Department of Clinical Laboratory, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan; International Research Center for Neurointelligence, The University of Tokyo, Tokyo, Japan.
Department of Clinical Laboratory, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan; Advanced Medical Science Research Center, Gunma Paz University, Gunma, Japan.
Neuropsychologia. 2020 Sep;146:107553. doi: 10.1016/j.neuropsychologia.2020.107553. Epub 2020 Jul 8.
The brain extracts statistical regularities from sequential information around our environment. This is referred to as statistical learning (SL). Statistical learning is considered an innate function in the human brain and contributes to the brain's development. Within the framework of predictive coding, this learning system allows us to predict a future state to minimize sensory reaction and resolve uncertainty around the world. By auditory statistical learning, over the brain's development, humans become able to comprehend language and music. An increasing number of studies has revealed that Western-classical musical training optimizes the brain's probabilistic model of music and enhances the accuracy of perceptive uncertainty (entropy) in newly encountered melody. No study, however, investigates how musical training modulates the probabilistic model of rhythm, and how the musical culture tunes them. The present study investigated how SL of temporal sequences with and without a beat is reflected in neural responses, and how the SL is modulated by the two types of musical training in different cultures: Western- and Japanese-classical music (i.e., Hougaku). The neural representation showed evidence that the SL effects of beat sequence were prominent in the left hemisphere. This finding was larger in Western- and Japanese-classical musicians compared with non-musicians. Further, the entropy (uncertainty) of the sequences negatively correlated with neural effects of SL, mainly in the left hemisphere of the both Western- and Japanese-classical musicians. These suggest that, regardless of musical culture, musical training may generally facilitate SL of rhythm. However, the specific neural components showed differences between groups of musicians: an earlier component, referred to as P1, represented the left lateralization for perceptive uncertainty in both groups of musicians, whereas a later component, referred to as N1, represented the left lateralization only in Japanese Classical musicians. These findings may suggest that the types of musical training differently modulate neural representation of underlying temporal SL, particularly global processing of uncertainty rather than local processing of transitional probability. The present study sheds new light on the neurophysiological account of Japanese classical music.
大脑从我们周围环境的序列信息中提取统计规律。这被称为统计学习（SL）。统计学习被认为是人类大脑的一种先天功能，有助于大脑的发育。在预测编码的框架内，这个学习系统允许我们预测未来的状态，以最小化感官反应并解决世界上的不确定性。通过听觉统计学习，在大脑的发育过程中，人类能够理解语言和音乐。越来越多的研究表明，西方古典音乐训练优化了大脑对音乐的概率模型，并提高了对新遇到的旋律感知不确定性（熵）的准确性。然而，没有研究调查音乐训练如何调节节奏的概率模型，以及音乐文化如何调整它们。本研究调查了有节奏和无节奏的时间序列的 SL 如何反映在神经反应中，以及两种不同文化的两种类型的音乐训练（即和乐）如何调节 SL：西方古典音乐和日本古典音乐（即和乐）。神经表现表明，有节奏的序列的 SL 效应在左半球中更为突出。与非音乐家相比，西方和日本古典音乐家中的这种发现更大。此外，序列的熵（不确定性）与 SL 的神经效应呈负相关，主要在西方和日本古典音乐家的左半球。这些表明，无论音乐文化如何，音乐训练通常都可以促进节奏的 SL。然而，特定的神经成分在音乐家群体之间存在差异：称为 P1 的较早成分表示两种音乐家群体的感知不确定性的左偏侧化，而称为 N1 的较晚成分仅表示日本古典音乐家的左偏侧化。这些发现可能表明，音乐训练的类型不同地调节潜在时间 SL 的神经表现，特别是对不确定性的全局处理，而不是对转移概率的局部处理。本研究为日本古典音乐的神经生理学解释提供了新的视角。
