Spatiotemporal brain hierarchies of auditory memory recognition and predictive coding.

Our brain is constantly extracting, predicting, and recognising key spatiotemporal features of the physical world in order to survive. While neural processing of visuospatial patterns has been extensively studied, the hierarchical brain mechanisms underlying conscious recognition of auditory sequences and the associated prediction errors remain elusive. Using magnetoencephalography (MEG), we describe the brain functioning of 83 participants during recognition of previously memorised musical sequences and systematic variations. The results show feedforward connections originating from auditory cortices, and extending to the hippocampus, anterior cingulate gyrus, and medial cingulate gyrus. Simultaneously, we observe backward connections operating in the opposite direction. Throughout the sequences, the hippocampus and cingulate gyrus maintain the same hierarchical level, except for the final tone, where the cingulate gyrus assumes the top position within the hierarchy. The evoked responses of memorised sequences and variations engage the same hierarchical brain network but systematically differ in terms of temporal dynamics, strength, and polarity. Furthermore, induced-response analysis shows that alpha and beta power is stronger for the variations, while gamma power is enhanced for the memorised sequences. This study expands on the predictive coding theory by providing quantitative evidence of hierarchical brain mechanisms during conscious memory and predictive processing of auditory sequences.

Moderate associations between BDNF Val66Met gene polymorphism, musical expertise, and mismatch negativity.

Auditory predictive processing relies on a complex interaction between environmental, neurophysiological, and genetic factors. In this view, the mismatch negativity (MMN) and intensive training on a musical instrument for several years have been used for studying environment-driven neural adaptations in audition. In addition, brain-derived neurotrophic factor (BDNF) has been shown crucial for both the neurogenesis and the later adaptation of the auditory system. The functional single-nucleotide polymorphism (SNP) Val66Met (rs6265) in the BDNF gene can affect BDNF protein levels, which are involved in neurobiological and neurophysiological processes such as neurogenesis and neuronal plasticity. In this study, we hypothesised that genetic variation within the BDNF gene would be associated with different levels of neuroplasticity of the auditory cortex in 74 musically trained participants. To achieve this goal, musicians and non-musicians were recruited and divided in Val/Val and Met- (Val/Met and Met/Met) carriers and their brain activity was measured with magnetoencephalography (MEG) while they listened to a regular auditory sequence eliciting different types of prediction errors. MMN responses indexing those prediction errors were overall enhanced in Val/Val carriers who underwent intensive musical training, compared to Met-carriers and non-musicians with either genotype. Although this study calls for replications with larger samples, our results provide a first glimpse of the possible role of gene-regulated neurotrophic factors in the neural adaptations of automatic predictive processing in the auditory domain after long-term training.

Singing training predicts increased insula connectivity with speech and respiratory sensorimotor areas at rest.

The insula contributes to the detection of salient events during goal-directed behavior and participates in the coordination of motor, multisensory, and cognitive systems. Recent task-fMRI studies with trained singers suggest that singing experience can enhance the access to these resources. However, the long-term effects of vocal training on insula-based networks are still unknown. In this study, we employed resting-state fMRI to assess experience-dependent differences in insula co-activation patterns between conservatory-trained singers and non-singers. Results indicate enhanced bilateral anterior insula connectivity in singers relative to non-singers with constituents of the speech sensorimotor network. Specifically, with the cerebellum (lobule V-VI) and the superior parietal lobes. The reversed comparison showed no effects. The amount of accumulated singing training predicted enhanced bilateral insula co-activation with primary sensorimotor areas representing the diaphragm and the larynx/phonation area-crucial regions for cortico-motor control of complex vocalizations-as well as the bilateral thalamus and the left putamen. Together, these findings highlight the neuroplastic effect of expert singing training on insula-based networks, as evidenced by the association between enhanced insula co-activation profiles in singers and the brain's speech motor system components.

Brain recognition of previously learned versus novel temporal sequences: a differential simultaneous processing.

Memory for sequences is a central topic in neuroscience, and decades of studies have investigated the neural mechanisms underlying the coding of a wide array of sequences extended over time. Yet, little is known on the brain mechanisms underlying the recognition of previously memorized versus novel temporal sequences. Moreover, the differential brain processing of single items in an auditory temporal sequence compared to the whole superordinate sequence is not fully understood. In this magnetoencephalography (MEG) study, the items of the temporal sequence were independently linked to local and rapid (2-8 Hz) brain processing, while the whole sequence was associated with concurrent global and slower (0.1-1 Hz) processing involving a widespread network of sequentially active brain regions. Notably, the recognition of previously memorized temporal sequences was associated to stronger activity in the slow brain processing, while the novel sequences required a greater involvement of the faster brain processing. Overall, the results expand on well-known information flow from lower- to higher order brain regions. In fact, they reveal the differential involvement of slow and faster whole brain processing to recognize previously learned versus novel temporal information.

A magnetoencephalography study of first-time mothers listening to infant cries.

Studies using magnetoencephalography (MEG) have identified the orbitofrontal cortex (OFC) to be an important early hub for a "parental instinct" in the brain. This complements the finding from functional magnetic resonance imaging studies linking reward, emotion regulation, empathy, and mentalization networks to the "parental brain." Here, we used MEG in 43 first-time mothers listening to infant and adult cry vocalizations to investigate the link with mother-infant postpartum bonding scores and their level of sleep deprivation (assessed using both actigraphy and sleep logs). When comparing brain responses to infant versus adult cry vocalizations, we found significant differences at around 800-1,000 ms after stimuli onset in the primary auditory cortex, superior temporal gyrus, hippocampal areas, insula, precuneus supramarginal gyrus, postcentral gyrus, and posterior cingulate gyrus. Importantly, mothers with weaker bonding scores showed decreased brain responses to infant cries in the auditory cortex, middle and superior temporal gyrus, OFC, hippocampal areas, supramarginal gyrus, and inferior frontal gyrus at around 100-300 ms after the stimulus onset. In contrast, we did not find correlations with sleep deprivation scores. The significant decreases in brain processing of an infant's distress signals could potentially be a novel signature of weaker infant bonding in new mothers and should be investigated in vulnerable populations.

Dissociated brain functional connectivity of fast versus slow frequencies underlying individual differences in fluid intelligence: a DTI and MEG study.

Brain network analysis represents a powerful technique to gain insights into the connectivity profile characterizing individuals with different levels of fluid intelligence (Gf). Several studies have used diffusion tensor imaging (DTI) and slow-oscillatory resting-state fMRI (rs-fMRI) to examine the anatomical and functional aspects of human brain networks that support intelligence. In this study, we expand this line of research by investigating fast-oscillatory functional networks. We performed graph theory analyses on resting-state magnetoencephalographic (MEG) signal, in addition to structural brain networks from DTI data, comparing degree, modularity and segregation coefficient across the brain of individuals with high versus average Gf scores. Our results show that high Gf individuals have stronger degree and lower segregation coefficient than average Gf participants in a significantly higher number of brain areas with regards to structural connectivity and to the slower frequency bands of functional connectivity. The opposite result was observed for higher-frequency (gamma) functional networks, with higher Gf individuals showing lower degree and higher segregation across the brain. We suggest that gamma oscillations in more intelligent individuals might support higher local processing in segregated subnetworks, while slower frequency bands would allow a more effective information transfer between brain subnetworks, and stronger information integration.

Data and model considerations for estimating time-varying functional connectivity in fMRI.

Functional connectivity (FC) in the brain has been shown to exhibit subtle but reliable modulations within a session. One way of estimating time-varying FC is by using state-based models that describe fMRI time series as temporal sequences of states, each with an associated, characteristic pattern of FC. However, the estimation of these models from data sometimes fails to capture changes in a meaningful way, such that the model estimation assigns entire sessions (or the largest part of them) to a single state, therefore failing to capture within-session state modulations effectively; we refer to this phenomenon as the model becoming static, or model stasis. Here, we aim to quantify how the nature of the data and the choice of model parameters affect the model's ability to detect temporal changes in FC using both simulated fMRI time courses and resting state fMRI data. We show that large between-subject FC differences can overwhelm subtler within-session modulations, causing the model to become static. Further, the choice of parcellation can also affect the model's ability to detect temporal changes. We finally show that the model often becomes static when the number of free parameters per state that need to be estimated is high and the number of observations available for this estimation is low in comparison. Based on these findings, we derive a set of practical recommendations for time-varying FC studies, in terms of preprocessing, parcellation and complexity of the model.

Rapid encoding of musical tones discovered in whole-brain connectivity.

Information encoding has received a wide neuroscientific attention, but the underlying rapid spatiotemporal brain dynamics remain largely unknown. Here, we investigated the rapid brain mechanisms for encoding of sounds forming a complex temporal sequence. Specifically, we used magnetoencephalography (MEG) to record the brain activity of 68 participants while they listened to a highly structured musical prelude. Functional connectivity analyses performed using phase synchronisation and graph theoretical measures showed a large network of brain areas recruited during encoding of sounds, comprising primary and secondary auditory cortices, frontal operculum, insula, hippocampus and basal ganglia. Moreover, our results highlighted the rapid transition of brain activity from primary auditory cortex to higher order association areas including insula and superior temporal pole within a whole-brain network, occurring during the first 220 ms of the encoding process. Further, we discovered that individual differences along cognitive abilities and musicianship modulated the degree centrality of the brain areas implicated in the encoding process. Indeed, participants with higher musical expertise presented a stronger centrality of superior temporal gyrus and insula, while individuals with high working memory abilities showed a stronger centrality of frontal operculum. In conclusion, our study revealed the rapid unfolding of brain network dynamics responsible for the encoding of sounds and their relationship with individual differences, showing a complex picture which extends beyond the well-known involvement of auditory areas. Indeed, our results expanded our understanding of the general mechanisms underlying auditory pattern encoding in the human brain.

Applying Spike-density component analysis for high-accuracy auditory event-related potentials in children.

OBJECTIVE: Overlapping neurophysiological signals are the main obstacle preventing from using cortical auditory event-related potentials (AEPs) in clinical settings. Children AEPs are particularly affected by this problem, as their cerebral cortex is still maturing. To overcome this problem, we applied a new version of Spike-density Component Analysis (SCA), an analysis method recently developed, to isolate with high accuracy the neural components of auditory responses of 8-year-old children. METHODS: Electroencephalography was used with 33 children to record AEPs to auditory stimuli varying in spectrotemporal features. Three different analysis approaches were adopted: the standard AEP analysis procedure, SCA with template-match (SCA-TM), and SCA with half-split average consistency (SCA-HSAC). RESULTS: SCA-HSAC most successfully allowed the extraction of AEPs for each child, revealing that the most consistent components were P1 and N2. An immature N1 component was also detected. CONCLUSION: Superior accuracy in isolating neural components at the individual level was demonstrated for SCA-HSAC over other SCA approaches even for children AEPs. SIGNIFICANCE: Reliable methods of extraction of neurophysiological signals at the individual level are crucial for the application of cortical AEPs for routine diagnostic exams in clinical settings both in children and adults.

Listeners with congenital amusia are sensitive to context uncertainty in melodic sequences.

In typical listeners, the perceptual salience of a surprising auditory event depends on the uncertainty of its context. For example, in melodies, pitch deviants are more easily detected and generate larger neural responses when the context is highly predictable than when it is less so. However, it is not known whether amusic listeners with abnormal pitch processing are sensitive to the degree of uncertainty of pitch sequences and, if so, whether they are to a different extent than typical non-musician listeners. To answer this question, we manipulated the uncertainty of short melodies while participants with and without congenital amusia underwent EEG recordings in a passive listening task. Uncertainty was manipulated by presenting melodies with different levels of complexity and familiarity, under the assumption that simpler and more familiar patterns would enhance pitch predictability. We recorded mismatch negativity (MMN) responses to pitch, intensity, timbre, location, and rhythm deviants as a measure of auditory surprise. In both participant groups, we observed reduced MMN amplitudes and longer peak latencies for all sound features with increasing levels of complexity, and putative familiarity effects only for intensity deviants. No significant group-by-complexity or group-by-familiarity interactions were detected. However, in contrast to previous studies, pitch MMN responses in amusics were disrupted in high complexity and unfamiliar melodies. The present results thus indicate that amusics are sensitive to the uncertainty of melodic sequences and that preattentive auditory change detection is greatly spared in this population across sound features and levels of predictability. However, our findings also hint at pitch-specific impairments in this population when uncertainty is high, thus suggesting that pitch processing under high uncertainty conditions requires an intact frontotemporal loop.

Brain predictive coding processes are associated to COMT gene Val158Met polymorphism.

Predicting events in the ever-changing environment is a fundamental survival function intrinsic to the physiology of sensory systems, whose efficiency varies among the population. Even though it is established that a major source of such variations is genetic heritage, there are no studies tracking down auditory predicting processes to genetic mutations. Thus, we examined the neurophysiological responses to deviant stimuli recorded with magnetoencephalography (MEG) in 108 healthy participants carrying different variants of Val158Met single-nucleotide polymorphism (SNP) within the catechol-O-methyltransferase (COMT) gene, responsible for the majority of catecholamines degradation in the prefrontal cortex. Our results showed significant amplitude enhancement of prediction error responses originating from the inferior frontal gyrus, superior and middle temporal cortices in heterozygous genotype carriers (Val/Met) vs homozygous (Val/Val and Met/Met) carriers. Integrating neurophysiology and genetics, this study shows how the neural mechanisms underlying optimal deviant detection vary according to the gene-determined cathecolamine levels in the brain.

Decomposing neural responses to melodic surprise in musicians and non-musicians: Evidence for a hierarchy of predictions in the auditory system.

Neural responses to auditory surprise are typically studied with highly unexpected, disruptive sounds. Consequently, little is known about auditory prediction in everyday contexts that are characterized by fine-grained, non-disruptive fluctuations of auditory surprise. To address this issue, we used IDyOM, a computational model of auditory expectation, to obtain continuous surprise estimates for a set of newly composed melodies. Our main goal was to assess whether the neural correlates of non-disruptive surprising sounds in a musical context are affected by musical expertise. Using magnetoencephalography (MEG), auditory responses were recorded from musicians and non-musicians while they listened to the melodies. Consistent with a previous study, the amplitude of the N1m component increased with higher levels of computationally estimated surprise. This effect, however, was not different between the two groups. Further analyses offered an explanation for this finding: Pitch interval size itself, rather than probabilistic prediction, was responsible for the modulation of the N1m, thus pointing to low-level sensory adaptation as the underlying mechanism. In turn, the formation of auditory regularities and proper probabilistic prediction were reflected in later components: The mismatch negativity (MMNm) and the P3am, respectively. Overall, our findings reveal a hierarchy of expectations in the auditory system and highlight the need to properly account for sensory adaptation in research addressing statistical learning.

Pessimistic outcome expectancy does not explain ambiguity aversion in decision-making under uncertainty.

When faced with a decision, most people like to know the odds and prefer to avoid ambiguity. It has been suggested that this aversion to ambiguity is linked to people's assumption of worst possible outcomes. We used two closely linked behavioural tasks in 78 healthy participants to investigate whether such pessimistic prior beliefs can explain ambiguity aversion. In the risk-taking task, participants had to decide whether or not they place a bet, while in the beliefs task, participants were asked what they believed would be the outcome. Unexpectedly, we found that in the beliefs task, participants were not overly pessimistic about the outcome in the ambiguity condition and in fact closer to optimal levels of decision-making than in the risk conditions. While individual differences in pessimism could explain outcome expectancy, they had no effect on ambiguity aversion. Consequently, ambiguity aversion is more likely caused by general caution than by expectation of negative outcomes despite pessimism-dependent subjective weighting of probabilities.

Discovery of key whole-brain transitions and dynamics during human wakefulness and non-REM sleep.

The modern understanding of sleep is based on the classification of sleep into stages defined by their electroencephalography (EEG) signatures, but the underlying brain dynamics remain unclear. Here we aimed to move significantly beyond the current state-of-the-art description of sleep, and in particular to characterise the spatiotemporal complexity of whole-brain networks and state transitions during sleep. In order to obtain the most unbiased estimate of how whole-brain network states evolve through the human sleep cycle, we used a Markovian data-driven analysis of continuous neuroimaging data from 57 healthy participants falling asleep during simultaneous functional magnetic resonance imaging (fMRI) and EEG. This Hidden Markov Model (HMM) facilitated discovery of the dynamic choreography between different whole-brain networks across the wake-non-REM sleep cycle. Notably, our results reveal key trajectories to switch within and between EEG-based sleep stages, while highlighting the heterogeneities of stage N1 sleep and wakefulness before and after sleep.

Auditory sensory memory and working memory skills: Association between frontal MMN and performance scores.

OBJECTIVE: Memory is the faculty responsible for encoding, storing and retrieving information, comprising several sub-systems such as sensory memory (SM) and working memory (WM). Some previous studies exclusively using clinical population revealed associations between these two memory systems. Here we aimed at investigating the relation between modality-general WM performance and auditory SM formation indexed by magnetic mismatch negativity (MMN) responses in a healthy population of young adults. METHODS: Using magnetoencephalography (MEG), we recorded MMN amplitudes to changes related to six acoustic features (pitch, timbre, location, intensity, slide, and rhythm) inserted in a 4-tone sequence in 86 adult participants who were watching a silent movie. After the MEG recordings, participants were administered the WM primary subtests (Spatial Span and Letter Number Sequencing) of Wechsler Memory Scale (WMS). RESULTS: We found significant correlations between frontal MMN amplitudes to intensity and slide deviants and WM performance. In case of intensity, the relation was revealed in all participants, while for slide only in individuals with a musical background. CONCLUSIONS: Automatic neural responses to auditory feature changes are increased in individuals with higher visual WM performance. SIGNIFICANCE: Conscious WM abilities might be linked to pre-attentive sensory-specific neural skills of prediction and short-term storage of environmental regularities.

Risk of depression enhances auditory Pitch discrimination in the brain as indexed by the mismatch negativity.

OBJECTIVE: Depression is a state of aversion to activity and low mood that affects behaviour, thoughts, feelings and sense of well-being. Moreover, the individual depression trait is associated with altered auditory cortex activation and appraisal of the affective content of sounds. METHODS: Mismatch negativity responses (MMNs) to acoustic feature changes (pitch, timbre, location, intensity, slide and rhythm) inserted in a musical sequence played in major or minor mode were recorded using magnetoencephalography (MEG) in 88 subclinical participants with depression risk. RESULTS: We found correlations between MMNs to slide and pitch and the level of depression risk reported by participants, indicating that higher MMNs correspond to higher risk of depression. Furthermore we found significantly higher MMN amplitudes to mistuned pitches within a major context compared to MMNs to pitch changes in a minor context. CONCLUSIONS: The brains of individuals with depression risk are more responsive to mistuned and fast pitch stimulus changes, even at a pre-attentive level. SIGNIFICANCE: Considering the altered appraisal of affective contents of sounds in depression and the relevance of spectral pitch features for those contents in music and speech, we propose that individuals with subclinical depression risk are more tuned to tracking sudden pitch changes.

Oxytocin improves synchronisation in leader-follower interaction.

The neuropeptide oxytocin has been shown to affect social interaction. Meanwhile, the underlying mechanism remains highly debated. Using an interpersonal finger-tapping paradigm, we investigated whether oxytocin affects the ability to synchronise with and adapt to the behaviour of others. Dyads received either oxytocin or a non-active placebo, intranasally. We show that in conditions where one dyad-member was tapping to another unresponsive dyad-member - i.e. one was following another who was leading/self-pacing - dyads given oxytocin were more synchronised than dyads given placebo. However, there was no effect when following a regular metronome or when both tappers were mutually adapting to each other. Furthermore, relative to their self-paced tapping partners, oxytocin followers were less variable than placebo followers. Our data suggests that oxytocin improves synchronisation to an unresponsive partner's behaviour through a reduction in tapping-variability. Hence, oxytocin may facilitate social interaction by enhancing sensorimotor predictions supporting interpersonal synchronisation. The study thus provides novel perspectives on how neurobiological processes relate to socio-psychological behaviour and contributes to the growing evidence that synchronisation and prediction are central to social cognition.

Cortical plasticity after cochlear implantation.

The most dramatic progress in the restoration of hearing takes place in the first months after cochlear implantation. To map the brain activity underlying this process, we used positron emission tomography at three time points: within 14 days, three months, and six months after switch-on. Fifteen recently implanted adult implant recipients listened to running speech or speech-like noise in four sequential PET sessions at each milestone. CI listeners with postlingual hearing loss showed differential activation of left superior temporal gyrus during speech and speech-like stimuli, unlike CI listeners with prelingual hearing loss. Furthermore, Broca's area was activated as an effect of time, but only in CI listeners with postlingual hearing loss. The study demonstrates that adaptation to the cochlear implant is highly related to the history of hearing loss. Speech processing in patients whose hearing loss occurred after the acquisition of language involves brain areas associated with speech comprehension, which is not the case for patients whose hearing loss occurred before the acquisition of language. Finally, the findings confirm the key role of Broca's area in restoration of speech perception, but only in individuals in whom Broca's area has been active prior to the loss of hearing.

It don't mean a thing... Keeping the rhythm during polyrhythmic tension, activates language areas (BA47).

Music is experienced and understood on the basis of foreground/background relationships created between actual music and the underlying meter. In contemporary styles of music so-called polyrhythmic, structures hence create tension between a counter pulse and the main pulse. This exerts a marked influence on the listener, particularly when the experience of the original meter is maintained during the counter pulse. We here demonstrate that Brodmann area 47, an area associated with higher processing of language, is activated bilaterally when musicians tap the main pulse in a polymetric context where the music emphasizes a counter meter. This suggests that the processing of metric elements of music relies on brain areas also involved in language comprehension. We propose that BA47 is involved in general neuronal processing of temporal coherence subserving both language and music.