Impaired Cortical Tracking of Speech in Children with Developmental Language Disorder.

In developmental language disorder (DLD), learning to comprehend and express oneself with spoken language is impaired, but the reason for this remains unknown. Using millisecond-scale magnetoencephalography recordings combined with machine learning models, we investigated whether the possible neural basis of this disruption lies in poor cortical tracking of speech. The stimuli were common spoken Finnish words (e.g., dog, car, hammer) and sounds with corresponding meanings (e.g., dog bark, car engine, hammering). In both children with DLD (10 boys and 7 girls) and typically developing (TD) control children (14 boys and 3 girls), aged 10-15 years, the cortical activation to spoken words was best modeled as time-locked to the unfolding speech input at ∼100 ms latency between sound and cortical activation. Amplitude envelope (amplitude changes) and spectrogram (detailed time-varying spectral content) of the spoken words, but not other sounds, were very successfully decoded based on time-locked brain responses in bilateral temporal areas; based on the cortical responses, the models could tell at ∼75-85% accuracy which of the two sounds had been presented to the participant. However, the cortical representation of the amplitude envelope information was poorer in children with DLD compared with TD children at longer latencies (at ∼200-300 ms lag). We interpret this effect as reflecting poorer retention of acoustic-phonetic information in short-term memory. This impaired tracking could potentially affect the processing and learning of words as well as continuous speech. The present results offer an explanation for the problems in language comprehension and acquisition in DLD.

Bayesian reduced rank regression models generalizable neural fingerprints that differentiate between individuals in magnetoencephalography data.

Recent magnetoencephalography (MEG) studies have reported that functional connectivity (FC) and power spectra can be used as neural fingerprints in differentiating individuals. Such studies have mainly used correlations between measurement sessions to distinguish individuals from each other. However, it has remained unclear whether such correlations might reflect a more generalizable principle of individually distinctive brain patterns. Here, we evaluated a machine-learning based approach, termed latent-noise Bayesian reduced rank regression (BRRR) as a means of modelling individual differences in the resting-state MEG data of the Human Connectome Project (HCP), using FC and power spectra as neural features. First, we verified that BRRR could model and reproduce the differences between metrics that correlation-based fingerprinting yields. We trained BRRR models to distinguish individuals based on data from one measurement and used the models to identify subsequent measurement sessions of those same individuals. The best performing BRRR models, using only 20 spatiospectral components, were able to identify subjects across measurement sessions with over 90% accuracy, approaching the highest correlation-based accuracies. Using cross-validation, we then determined whether that BRRR model could generalize to unseen subjects, successfully classifying the measurement sessions of novel individuals with over 80% accuracy. The results demonstrate that individual neurofunctional differences can be reliably extracted from MEG data with a low-dimensional predictive model and that the model is able to classify novel subjects.

Beta- and gamma-band cortico-cortical interactions support naturalistic reading of continuous text.

Large-scale integration of information across cortical structures, building on neural connectivity, has been proposed to be a key element in supporting human cognitive processing. In electrophysiological neuroimaging studies of reading, quantification of neural interactions has been limited to the level of isolated words or sentences due to artefacts induced by eye movements. Here, we combined magnetoencephalography recording with advanced artefact rejection tools to investigate both cortico-cortical coherence and directed neural interactions during naturalistic reading of full-page texts. Our results show that reading versus visual scanning of text was associated with wide-spread increases of cortico-cortical coherence in the beta and gamma bands. We further show that the reading task was linked to increased directed neural interactions compared to the scanning task across a sparse set of connections within a wide range of frequencies. Together, the results demonstrate that neural connectivity flexibly builds on different frequency bands to support continuous natural reading.

Human Sensorimotor Beta Event Characteristics and Aperiodic Signal Are Highly Heritable.

Individuals' phenotypes, including the brain's structure and function, are largely determined by genes and their interplay. The resting brain generates salient rhythmic patterns that can be characterized noninvasively using functional neuroimaging such as magnetoencephalography (MEG). One of these rhythms, the somatomotor (rolandic) beta rhythm, shows intermittent high amplitude "events" that predict behavior across tasks and species. Beta rhythm is altered in neurological disease. The aperiodic (1/f) signal present in electrophysiological recordings is also modulated by some neurological conditions and aging. Both sensorimotor beta and aperiodic signal could thus serve as biomarkers of sensorimotor function. Knowledge about the extent to which these brain functional measures are heritable could shed light on the mechanisms underlying their generation. We investigated the heritability and variability of human spontaneous sensorimotor beta rhythm events and aperiodic activity in 210 healthy male and female adult siblings' spontaneous MEG activity. The most heritable trait was the aperiodic 1/f signal, with a heritability of 0.87 in the right hemisphere. Time-resolved beta event amplitude parameters were also highly heritable, whereas the heritabilities for overall beta power, peak frequency, and measures of event duration remained nonsignificant. Human sensorimotor neural activity can thus be dissected into different components with variable heritability. We postulate that these differences partially reflect different underlying signal-generating mechanisms. The 1/f signal and beta event amplitude measures may depend more on fixed, anatomical parameters, whereas beta event duration and its modulation reflect dynamic characteristics, guiding their use as potential disease biomarkers.

Cortical time-course of evidence accumulation during semantic processing.

Our understanding of the surrounding world and communication with other people are tied to mental representations of concepts. In order for the brain to recognize an object, it must determine which concept to access based on information available from sensory inputs. In this study, we combine magnetoencephalography and machine learning to investigate how concepts are represented and accessed in the brain over time. Using brain responses from a silent picture naming task, we track the dynamics of visual and semantic information processing, and show that the brain gradually accumulates information on different levels before eventually reaching a plateau. The timing of this plateau point varies across individuals and feature models, indicating notable temporal variation in visual object recognition and semantic processing.

Semantic feature norms: a cross-method and cross-language comparison.

The ability to assign meaning to perceptual stimuli forms the basis of human behavior and the ability to use language. The meanings of things have primarily been probed using behavioral production norms and corpus-derived statistical methods. However, it is not known to what extent the collection method and the language being probed influence the resulting semantic feature vectors. In this study, we compare behavioral with corpus-based norms, across Finnish and English, using an all-to-all approach. To complete the set of norms required for this study, we present a new set of Finnish behavioral production norms, containing both abstract and concrete concepts. We found that all the norms provide largely similar information about the relationships of concrete objects and allow item-level mapping across norms sets. This validates the use of the corpus-derived norms which are easier to obtain than behavioral norms, which are labor-intensive to collect, for studies that do not depend on subtle differences in meaning between close semantic neighbors.

Sensor-level MEG combined with machine learning yields robust classification of mild traumatic brain injury patients.

OBJECTIVE: Diagnosis of mild traumatic brain injury (mTBI) is challenging despite its high incidence, due to the unspecificity and variety of symptoms and the frequent lack of structural imaging findings. There is a need for reliable and simple-to-use diagnostic tools that would be feasible across sites and patient populations. METHODS: We evaluated linear machine learning (ML) methods' ability to separate mTBI patients from healthy controls, based on their sensor-level magnetoencephalographic (MEG) power spectra in the subacute phase (<2 months) after a head trauma. We recorded resting-state MEG data from 25 patients and 25 age-sex matched controls and utilized a previously collected data set of 20 patients and 20 controls from a different site. The data sets were analyzed separately with three ML methods. RESULTS: The median classification accuracies varied between 80 and 95%, without significant differences between the applied ML methods or data sets. The classification accuracies were significantly higher with ML than with traditional sensor-level MEG analysis based on detecting pathological low-frequency activity. CONCLUSIONS: Easily applicable linear ML methods provide reliable and replicable classification of mTBI patients using sensor-level MEG data. SIGNIFICANCE: Power spectral estimates combined with ML can classify mTBI patients with high accuracy and have high promise for clinical use.

The relationship between electrophysiological and hemodynamic measures of neural activity varies across picture naming tasks: A multimodal magnetoencephalography-functional magnetic resonance imaging study.

Different neuroimaging methods can yield different views of task-dependent neural engagement. Studies examining the relationship between electromagnetic and hemodynamic measures have revealed correlated patterns across brain regions but the role of the applied stimulation or experimental tasks in these correlation patterns is still poorly understood. Here, we evaluated the across-tasks variability of MEG-fMRI relationship using data recorded during three distinct naming tasks (naming objects and actions from action images, and objects from object images), from the same set of participants. Our results demonstrate that the MEG-fMRI correlation pattern varies according to the performed task, and that this variability shows distinct spectral profiles across brain regions. Notably, analysis of the MEG data alone did not reveal modulations across the examined tasks in the time-frequency windows emerging from the MEG-fMRI correlation analysis. Our results suggest that the electromagnetic-hemodynamic correlation could serve as a more sensitive proxy for task-dependent neural engagement in cognitive tasks than isolated within-modality measures.

Attentional modulation of interhemispheric (a)symmetry in children with developmental language disorder.

The nature of auditory processing problems in children with developmental language disorder (DLD) is still poorly understood. Much research has been devoted to determining the extent to which DLD is associated with general auditory versus language-specific dysfunction. However, less emphasis has been given to the role of different task conditions in these dysfunctions. We explored whether children with DLD demonstrate atypical interhemispheric asymmetry during the auditory processing of speech and non-speech sounds and whether this interhemispheric balance is modulated by attention. Magnetoencephalography was used to record auditory evoked fields in 18 children (9 to 10 years old), 9 with DLD and 9 with language typical development, during active or passive listening to speech and non-speech sounds. A linear mixed model analysis revealed a bilateral effect of attention in both groups. Participants with DLD demonstrated atypical interhemispheric asymmetry, specifically in the later (185-600 ms) time window but only during the passive listening condition. During the active task, the DLD group did not differ from the typically developed children in terms of hemispheric balance of activation. Our results support the idea of an altered interhemispheric balance in passive auditory response properties in DLD. We further suggest that an active task condition, or top-down attention, can help to regain leftward lateralization, particularly in a later stage of activation. Our study highlights the highly dynamic and interhemispheric nature of auditory processing, which may contribute to the variability in reports of auditory language processing deficits in DLD.

Moving in Semantic Space in Prodromal and Very Early Alzheimer's Disease: An Item-Level Characterization of the Semantic Fluency Task.

The semantic fluency task is a widely used clinical tool in the diagnostic process of Alzheimer's disease. The task requires efficient mapping of the semantic space to produce as many items as possible within a semantic category. We examined whether healthy volunteers (n = 42) and patients with early Alzheimer's disease (24 diagnosed with amnestic Mild Cognitive Impairment and 18 with early Alzheimer's dementia) take advantage of and travel in the semantic space differently. With focus on the animal fluency task, we sought to emulate the detailed structure of the multidimensional semantic space by utilizing word2vec-method from the natural language processing domain. To render the resulting multidimensional semantic space visually comprehensible, we applied a dimensionality reduction algorithm (t-SNE), which enabled a straightforward division of the semantic space into sub-categories. Moving in semantic space was quantified with the number of items created, sub-categories visited, and switches and returns to these sub-categories. Multinomial logistic regression models were used to predict the diagnostic group with these independent variables. We found that returning to a sub-category provided additional information, besides the number of words produced in the task, to differentiate patients with Alzheimer's dementia from both amnestic Mild Cognitive Impairment patients and healthy controls. The results suggest that the frequency of returning to a sub-category may serve as an additional aid for clinicians in diagnosing early Alzheimer's disease. Moreover, our results imply that the combination of word2vec and subsequent t-SNE-visualization may offer a valuable tool for examining the semantic space and its sub-categories.

Removing ocular artifacts from magnetoencephalographic data on naturalistic reading of continuous texts.

Naturalistic reading paradigms and stimuli consisting of long continuous texts are essential for characterizing the cortical basis of reading. Due to the highly dynamic nature of the reading process, electrophysiological brain imaging methods with high spatial and temporal resolution, such as magnetoencephalography (MEG), are ideal for tracking them. However, as electrophysiological recordings are sensitive to electromagnetic artifacts, data recorded during naturalistic reading is confounded by ocular artifacts. In this study, we evaluate two different pipelines for removing ocular artifacts from MEG data collected during continuous, naturalistic reading, with the focus on saccades and blinks. Both pipeline alternatives are based on blind source separation methods but differ fundamentally in their approach. The first alternative is a multi-part process, in which saccades are first extracted by applying Second-Order Blind Identification (SOBI) and, subsequently, FastICA is used to extract blinks. The other alternative uses a single powerful method, Adaptive Mixture ICA (AMICA), to remove all artifact types at once. The pipelines were tested, and their effects compared on MEG data recorded from 13 subjects in a naturalistic reading task where the subjects read texts with the length of multiple pages. Both pipelines performed well, extracting the artifacts in a single component per artifact type in most subjects. Signal power was reduced across the whole cortex in all studied frequency bands from 1 to 90 Hz, but especially in the frontal cortex and temporal pole. The results were largely similar for the two pipelines, with the exception that SOBI-FastICA reduced signal in the right frontal cortex in all studied frequency bands more than AMICA. However, there was considerable interindividual variation in the effects of the pipelines. As a holistic conclusion, we choose to recommend AMICA for removing artifacts from MEG data on naturalistic reading but note that the SOBI-FastICA pipeline has also various favorable characteristics.

Selective auditory attention within naturalistic scenes modulates reactivity to speech sounds.

Rapid recognition and categorization of sounds are essential for humans and animals alike, both for understanding and reacting to our surroundings and for daily communication and social interaction. For humans, perception of speech sounds is of crucial importance. In real life, this task is complicated by the presence of a multitude of meaningful non-speech sounds. The present behavioural, magnetoencephalography (MEG) and functional magnetic resonance imaging (fMRI) study was set out to address how attention to speech versus attention to natural non-speech sounds within complex auditory scenes influences cortical processing. The stimuli were superimpositions of spoken words and environmental sounds, with parametric variation of the speech-to-environmental sound intensity ratio. The participants' task was to detect a repetition in either the speech or the environmental sound. We found that specifically when participants attended to speech within the superimposed stimuli, higher speech-to-environmental sound ratios resulted in shorter sustained MEG responses and stronger BOLD fMRI signals especially in the left supratemporal auditory cortex and in improved behavioural performance. No such effects of speech-to-environmental sound ratio were observed when participants attended to the environmental sound part within the exact same stimuli. These findings suggest stronger saliency of speech compared with other meaningful sounds during processing of natural auditory scenes, likely linked to speech-specific top-down and bottom-up mechanisms activated during speech perception that are needed for tracking speech in real-life-like auditory environments.

The neural representation of abstract words may arise through grounding word meaning in language itself.

In order to describe how humans represent meaning in the brain, one must be able to account for not just concrete words but, critically, also abstract words, which lack a physical referent. Hebbian formalism and optimization are basic principles of brain function, and they provide an appealing approach for modeling word meanings based on word co-occurrences. We provide proof of concept that a statistical model of the semantic space can account for neural representations of both concrete and abstract words, using MEG. Here, we built a statistical model using word embeddings extracted from a text corpus. This statistical model was used to train a machine learning algorithm to successfully decode the MEG signals evoked by written words. In the model, word abstractness emerged from the statistical regularities of the language environment. Representational similarity analysis further showed that this salient property of the model co-varies, at 280-420 ms after visual word presentation, with activity in regions that have been previously linked with processing of abstract words, namely the left-hemisphere frontal, anterior temporal and superior parietal cortex. In light of these results, we propose that the neural encoding of word meanings can arise through statistical regularities, that is, through grounding in language itself.

Picture naming yields highly consistent cortical activation patterns: Test-retest reliability of magnetoencephalography recordings.

Reliable paradigms and imaging measures of individual-level brain activity are paramount when reaching from group-level research studies to clinical assessment of individual patients. Magnetoencephalography (MEG) provides a direct, non-invasive measure of cortical processing with high spatiotemporal accuracy, and is thus well suited for assessment of functional brain damage in patients with language difficulties. This MEG study aimed to identify, in a delayed picture naming paradigm, source-localized evoked activity and modulations of cortical oscillations that show high test-retest reliability across measurement days in healthy individuals, demonstrating their applicability in clinical settings. For patients with a language disorder picture naming can be a challenging task. Therefore, we also determined whether a semantic judgment task ('Is this item living?') with a spoken response ("yes"/"no") would suffice to induce comparably consistent activity within brain regions related to language production. The MEG data was collected from 19 healthy participants on two separate days. In picture naming, evoked activity was consistent across measurement days (intraclass correlation coefficient (ICC)>0.4) in the left frontal (400-800 ms after image onset), sensorimotor (200-800 ms), parietal (200-600 ms), temporal (200-800 ms), occipital (400-800 ms) and cingulate (600-800 ms) regions, as well as the right temporal (600-800 ms) region. In the semantic judgment task, consistent evoked activity was spatially more limited, occurring in the left temporal (200-800 ms), sensorimotor (400-800 ms), occipital (400-600 ms) and subparietal (600-800 ms) regions, and the right supramarginal cortex (600-800 ms). The delayed naming task showed typical beta oscillatory suppression in premotor and sensorimotor regions (800-1200 ms) but other consistent modulations of oscillatory activity were mostly observed in posterior cortical regions that have not typically been associated with language processing. The high test-retest consistency of MEG evoked activity in the picture naming task testifies to its applicability in clinical evaluations of language function, as well as in longitudinal MEG studies of language production in clinical and healthy populations.

Issues and recommendations from the OHBM COBIDAS MEEG committee for reproducible EEG and MEG research.

The Organization for Human Brain Mapping (OHBM) has been active in advocating for the instantiation of best practices in neuroimaging data acquisition, analysis, reporting and sharing of both data and analysis code to deal with issues in science related to reproducibility and replicability. Here we summarize recommendations for such practices in magnetoencephalographic (MEG) and electroencephalographic (EEG) research, recently developed by the OHBM neuroimaging community known by the abbreviated name of COBIDAS MEEG. We discuss the rationale for the guidelines and their general content, which encompass many topics under active discussion in the field. We highlight future opportunities and challenges to maximizing the sharing and exploitation of MEG and EEG data, and we also discuss how this 'living' set of guidelines will evolve to continually address new developments in neurophysiological assessment methods and multimodal integration of neurophysiological data with other data types.

Post-hoc modification of linear models: Combining machine learning with domain information to make solid inferences from noisy data.

Linear machine learning models "learn" a data transformation by being exposed to examples of input with the desired output, forming the basis for a variety of powerful techniques for analyzing neuroimaging data. However, their ability to learn the desired transformation is limited by the quality and size of the example dataset, which in neuroimaging studies is often notoriously noisy and small. In these cases, it is desirable to fine-tune the learned linear model using domain information beyond the example dataset. To this end, we present a framework that decomposes the weight matrix of a fitted linear model into three subcomponents: the data covariance, the identified signal of interest, and a normalizer. Inspecting these subcomponents in isolation provides an intuitive way to inspect the inner workings of the model and assess its strengths and weaknesses. Furthermore, the three subcomponents may be altered, which provides a straightforward way to inject prior information and impose additional constraints. We refer to this process as "post-hoc modification" of a model and demonstrate how it can be used to achieve precise control over which aspects of the model are fitted to the data through machine learning and which are determined through domain information. As an example use case, we decode the associative strength between words from electroencephalography (EEG) reading data. Our results show how the decoding accuracy of two example linear models (ridge regression and logistic regression) can be boosted by incorporating information about the spatio-temporal nature of the data, domain information about the N400 evoked potential and data from other participants.

Task-Modulated Corticocortical Synchrony in the Cognitive-Motor Network Supporting Handwriting.

Both motor and cognitive aspects of behavior depend on dynamic, accurately timed neural processes in large-scale brain networks. Here, we studied synchronous interplay between cortical regions during production of cognitive-motor sequences in humans. Specifically, variants of handwriting that differed in motor variability, linguistic content, and memorization of movement cues were contrasted to unveil functional sensitivity of corticocortical connections. Data-driven magnetoencephalography mapping (n = 10) uncovered modulation of mostly left-hemispheric corticocortical interactions, as quantified by relative changes in phase synchronization. At low frequencies (~2-13 Hz), enhanced frontoparietal synchrony was related to regular handwriting, whereas premotor cortical regions synchronized for simple loop production and temporo-occipital areas for a writing task substituting normal script with loop patterns. At the beta-to-gamma band (~13-45 Hz), enhanced synchrony was observed for regular handwriting in the central and frontoparietal regions, including connections between the sensorimotor and supplementary motor cortices and between the parietal and dorsal premotor/precentral cortices. Interpreted within a modular framework, these modulations of synchrony mainly highlighted interactions of the putative pericentral subsystem of hand coordination and the frontoparietal subsystem mediating working memory operations. As part of cortical dynamics, interregional phase synchrony varies depending on task demands in production of cognitive-motor sequences.

Statistical models of morphology predict eye-tracking measures during visual word recognition.

We studied how statistical models of morphology that are built on different kinds of representational units, i.e., models emphasizing either holistic units or decomposition, perform in predicting human word recognition. More specifically, we studied the predictive power of such models at early vs. late stages of word recognition by using eye-tracking during two tasks. The tasks included a standard lexical decision task and a word recognition task that assumedly places less emphasis on postlexical reanalysis and decision processes. The lexical decision results showed good performance of Morfessor models based on the Minimum Description Length optimization principle. Models which segment words at some morpheme boundaries and keep other boundaries unsegmented performed well both at early and late stages of word recognition, supporting dual- or multiple-route cognitive models of morphological processing. Statistical models based on full forms fared better in late than early measures. The results of the second, multi-word recognition task showed that early and late stages of processing often involve accessing morphological constituents, with the exception of short complex words. Late stages of word recognition additionally involve predicting upcoming morphemes on the basis of previous ones in multimorphemic words. The statistical models based fully on whole words did not fare well in this task. Thus, we assume that the good performance of such models in global measures such as gaze durations or reaction times in lexical decision largely stems from postlexical reanalysis or decision processes. This finding highlights the importance of considering task demands in the study of morphological processing.

Dynamics of brain activation during learning of syllable-symbol paired associations.

Initial stages of reading acquisition require the learning of letter and speech sound combinations. While the long-term effects of audio-visual learning are rather well studied, relatively little is known about the short-term learning effects at the brain level. Here we examined the cortical dynamics of short-term learning using magnetoencephalography (MEG) and electroencephalography (EEG) in two experiments that respectively addressed active and passive learning of the association between shown symbols and heard syllables. In experiment 1, learning was based on feedback provided after each trial. The learning of the audio-visual associations was contrasted with items for which the feedback was meaningless. In experiment 2, learning was based on statistical learning through passive exposure to audio-visual stimuli that were consistently presented with each other and contrasted with audio-visual stimuli that were randomly paired with each other. After 5-10 min of training and exposure, learning-related changes emerged in neural activation around 200 and 350 ms in the two experiments. The MEG results showed activity changes at 350 ms in caudal middle frontal cortex and posterior superior temporal sulcus, and at 500 ms in temporo-occipital cortex. Changes in brain activity coincided with a decrease in reaction times and an increase in accuracy scores. Changes in EEG activity were observed starting at the auditory P2 response followed by later changes after 300 ms. The results show that the short-term learning effects emerge rapidly (manifesting in later stages of audio-visual integration processes) and that these effects are modulated by selective attention processes.

Children show hemispheric differences in the basic auditory response properties.

Auditory cortex in each hemisphere shows preference to sounds from the opposite hemifield in the auditory space. Besides this contralateral dominance, the auditory cortex shows functional and structural lateralization, presumably influencing the features of subsequent auditory processing. Children have been shown to differ from adults in the hemispheric balance of activation in higher-order auditory based tasks. We studied, first, whether the contralateral dominance can be detected in 7- to 8-year-old children and, second, whether the response properties of auditory cortex in children differ between hemispheres. Magnetoencephalography (MEG) responses to simple tones revealed adult-like contralateral preference that was, however, extended in time in children. Moreover, we found stronger emphasis towards mature response properties in the right than left hemisphere, pointing to faster maturation of the right-hemisphere auditory cortex. The activation strength of the child-typical prolonged response was significantly decreased with age, within the narrow age-range of the studied child population. Our results demonstrate that although the spatial sensitivity to the opposite hemifield has emerged by 7 years of age, the population-level neurophysiological response shows salient immature features, manifested particularly in the left hemisphere. The observed functional differences between hemispheres may influence higher-level processing stages, for example, in language function.