Training under high interference conditions through interleaved practice (IP) results in performance suppression during training but enhances long-term performance relative to repetitive practice (RP) involving low interference. Previous neuroimaging work addressing this contextual interference effect of motor learning has relied heavily on the blood-oxygen-level-dependent (BOLD) response using functional magnetic resonance imaging (fMRI) methodology resulting in mixed reports of prefrontal cortex (PFC) recruitment under IP and RP conditions. We sought to clarify these equivocal findings by imaging bilateral PFC recruitment using functional near-infrared spectroscopy (fNIRS) while discrete key pressing sequences were trained under IP and RP schedules and subsequently tested following a 24-hour delay. An advantage of fNIRS over the fMRI BOLD response, is that the former measures oxygenated and deoxygenated hemoglobin changes independently allowing for assessment of cortical hemodynamics even when there is neurovascular decoupling. Despite slower sequence performance durations under IP, bilateral PFC oxygenated and deoxygenated hemoglobin values did not differ between practice conditions. At test, however, slower performance from those previously trained under RP coincided with hemispheric asymmetry in PFC recruitment. Specifically, following RP, test deoxygenated hemoglobin values were significantly lower in the right PFC. The present findings contrast with previous behavioral demonstrations of increased cognitive demand under IP to illustrate a more complex involvement of the PFC in the contextual interference effect. IP and RP incur similar levels of bilateral PFC recruitment, but the processes underlying this recruitment are dissimilar. PFC recruitment during IP supports action reconstruction and memory elaboration while RP relies on PFC recruitment to maintain task variation information in working memory from trial to trial. While PFC recruitment under RP serves to enhance immediate performance, it does not support long-term performance.

中文翻译：

---

交错和重复练习下运动序列学习过程中前额叶皮层的激活：双通道近红外光谱研究

通过交错练习 (IP) 在高干扰条件下进行训练会导致训练期间的表现受到抑制，但相对于涉及低干扰的重复练习 (RP) 而言，可以提高长期表现。先前针对运动学习的这种背景干扰效应的神经影像学工作在很大程度上依赖于使用功能磁共振成像（fMRI）方法的血氧水平依赖性（BOLD）反应，导致前额皮质（PFC）在IP和RP条件。我们试图通过使用功能性近红外光谱 (fNIRS) 对双侧 PFC 募集进行成像来澄清这些模棱两可的发现，同时在 IP 和 RP 计划下训练离散按键序列，并随后在 24 小时延迟后进行测试。fNIRS 相对于 fMRI BOLD 响应的优势在于，前者独立测量氧合和脱氧血红蛋白变化，即使存在神经血管解耦，也可以评估皮质血流动力学。尽管 IP 下的序列执行持续时间较慢，但双侧 PFC 含氧和脱氧血红蛋白值在实践条件下没有差异。然而，在测试中，那些先前在 RP 下训练的人表现较慢，与 PFC 募集的半球不对称相一致。具体来说，在 RP 后，右侧 PFC 的测试脱氧血红蛋白值显着降低。目前的研究结果与之前 IP 下认知需求增加的行为论证形成对比，说明 PFC 在情境干扰效应中的参与更为复杂。IP 和 RP 的双边 PFC 招募水平相似，但这种招募的过程不同。IP 期间的 PFC 招募支持动作重建和记忆细化，而 RP 则依赖 PFC 招募来维护工作记忆中每次试验的任务变化信息。虽然 RP 下的 PFC 招募有助于提高即时绩效，但它并不支持长期绩效。