Single, but not dual, attention facilitates statistical learning of two concurrent auditory sequences

Statistical learning at a virtual cocktail party

Statistical learning - the ability to extract distributional regularities from input - is suggested to be key to language acquisition. Yet, evidence for the human capacity for statistical learning comes mainly from studies conducted in carefully controlled settings without auditory distraction. While such conditions permit careful examination of learning, they do not reflect the naturalistic language learning experience, which is replete with auditory distraction - including competing talkers. Here, we examine how statistical language learning proceeds in a virtual cocktail party environment, where the to-be-learned input is presented alongside a competing speech stream with its own distributional regularities. During exposure, participants in the Dual Talker group concurrently heard two novel languages, one produced by a female talker and one by a male talker, with each talker virtually positioned at opposite sides of the listener (left/right) using binaural acoustic manipulations. Selective attention was manipulated by instructing participants to attend to only one of the two talkers. At test, participants were asked to distinguish words from part-words for both the attended and the unattended languages. Results indicated that participants' accuracy was significantly higher for trials from the attended vs. unattended language. Further, the performance of this Dual Talker group was no different compared to a control group who heard only one language from a single talker (Single Talker group). We thus conclude that statistical learning is modulated by selective attention, being relatively robust against the additional cognitive load provided by competing speech, emphasizing its efficiency in naturalistic language learning situations.

Relationship between schematic and dynamic expectations of melodic patterns in music perception

Prediction is fundamental in music listening. Two types of expectations have been proposed: schematic expectations, which arise from knowledge of tonal regularities (e.g., harmony and key) acquired through long-term plasticity and learning, and dynamic expectations, which arise from short-term regularity representations (e.g., rhythmic patterns and melodic contours) extracted from ongoing musical contexts. Although both expectations are indispensable in music listening, how they interact with each other in music prediction remains unclear. The present study examined the relationship between schematic and dynamic expectations in music processing using event-related potentials (ERPs). At the ending note of the melodies, the schematic expectation was violated by presenting a note with music-syntactic irregular (i.e., outof- key note), while the dynamic expectation was violated by presenting a contour deviant based on online statistical learning of melodic patterns. Schematic and dynamic expectations were manipulated to predict the same note. ERPs were recorded for the music-syntactic irregularity and the contour deviant, which occurred independently or simultaneously. The results showed that the music-syntactic irregularity elicited an early right anterior negativity (ERAN), reflecting the prediction error in the schematic expectation, while the contour deviant elicited a mismatch negativity (MMN), reflecting the prediction error in the dynamic expectation. Both components occurred within a similar latency range. Moreover, the ERP amplitude was multiplicatively increased when the irregularity and deviance occurred simultaneously. These findings suggest that schematic and dynamic expectations function concurrently in an interactive manner when both expectations predict the same note.

Seeking the neural representation of statistical properties in print during implicit processing of visual words

Statistical learning (SL) plays a key role in literacy acquisition. Studies have increasingly revealed the influence of distributional statistical properties of words on visual word processing, including the effects of word frequency (lexical level) and mappings between orthography, phonology, and semantics (sub-lexical level). However, there has been scant evidence to directly confirm that the statistical properties contained in print can be directly characterized by neural activities. Using time-resolved representational similarity analysis (RSA), the present study examined neural representations of different types of statistical properties in visual word processing. From the perspective of predictive coding, an equal probability sequence with low built-in prediction precision and three oddball sequences with high built-in prediction precision were designed with consistent and three types of inconsistent (orthographically inconsistent, orthography-to-phonology inconsistent, and orthography-to-semantics inconsistent) Chinese characters as visual stimuli. In the three oddball sequences, consistent characters were set as the standard stimuli (probability of occurrence p = 0.75) and three types of inconsistent characters were set as deviant stimuli (p = 0.25), respectively. In the equal probability sequence, the same consistent and inconsistent characters were presented randomly with identical occurrence probability (p = 0.25). Significant neural representation activities of word frequency were observed in the equal probability sequence. By contrast, neural representations of sub-lexical statistics only emerged in oddball sequences where short-term predictions were shaped. These findings reveal that the statistical properties learned from long-term print environment continues to play a role in current word processing mechanisms and these mechanisms can be modulated by short-term predictions.

Perceptual uncertainty modulates auditory statistical learning: A magnetoencephalography study

Statistical learning allows comprehension of structured information, such as that in language and music. The brain computes a sequence's transition probability and predicts future states to minimise sensory reaction and derive entropy (uncertainty) from sequential information. Neurophysiological studies have revealed that early event-related neural responses (P1 and N1) reflect statistical learning - when the brain encodes transition probability in stimulus sequences, it predicts an upcoming stimulus with a high transition probability and suppresses the early event-related responses to a stimulus with a high transition probability. This amplitude difference between high and low transition probabilities reflects statistical learning effects. However, how a sequence's transition probability ratio affects neural responses contributing to statistical learning effects remains unknown. This study investigated how transition-probability ratios or conditional entropy (uncertainty) in auditory sequences modulate the early event-related neuromagnetic responses of P1m and N1m. Sequence uncertainties were manipulated using three different transition-probability ratios: 90:10%, 80:20%, and 67:33% (conditional entropy: 0.47, 0.72, and 0.92 bits, respectively). Neuromagnetic responses were recorded when participants listened to sequential sounds with these three transition probabilities. Amplitude differences between lower and higher probabilities were larger in sequences with transition-probability ratios of 90:10% and smaller in sequences with those of 67:33%, compared to sequences with those of 80:20%. This suggests that the transition-probability ratio finely tunes P1m and N1m. Our study also showed larger amplitude differences between frequent- and rare-transition stimuli in P1m than in N1m. This indicates that information about transition-probability differences may be calculated in earlier cognitive processes.

Statistical Properties in Jazz Improvisation Underline Individuality of Musical Representation

Statistical learning is an innate function in the brain and considered to be essential for producing and comprehending structured information such as music. Within the framework of statistical learning the brain has an ability to calculate the transitional probabilities of sequences such as speech and music, and to predict a future state using learned statistics. This paper computationally examines whether and how statistical learning and knowledge partially contributes to musical representation in jazz improvisation. The results represent the time-course variations in a musician’s statistical knowledge. Furthermore, the findings show that improvisational musical representation might be susceptible to higher- but not lower-order statistical knowledge (i.e., knowledge of higher-order transitional probability). The evidence also demonstrates the individuality of improvisation for each improviser, which in part depends on statistical knowledge. Thus, this study suggests that statistical properties in jazz improvisation underline individuality of musical representation.

Statistical learning and the uncertainty of melody and bass line in music

Statistical learning is the ability to learn based on transitional probability (TP) in sequential information, which has been considered to contribute to creativity in music. The interdisciplinary theory of statistical learning examines statistical learning as a mechanism of human learning. This study investigated how TP distribution and conditional entropy in TP of the melody and bass line in music interact with each other, using the highest and lowest pitches in Beethoven’s piano sonatas and Johann Sebastian Bach’s Well-Tempered Clavier. Results for the two composers were similar. First, the results detected specific statistical characteristics that are unique to each melody and bass line as well as general statistical characteristics that are shared between the melody and bass line. Additionally, a correlation of the conditional entropies sampled from the TP distribution could be detected between the melody and bass line. This suggests that the variability of entropies interacts between the melody and bass line. In summary, this study suggested that TP distributions and the entropies of the melody and bass line interact with but are partly independent of each other.

Tonality Tunes the Statistical Characteristics in Music: Computational Approaches on Statistical Learning

Statistical learning is a learning mechanism based on transition probability in sequences such as music and language. Recent computational and neurophysiological studies suggest that the statistical learning contributes to production, action, and musical creativity as well as prediction and perception. The present study investigated how statistical structure interacts with tonalities in music based on various-order statistical models. To verify this in all 24 major and minor keys, the transition probabilities of the sequences containing the highest pitches in Bach's Well-Tempered Clavier, which is a collection of two series (No. 1 and No. 2) of preludes and fugues in all of the 24 major and minor keys, were calculated based on nth-order Markov models. The transition probabilities of each sequence were compared among tonalities (major and minor), two series (No. 1 and No. 2), and music types (prelude and fugue). The differences in statistical characteristics between major and minor keys were detected in lower- but not higher-order models. The results also showed that statistical knowledge in music might be modulated by tonalities and composition periods. Furthermore, the principal component analysis detected the shared components of related keys, suggesting that the tonalities modulate statistical characteristics in music. The present study may suggest that there are at least two types of statistical knowledge in music that are interdependent on and independent of tonality, respectively.

Concurrent Statistical Learning of Ignored and Attended Sound Sequences: An MEG Study

In an auditory environment, humans are frequently exposed to overlapping sound sequences such as those made by human voices and musical instruments, and we can acquire information embedded in these sequences via attentional and nonattentional accesses. Whether the knowledge acquired by attentional accesses interacts with that acquired by nonattentional accesses is unknown, however. The present study examined how the statistical learning (SL) of two overlapping sound sequences is reflected in neurophysiological and behavioral responses, and how the learning effects are modulated by attention to each sequence. SL in this experimental paradigm was reflected in a neuromagnetic response predominantly in the right hemisphere, and the learning effects were not retained when attention to the tone streams was switched during the learning session. These results suggest that attentional and nonattentional learning scarcely interact with each other and that there may be a specific system for nonattentional learning, which is independent of attentional learning.

Depth and the Uncertainty of Statistical Knowledge on Musical Creativity Fluctuate Over a Composer's Lifetime

Brain models music as a hierarchy of dynamical systems that encode probability distributions and complexity (i.e., entropy and uncertainty). Through musical experience over lifetime, a human is intrinsically motivated in optimizing the internalized probabilistic model for efficient information processing and the uncertainty resolution, which has been regarded as rewords. Human's behavior, however, appears to be not necessarily directing to efficiency but sometimes act inefficiently in order to explore a maximum rewards of uncertainty resolution. Previous studies suggest that the drive for novelty seeking behavior (high uncertain phenomenon) reflects human's curiosity, and that the curiosity rewards encourage humans to create and learn new regularities. That is to say, although brain generally minimizes uncertainty of music structure, we sometimes derive pleasure from music with uncertain structure due to curiosity for novelty seeking behavior by which we anticipate the resolution of uncertainty. Few studies, however, investigated how curiosity for uncertain and novelty seeking behavior modulates musical creativity. The present study investigated how the probabilistic model and the uncertainty in music fluctuate over a composer's lifetime (all of the 32 piano sonatas by Ludwig van Beethoven). In the late periods of the composer's lifetime, the transitional probabilities (TPs) of sequential patterns that ubiquitously appear in all of his music (familiar phrase) were decreased, whereas the uncertainties of the whole structure were increased. Furthermore, these findings were prominent in higher-, rather than lower-, order models of TP distribution. This may suggest that the higher-order probabilistic model is susceptible to experience and psychological phenomena over the composer's lifetime. The present study first suggested the fluctuation of uncertainty of musical structure over a composer's lifetime. It is suggested that human's curiosity for uncertain and novelty seeking behavior may modulate optimization and creativity in human's brain.

Entropy, Uncertainty, and the Depth of Implicit Knowledge on Musical Creativity: Computational Study of Improvisation in Melody and Rhythm

Recent neurophysiological and computational studies have proposed the hypothesis that our brain automatically codes the nth-order transitional probabilities (TPs) embedded in sequential phenomena such as music and language (i.e., local statistics in nth-order level), grasps the entropy of the TP distribution (i.e., global statistics), and predicts the future state based on the internalized nth-order statistical model. This mechanism is called statistical learning (SL). SL is also believed to contribute to the creativity involved in musical improvisation. The present study examines the interactions among local statistics, global statistics, and different levels of orders (mutual information) in musical improvisation interact. Interactions among local statistics, global statistics, and hierarchy were detected in higher-order SL models of pitches, but not lower-order SL models of pitches or SL models of rhythms. These results suggest that the information-theoretical phenomena of local and global statistics in each order may be reflected in improvisational music. The present study proposes novel methodology to evaluate musical creativity associated with SL based on information theory.

Musical Creativity and Depth of Implicit Knowledge: Spectral and Temporal Individualities in Improvisation

It has been suggested that musical creativity is mainly formed by implicit knowledge. However, the types of spectro-temporal features and depth of the implicit knowledge forming individualities of improvisation are unknown. This study, using various-order Markov models on implicit statistical learning, investigated spectro-temporal statistics among musicians. The results suggested that lower-order models on implicit knowledge represented general characteristics shared among musicians, whereas higher-order models detected specific characteristics unique to each musician. Second, individuality may essentially be formed by pitch but not rhythm, whereas the rhythms may allow the individuality of pitches to strengthen. Third, time-course variation of musical creativity formed by implicit knowledge and uncertainty (i.e., entropy) may occur in a musician's lifetime. Individuality of improvisational creativity may be formed by deeper but not superficial implicit knowledge of pitches, and that the rhythms may allow the individuality of pitches to strengthen. Individualities of the creativity may shift over a musician's lifetime via experience and training.

Motor Reproduction of Time Interval Depends on Internal Temporal Cues in the Brain: Sensorimotor Imagery in Rhythm

How the human brain perceives time intervals is a fascinating topic that has been explored in many fields of study. This study examined how time intervals are replicated in three conditions: with no internalized cue (PT), with an internalized cue without a beat (AS), and with an internalized cue with a beat (RS). In PT, participants accurately reproduced the time intervals up to approximately 3 s. Over 3 s, however, the reproduction errors became increasingly negative. In RS, longer presentations of over 5.6 s and 13 beats induced accurate time intervals in reproductions. This suggests longer exposure to beat presentation leads to stable internalization and efficiency in the sensorimotor processing of perception and reproduction. In AS, up to approximately 3 s, the results were similar to those of RS whereas over 3 s, the results shifted and became similar to those of PT. The time intervals between the first two stimuli indicate that the strategies of time-interval reproduction in AS may shift from RS to PT. Neural basis underlying the reproduction of time intervals without a beat may depend on length of time interval between adjacent stimuli in sequences.

Auditory Statistical Learning During Concurrent Physical Exercise and the Tolerance for Pitch, Tempo, and Rhythm Changes

Previous studies suggest that statistical learning is preserved when acoustic changes are made to auditory sequences. However, statistical learning effects can vary with and without concurrent exercise. The present study examined how concurrent physical exercise influences auditory statistical learning when acoustical and temporal changes are made to auditory sequences. Participants were presented with the 500-tone sequences based on a Markov chain while cycling or resting in ignored and attended conditions. Learning effects were evaluated using a familiarity test with four types of short tone series: tone series in which stimuli were same as 500-tone sequence and three tone series in which frequencies, tempo, or rhythm was changed. We suggested that, regardless of attention, concurrent exercise interferes with tolerance in statistical learning for rhythm, rather than tempo changes. There may be specific relationships among statistical learning, rhythm perception, and motor system underlying physical exercise.

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

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.

Time-course variation of statistics embedded in music: Corpus study on implicit learning and knowledge

Learning and knowledge of transitional probability in sequences like music, called statistical learning and knowledge, are considered implicit processes that occur without intention to learn and awareness of what one knows. This implicit statistical knowledge can be alternatively expressed via abstract medium such as musical melody, which suggests this knowledge is reflected in melodies written by a composer. This study investigates how statistics in music vary over a composer's lifetime. Transitional probabilities of highest-pitch sequences in Ludwig van Beethoven's Piano Sonata were calculated based on different hierarchical Markov models. Each interval pattern was ordered based on the sonata opus number. The transitional probabilities of sequential patterns that are musical universal in music gradually decreased, suggesting that time-course variations of statistics in music reflect time-course variations of a composer's statistical knowledge. This study sheds new light on novel methodologies that may be able to evaluate the time-course variation of composer's implicit knowledge using musical scores.