Learning in chemistry and other areas of science involves developing one's mental models of invisible processes and manipulating temporal and spatial domains during visual information processing. While some aspects learning have been well studied by EEG (e.g., theta and gamma oscillations), the role of spontaneous and scale-free brain activity remains unclear. We used a continuous chemistry learning EEG paradigm to explore how scale-free brain activity is related learning. We found a learning effect in participants (N = 22) with an increase in test accuracy (learning gain) and decrease in test question response times in a counterbalanced pre/post-test experiment. In the brain we found increased overall (mixed) broadband power (1-50 Hz) during learning compared to rest. We then used the IRASA method to separate oscillatory and fractal (i.e. scale-free) spectral components and observed an increase in low-frequency oscillatory band powers during learning. More importantly, we found that fractal power increased during the learning sessions relative to oscillatory power. Finally, the structure of the fractal power spectra (PLE) correlated to the individual participants' learning gains. These findings support the importance of scale-free activity for learning from a complex visual paradigm. We tentatively hypothesize that this fractal component is involved in integrating the different time scales of the learning material with those of the spontaneous activity during learning and mental model shaping.

中文翻译：

---

学习是无尺度的吗？化学学习增加 EEG 分形功率并改变幂律指数

化学和其他科学领域的学习涉及开发一个人的无形过程的心理模型，并在视觉信息处理过程中操纵时空域。虽然 EEG 已经很好地研究了学习的某些方面（例如，theta 和 gamma 振荡），但自发和无标度大脑活动的作用仍不清楚。我们使用连续化学学习 EEG 范式来探索无标度大脑活动如何与学习相关。我们发现参与者（N = 22）的学习效果随着测试准确性（学习增益）的增加和测试问题响应时间的减少在平衡的前/后测试实验中产生。在大脑中，我们发现与休息相比，学习期间的整体（混合）宽带功率（1-50 Hz）增加。然后我们使用 IRASA 方法来分离振荡和分形（即 无标度）频谱分量并观察到在学习过程中低频振荡带功率的增加。更重要的是，我们发现相对于振荡功率，在学习过程中分形功率增加。最后，分形功率谱 (PLE) 的结构与个体参与者的学习收益相关。这些发现支持了无标度活动对于从复杂的视觉范式中学习的重要性。我们初步假设这个分形成分参与了学习材料的不同时间尺度与学习和心智模型塑造过程中自发活动的时间尺度的整合。我们发现在学习过程中分形功率相对于振荡功率有所增加。最后，分形功率谱 (PLE) 的结构与个体参与者的学习收益相关。这些发现支持了无标度活动对于从复杂的视觉范式中学习的重要性。我们初步假设这个分形成分参与了学习材料的不同时间尺度与学习和心智模型塑造过程中自发活动的时间尺度的整合。我们发现在学习过程中分形功率相对于振荡功率有所增加。最后，分形功率谱 (PLE) 的结构与个体参与者的学习收益相关。这些发现支持了无标度活动对于从复杂的视觉范式中学习的重要性。我们初步假设这个分形成分参与了学习材料的不同时间尺度与学习和心智模型塑造过程中自发活动的时间尺度的整合。