Neurophysiological Markers of Statistical Learning in Music and Language: Hierarchy, Entropy, and Uncertainty

Abstract

Statistical learning (SL) is a method of learning based on the transitional probabilities embedded in sequential phenomena such as music and language. It has been considered an implicit and domain-general mechanism that is innate in the human brain and that functions independently of intention to learn and awareness of what has been learned. SL is an interdisciplinary notion that incorporates information technology, artificial intelligence, musicology, and linguistics, as well as psychology and neuroscience. A body of recent study has suggested that SL can be reflected in neurophysiological responses based on the framework of information theory. This paper reviews a range of work on SL in adults and children that suggests overlapping and independent neural correlations in music and language, and that indicates disability of SL. Furthermore, this article discusses the relationships between the order of transitional probabilities (TPs) (i.e., hierarchy of local statistics) and entropy (i.e., global statistics) regarding SL strategies in human’s brains; claims importance of information-theoretical approaches to understand domain-general, higher-order, and global SL covering both real-world music and language; and proposes promising approaches for the application of therapy and pedagogy from various perspectives of psychology, neuroscience, computational studies, musicology, and linguistics.

Towards a standard model of musical improvisation

“…For high‐level features that may link the parsing and planning networks in improvisation, we can take cues from work on statistical learning in music. Hansen and Pearce () and Daikoku () both detail the utility of Markov models in predicting uncertainty over time in musical output. Hansen and Pearce () found training‐based differences in expectations from musicians and non‐musicians with musicians displaying lower degrees of uncertainty for pieces of varying complexity during music listening.…”

“…Hansen and Pearce () found training‐based differences in expectations from musicians and non‐musicians with musicians displaying lower degrees of uncertainty for pieces of varying complexity during music listening. Daikoku () expanded this further by applying two‐process statistical learning proposed by Thiessen, Kronstein, and Hufnagle () to music. In this framework, local statistics (transitional probabilities) are integrated into summary statistics to generate predictions for novel sequences.…”

Toward a standard model of musical improvisation

“…For high-level features that may link the parsing and planning networks in improvisation, we can take cues from work on statistical learning in music. Hansen and Pearce (2014) and Daikoku (2018) both detail the utility of Markov models in predicting uncertainty over time in musical output. Hansen and Pearce (2014) found training-based differences in expectations from musicians and non-musicians with musicians displaying lower degrees of uncertainty for pieces of varying complexity during music listening.…”

“…Hansen and Pearce (2014) found training-based differences in expectations from musicians and non-musicians with musicians displaying lower degrees of uncertainty for pieces of varying complexity during music listening. Daikoku (2018) expanded this further applying two-process statistical learning proposed by Thiessen et al (2013) to music. In this framework, local statistics (transitional probabilities) are integrated into summary statistics to generate predictions for novel sequences.…”

Musical expertise is associated with improved neural statistical learning

“…Musicians have better performances than non-musicians in tasks that implicate different levels of SL complexity. For example, musicians have a greater neural sensitivity to the frequency of occurrence of items than non-musicians 7 , they are better at segmenting words from an artificial language stream 8 , they have a greater neural sensitivity to 1 st order Markov probability 9 and to more complex statistics [10][11][12] . However, it is not clear whether these differences arise from an improved ability to learn sequence statistics -and if so, at which level of complexity -or from an improved ability to process sensory information.…”