is or whether the role depends on the abilities of the learner. At the individual level, large individual differences have been shown in amount of learning and degree of ultimate attainment even under controlled laboratory training paradigms. And while there are indications that initial aptitude predicts learning, it is not clear what predicts initial aptitude. Some studies have found associations between long term outcomes and cognitive factors such as non-speech auditory processing or attention. However, it's not clear if these factors are predictive of learning or if exposure to a second language may affect cognitive skills. The long-term objective of this research program is to understand the cognitive factors affecting learning of second language speech sounds. This includes both individual trait-level factors and training-level factors. This proposal has three specific short-term goals: Aim 1: To understand whether and to what extent exposure to varied speech sounds is critical to robust learning (training level). Aim 2: To test which individual difference factors predict aptitude for learning at the initial stages (trait-level) Aim 3: To disentangle the predictive role of ID factors versus bilingualism effects on cognitive abilities (both levels) The importance of variability in learning new sounds has influenced models of speech perception and infant language development. Thus, a more thorough understanding of the role of multi-talker variation on learning will have impacts on these models as well as on our models of learning new challenging skills. Identifying the cognitive factors that predict initial performance and learning tells us what mechanisms and skills underlie learning of novel complex auditory categories. It also points towards which individuals will need extra support and possibly extra input in a learning context. Finally, the results will have implications for second language learning, both in educational settings and for app-based and computer-based training programs.