Evidence for A Neural System For Learning

我们如何学习？大脑中的哪些神经系统负责人类的学习？我们可以使用脑电图（EEG;即，脑电波）来找到一个可靠的神经标记诱发的处理性能反馈，这是一个潜在的学习过程的指示？在过去的20年里，大量的研究已经使用EEG来检查由表现反馈引起的神经信号。特别是，在人类中使用EEG的研究揭示了被认为与学习有关的表现反馈引起的两个关键信号：（1）反馈相关的负性（FRN）和（2）奖励积极性（RP）。理论解释认为，FRN和RP反映了人类内侧额叶皮层内强化学习系统的功能。然而，到目前为止，它仍然不清楚是否FRN和RP只是由反馈引起的，或者它们是否实际上反映了一个潜在的学习过程。这项研究试图回答这个问题，并证明FRN和RP是人类内侧额叶皮层内学习系统的神经标记。 拟议的研究有三个短期目标的基础上脑电图方法。首先，我将证明FRN和RP随时间的变化反映了理论预测。具体来说，在这里，我们将证明FRN和RP随时间的变化类似于行为数据中的学习曲线。第二，我将检查在FRN和RP中看到的与学习相关的变化是否稳定，并且对影响保留的不同实践结构敏感。例如，人们早就知道，间隔开练习会话（即，实践的分布）导致更多的学习。因此，如果FRN和RP反映了潜在的过程，则这些EEG分量应该对练习的分布敏感。第三，我将确定FRN和RP等信号是否与随后的行为变化相关并可预测随后的行为变化。事实上，如果FRN和RP反映了学习过程，那么这些EEG分量的大小应该与行为修改的大小相关-较大的错误信号应该与较大的行为变化相关。 总之，这三个研究目标的答案-如果支持的话-将证明FRN和RP是人类大脑内神经学习系统的标志物。重要的是，这些发现将证明我们有一个可靠的神经标记，使我们能够直接衡量人类的学习，而不是仅仅依靠间接的行为措施-这一结果将对加拿大的教育和工业产生巨大影响。

How do we learn? What neural systems within the brain are responsible for human learning? Can we use electroencephalography (EEG; i.e., brain waves) to find a reliable neural marker evoked by the processing of performance feedback that is indicative of an underlying learning process? In the past 20 years, a multitude of studies have used EEG to examine the neural signals elicited by performance feedback. In particular, studies using EEG in humans have revealed two key signals evoked by performance feedback that are thought to relate to learning: (1) the feedback-related negativity (FRN) and (2) and reward positivity (RP). Theoretical accounts posit that the FRN and RP reflect the functioning of a reinforcement learning system within the human medial-frontal cortex. However, to date it is still unclear whether the FRN and RP are simply evoked by feedback or whether they are actually reflective of an underlying learning process. The proposed research seeks to answer this question and demonstrate that the FRN and RP are neural markers for a learning system within the human-medial-frontal cortex. The proposed research has three short-term objectives based on EEG methodologies. First, I will demonstrate that changes in the FRN and RP over time mirror theoretical predictions. Specifically, here we will demonstrate that changes in the FRN and RP over time resemble the learning curves seen in behavioural data. Second, I will examine if learning-related changes seen in the FRN and RP are stable and sensitive to different practice constructs that impact retention. For example, it has long been known that spacing out practice sessions (i.e., distribution of practice) results in greater learning. As such, if the FRN and RP are reflective of an underlying process these EEG components should be sensitive to distribution of practice. Third, I will determine if signals such as the FRN and RP are related to and predictive of subsequent changes in behaviour. Indeed, if the FRN and RP are reflective of a learning process then the size of these EEG components should be related to the size of behavioural modifications - larger error signals should relate to larger changes in behaviour. Together, the answers of these three research objectives - if supportive - will demonstrate that the FRN and RP are markers for a neural learning system within the human brain. Importantly, these findings will demonstrate that we have a reliable neural marker that will allow us to gauge human learning directly instead of relying solely on indirect behavioural measures - an outcome that will have a huge impact in education and industry across Canada.