Imaging Neuroscience

Task-based precision mapping has become a promising technique in functional MRI (fMRI) to robustly characterize and map an individuals unique activity patterns. These experiments consist of acquiring extensive imaging data in one participant, ultimately improving the sensitivity and specificity of individual-specific functional localization. Despite its advantages, studies have primarily focused on understanding individual-specific cortical activation, preventing a holistic view of a systems-level functional response, and to date, best approaches for the statistical analysis of controlled task-based, densely sampled, whole-brain data have not yet been fully established. Therefore, in this study, we collected whole-brain (i.e. covering cortex, cerebellum, and brainstem) multi-echo densely sampled data of the auditory system, a system with major subcortical components, and evaluated activation sensitivity as well as activation stability across data subsets of commonly-used whole-brain and region-specific inference testing approaches. The whole-brain approaches involved standard voxel-level and cluster-level inference schemes with varying statistical thresholds and a non-parametric permutation inference approach. The region-specific approaches involved an exploratory top % t-statistics methods and non-parametric permutation inference approaches. We found that a whole-brain voxel-level approach with a false discovery rate (FDR) correction (p<0.05) presented highest sensitivity across regions and subjects as well as most consistent detection of expected auditory regions, even with lower scan duration. In addition, we found that a region-specific top % t-statistic approach may be a useful exploratory functional localization tool and a complementary method to standard inference testing approaches.

Optically pumped magnetometer magnetoencephalography (OPM-MEG) offers a wearable, movement-tolerant alternative to conventional cryogenic MEG, placing sensors closer to the scalp and, in principle, improving sensitivity to deep sources. This is advantageous for examining subcortical structures that are affected by ageing, disorders and disease, such as the hippocampus. However, it remains unclear whether well-established activity (such as the attenuation of theta oscillations during the imagination of novel scenes) can be recovered from the medial temporal lobe (MTL) with OPM-MEG, and whether an individual structural MRI is required. Here, fifteen adults completed a scene imagination task. Initially we applied a 12-parameter template warping coregistration pipeline to the full sample. Following source reconstruction, we recovered the expected attenuation of theta (4-8 Hz) power during scene imagination compared to a counting baseline, with a significant cluster of activity peaking in the left parahippocampal gyrus. The clusters centre of mass was localised to the left hippocampus (t = -3.55, p = 0.048, whole-brain FWE-corrected) and was mostly confined to the left medial temporal lobe. We further supported our findings by using an individual T1-weighted MRI reconstruction pipeline in six participants who had these scans available. The two approaches produced similar whole-brain topographies and localised the peak MTL theta effect to left hippocampus, with temporal-lobe conjunction centroids 3-mm apart. These findings provide evidence that the theta attenuation of the scene construction network can be recovered at the group level with OPM-MEG, without an individual MRI.

Portable low-field MRI systems are a promising complement to conventional high-field systems, enabling broader access to MRI. However, correspondence in cortical thickness estimates between low- and high-field MRI in young people remains limited despite its importance for neurodevelopment and psychopathology. To evaluate how multiple low-field image processing approaches improve cortical thickness correspondence with high-field MRI in a large sample of young individuals, we collected ultra-low-field (64mT) and high-field (3T) MRI data from a community sample of young people. We applied deep learning-based image processing approaches (SynthSR v1.0, SynthSR v2.0, recon-all-clinical, and recon-any) to low-field data acquired across multiple sequences (T1- and T2-weighted) and orientations (axial, coronal, sagittal, and multi-orientation), with and without resampling and/or co-registration. We assessed global, lobar, and regional cortical thickness correspondence with 3T MRI measures using Pearson and intraclass correlations. We compared pipelines using Steigers Z-tests and Fishers Z-tests. A total of 150 individuals (mean age, 18.63{+/-}5.07; 80 female) were included. We observed the highest global correspondence with recon-all-clinical applied to coronal T1-weighted images (r=0.40, pFDR=2.6e-05). At the lobar and regional levels, multi-orientation T2-weighted images processed with recon-all-clinical showed the highest correspondence across the greatest number of regions (4/12 lobes; 13/68 regions). The highest correspondence and largest improvements were in frontal, cingulate, and temporal regions, including the right pars triangularis (r=0.52, pFDR=4.78e-11; Z=4.78, pFDR=4.25e-06), right caudal anterior cingulate (r=0.47, pFDR=3.83e-09; Z=5.46, pFDR=1.32e-07), and left parahippocampal (r=0.58, pFDR=2.98e-14; Z=5.17, pFDR=6.01e-07). We observed significantly improved cortical thickness correspondence in low-field MRI in young people. The recon-all-clinical pipeline yielded moderate correspondence, particularly in frontal, cingulate, and temporal regions. Our results highlight the potential of low-field MRI as an affordable and scalable approach for assessing cortical thickness in young people.

Individualized resting-state functional magnetic resonance imaging (rs-fMRI) is increasingly used to guide neuromodulation target selection. However, clinical scans are often short and noisy, and standard pipelines for functional network identification do not provide information about confidence of network assignment. With limited data, unstable network assignments can misdirect stimulation toward off-target regions, making it critical to know which assignments can be trusted. We developed Precision Confidence Mapping (PCM), a bootstrap-based framework that makes this uncertainty explicit and actionable. PCM repeatedly resamples the time series and reruns network detection to estimate how consistently each vertex is assigned to a given network. The resulting confidence maps can be thresholded to exclude less stable regions. We evaluated PCM across scan durations from 5 to 70 minutes using positive predictive value (PPV) as the primary measure of network-assignment precision. PPV quantified the proportion of vertices assigned to a network that received the same label in an independent within-subject 70 minute reference map. Confidence thresholding markedly improved PPV across functional networks, with the largest gains for short scan durations. Compared with standard network assignment, PCM significantly increased agreement with this independent reference. Within-subject agreement remained greater than between-subject agreement, indicating that thresholding preserved individual-specific network topography. These precision gains came with modest reductions in reference-network coverage, particularly at shorter scan durations. This tradeoff may be acceptable for neuromodulation applications that prioritize minimizing off-network assignments. By adding a reliability layer to individualized mapping, PCM supports more cautious and precise neuromodulation targeting under real-world clinical scan constraints.

Neurological and neuropsychiatric conditions affect 43% of the global population, many shaped by modifiable lifestyle exposures, yet their relationship to cortical haemodynamics is poorly characterised. The dorsolateral prefrontal cortex (dlPFC) is a particularly tractable target: it underpins executive function, is disrupted across neuropsychiatric and age-related conditions, and lies on the cortical surface, within reach of scalable, wearable-grade optical neuroimaging. We present LUCID, a longitudinal study of 92 healthy adults combining consumer wearable sleep and physical activity metrics with task-evoked dlPFC haemodynamics, measured by time-domain functional near-infrared spectroscopy (TD-fNIRS). Log-transformed peak dlPFC activation was negatively associated with reaction time (RT) across the 2N-Back and Stroop tasks and both hemispheres (r = -0.37 to -0.53), greater activation accompanying faster responses, consistent with a capacity/recruitment account. Activation showed moderate test-retest reliability (intraclass correlation coefficient, ICC = 0.56-0.71), with between-person variance exceeding within-person fluctuation, indicating stable individual differences. Demographic and lifestyle features incrementally predicted activation, with age the strongest predictor and modest contributions from sleep and physical activity. These findings establish TD-fNIRS dlPFC activation as a longitudinally stable, behaviourally relevant functional neural marker for scalable tracking of modifiable risk.

Dynamic functional connectivity (dFC) -- the time-varying re-configuration of brain network interactions -- has become a widely adopted method for studying how neural dynamics reflect ongoing cognition. Yet a fundamental question remains unresolved: can dFC reliably track when a person is cognitively engaged, and if not, why does it fail? Here, we address this question through a large-scale benchmark of seven widely used dFC methods, evaluating how well each predicts task presence across 16 fMRI datasets encompassing over 1,500 participants and 28 distinct experimental settings, complemented by realistic simulated data. Across experimental data, dFC-based tracking of cognitive engagement was unreliable in many cases: most method-experiment combinations performed near chance, and no single method succeeded across all contexts. This failure, however, was not uniform. Both experimental and simulated data showed that decoding performance varied systematically with three interacting factors -- experimental design, data quality, and the choice of dFC method -- rather than depending on dFC features alone. Critically, we identify specific experimental design conditions associated with more reliable tracking: paradigms with longer, more regular task blocks and fewer task-rest transitions were substantially more decodable, while data quality independently influenced performance across methods. These findings offer actionable principles for when dFC can -- and cannot -- be expected to serve as a reliable marker of underlying cognitive states.

Large-scale neuroimaging datasets are increasingly used to map relationships between brain structure, function, and behavior across the human lifespan. Routinely, analyses exclude participants who moved too much during imaging. While this decision is framed as quality control, it is increasingly recognized that head motion is not randomly distributed across individuals within a study, and so motion-based exclusion may preferentially remove people with particular characteristics relevant to the scientific goals of the study. Here we survey head motion and how it relates to participant characteristics across six large, publicly available datasets spanning nearly the entire human lifespan, namely the Human Connectome Project (HCP) Young Adult, HCP in Development, HCP in Aging, Adolescent Brain Cognitive DevelopmentSM Study, UK Biobank, and Spatial Topology project. These six datasets comprise more than 50,000 unique participants and 300,000 scans. We further benchmark our findings against motion distributions aggregated by MRIQC across more than 1.5 million scans. Under commonly applied strict exclusion thresholds, large fractions of participants would be removed (exceeding 80 % in the UK Biobank task data), and these removals were demographically structured, disproportionately excluding younger and older participants, those with higher BMI, and those with motion-associated clinical conditions. Respiratory pseudo-motion inflated estimates of head motion in adult cohorts, and applying notch filtering to remove respiratory frequencies from these estimates meaningfully reduced exclusion rates. Exclusion also carried downstream consequences. Strict thresholds reduced statistical power, inflated study costs, and altered the apparent predictability of behavioral phenotypes by removing a non-random, behaviorally distinct subgroup. These findings demonstrate that motion exclusion thresholds are not neutral quality-control decisions but structured selection mechanisms that reshape the composition of neuroimaging samples. We recommend that studies report the demographic characteristics of excluded participants, prefer data-driven censoring methods over fixed motion cutoffs, and clarify the target population while considering appropriate weighting techniques.

The human brain undergoes rapid developmental changes through early life, underpinning the emergence of function but also marking a period of vulnerability to a range of neurodevelopmental disorders. With dynamic changes to brain size, morphology, and imaging contrast, consistent and accurate computational neuroanatomy remains a challenge. Deep learning tools for segmentation, like SynthSeg, offer robustness to heterogeneously acquired MRI contrast but remain unproven in early development. Here, we aggregated a large cohort (26k) of MRI scans spanning infant to adult development, and evaluated SynthSeg performance. Automated quality control scores, visual inspection, and spatial overlap with expert-segmented MRI scans revealed poor quality output segmentations during development. In the infant period only 36% of scans (1094/3069) passed automated QC. Rescaling infant scans to adult brain sizes significantly improved spatial overlap, and cropping scans to match adult fields of view retrieved automated quality control. Evaluation of the SynthSeg rescale + crop pipeline demonstrated visible and quantitative improvements in segmentation throughout infancy and childhood. There were marked increases in successful segmentations in infant scans, with 91% of scans now passing QC (2803/3069). These findings facilitate computational analysis of typical and disrupted neurodevelopment and should be considered when training the next generation of computational tools.

Functional near-infrared spectroscopy (fNIRS) and functional magnetic resonance imaging (fMRI) both rely on the phenomenon of neurovascular coupling (NVC) to probe brain activity through their sensitivity to cerebral blood oxygenation. However, the relationship between fNIRS chromophores (oxy- and deoxyhaemoglobin, HbO and HbR), and fMRI (Blood Oxygen Level Dependent and Arterial Spin Labeling, BOLD and ASL) measurements, and whether this relationship remains consistent across subjects and physiological conditions, has only been partially characterised.. We acquired concurrent continuous-wave fNIRS and gradient-echo (GE) and spin-echo (SE) BOLD-ASL fMRI in healthy adults (n = 10) during visual stimulation. By applying calibrated fMRI methodology, we examined the relationships between fNIRS-derived haemoglobin modulations and fMRI-derived modulations in macrovascular (GE-) and microvascular (SE-) BOLD signals, cerebral blood flow (CBF), and oxygen metabolism (CMRO2). Group-level results showed strong temporal cross-modal agreement, with HbO and HbR tightly mirroring all fMRI signal time-courses (|r| > 0.8). A quantitative analysis of trial-by-trial modulations revealed distinct state-dependent behaviours: HbO maintained a stable relationship with the fMRI-derived metrics across conditions, whereas cross-modal relationships between HbR and fMRI-derived metrics substantially strengthened at higher flow-metabolism coupling (FMC), the ratio of CBF to CMRO2 change, an index of the strength of NVC. Both HbO and HbR were more strongly associated with GE-BOLD than with SE-BOLD. These findings provide a rigorous physiological grounding for fNIRS signal interpretation, demonstrating its utility as a surrogate marker for specific haemodynamic and metabolic parameters.

Optically pumped magnetometers (OPMs) can be used for magnetoencephalography (MEG) with equivalent or improved signal to noise ratio, relative to cryogenic MEG, when sensors are placed close to the scalp. OPM-based MEG can also be used in mobile contexts if sensors are placed in lightweight, wearable arrays. Individually tailored, rigid helmets known as scannercasts are currently the only method capable of achieving on-scalp, mobile recordings with high precision. However, these scannercasts are expensive to produce, require structural imaging in advance of the experiment, and can incur lengthy downtime while sensors are transferred between scannercasts. Here, we introduce a solution to these challenges that retains the advantages of scannercasts. We provide detailed steps for constructing a modular, cap-based design, suitable for all head sizes. Using simulations, we compare the leadfield power of this array against an idealised array and a commercially available mobile solution. We then validate our proposed solution empirically, in five participants, and provide a complete data preparation and analysis pipeline. Our design expands the accessibility of OPM-based MEG, and increases participant throughput to levels comparable to other imaging modalities. Crucially, it removes the trade-off between signal quality, mobility and practicality, promoting the unique potential of OPM-based MEG as a tool for studying naturalistic behaviour, and clinical assessment with high precision.

Recent advances in fast fMRI now enable whole-brain imaging with a TR of [&le;]1 s, which has helped to rekindle interest in characterizing the blood oxygen level dependent hemodynamic response function (BOLD HRF). Recent studies in the visual system have found intra-areal differences in temporal response characteristics, as well as HRFs that were faster and narrower than predicted by standard models. The auditory system presents a unique challenge, in that neuronal populations must operate across timescales of microseconds to minutes, and the surface of auditory cortex in particular is intricately and heavily vascularized. Here, we used fast fMRI to characterise voxelwise auditory HRFs evoked by short naturalistic sounds, assessing HRF reproducibility and variability across sessions, participants, and independent datasets. Across two studies at 3 T, participants passively listened to short environmental sounds while fMRI and quantitative MRI data were acquired with a 1s TR. Voxelwise HRFs were estimated via novel cross-session alignment routine, and exclusion of large vascular contributions. We identified a diverse set of hemodynamically plausible response shapes, which were not consistently captured by standard HRF approaches. These responses were reproducible within participants across sessions and robust across two independent acquisitions. Using data-driven gamma models, we achieved stable estimates with relatively few runs, particularly in auditory temporal regions. Within auditory cortex, we observed reproducible spatial gradients in response timing and shape, with faster and higher magnitude responses in medial regions, and slower and lower magnitude responses laterally. Together, these findings demonstrate that auditory HRFs are diverse, reliable, and regionally specific, and highlight the value of fast fMRI and data-driven modelling for advancing interpretation of fMRI data.

Introduction: Structural neuroimaging relies on T1-weighted (T1w) magnetic resonance imaging (MRI) for brain morphometry, yet at 7 Tesla (7 T) transmit field (B1+) inhomogeneity remains a major source of bias. Although Magnetization Prepared 2 Rapid Acquisition Gradient Echoes (MP2RAGE) improves the tissue contrast, residual B1+ effects may persist and may be exacerbated in aging or clinical populations, where anatomical and physiological factors further challenge image quality and preprocessing. The impact of B1+ inhomogeneity on automated quality assessment and morphometric statistical inference remains insufficiently understood. Methods: Submillimeter 7 T MP2RAGE brain acquisitions from carriers of a mitochondrial gene mutation (m.3243A>G) and controls were retrieved from previous studies. Image quality before and after B1+ inhomogeneity correction was assessed by multiple automated pipelines. Case-control morphometric studies, including regional volume and mean cortical thickness, were analyzed in both registration based and deep learning based segmentation frameworks. Changes in image quality metrics (IQMs) and morphometric statistical significance were evaluated to determine the impact of B1+ inhomogeneity correction. Results: Overall image quality rating and metrics sensitive to intensity non-uniformity and topological integrity consistently improved after B1+ inhomogeneity correction. However, its impact on morphometric statistical inferences was strongly method-dependent. Some pipelines showed redistribution of significant regions, whereas others predominantly demonstrated increased effects in sensitivity. Across methods, B1+ inhomogeneity correction altered the findings of morphometric analyses, particularly in cortical regions. Conclusion: Residual B1+ inhomogeneity at 7 T substantially influences both image quality control and morphometric evaluations. Current automated quality control approaches can hardly capture these effects reliably. B1+ inhomogeneity correction will not only improve intensity uniformity, but also change sensitivity of morphometric statistical inferences. To establish reliable morphometric biomarkers at UHF strengths, explicit B1+ correction and customized preprocessing are practically necessary and highly recommended.

Transcranial magnetic stimulation (TMS) is commonly dosed by setting stimulation intensity as a fixed percentage of the resting motor threshold (RMT), although a motor-derived intensity may not produce comparable neural recruitment across non-motor targets. We present TIDE (Tractography-Informed Dose Estimation), an open-source, SimNIBS-based pipeline designed to derive individualised stimulation intensities for non-motor white-matter targets. TIDE combines individual RMT measurements, finite-element electric-field modelling and diffusion MRI tractography to rescale the stimulation intensity according to the geometry and stimulation efficiency of the pathway of interest. Specifically, it computes the activating function along subject-specific streamlines and estimates the stimulator output, expressed as a percentage of maximum stimulator output, required for the target pathway to reach the activation level produced in the corticospinal tract at RMT. In an independent dataset of 19 participants, in which stimulation had been dosed conventionally as a fixed percentage of RMT, the relative difference between delivered and TIDE-estimated intensity was associated with the magnitude of TMS-induced behavioural effects at two frontal aslant tract (FAT) stimulation sites, while the delivered intensity alone was not. TIDE therefore extends conventional E-field dosing from cortical field magnitude to subject-specific pathway geometry, providing a method to move beyond the assumption of homogeneous pathway engagement while accounting for inter-individual variability in pathway-specific stimulation efficiency.

Introduction: A new method that corrects for time-dependent bias in standardized-uptake value ratios (SUVRs) was adapted and optimized for [11C]PiB (PiB) amyloid-beta (A{beta}) PET, across low-to-high A{beta} loads, relying only on PET data collected during the SUVR time-window. This modeling approach was evaluated in cross-sectional and longitudinal cohorts for earlier and shorter SUVR time-windows (30-45 min, 45-60 min) than commonly applied, to enable higher throughput imaging. Methods: The SUVR correction (SUVRc) approach was optimized and tested on separate cross-sectional (n=88), and longitudinal (36 participants, two time-points, 72 images) cohorts from the Harvard Aging Brain Study. The cross-sectional cohort spanned low, intermediate and high levels of cortical A{beta} pathology and the longitudinal images included two cohorts with low (5-10%) and high levels (~40%) of A{beta} change. SUVR and SUVRc were compared against SRTM DVR (0-60 min) to quantify A{beta} burden through Pearson's and Lin's correlations, difference plots and longitudinal change. Results: The mean regional bias in PiB SUVR (5-15%, depending on time-window and A{beta} burden) was significantly reduced to < 3% by SUVRc (corrected p < 0.05) in the cross-sectional cohorts for all time-windows, along with reductions in bias variability. SUVRc also showed higher Pearson's correlation (r) and Lin's concordance (LCC) with DVR across time-windows (r=0.98, LCC=0.99 at 30-45 min and 45-60 min) compared to uncorrected SUVR (r=0.96, LCC=0.95 at 30-45 min, r=0.97, LCC=0.92 at 45-60 min). Bland-Altman plots confirmed better agreement between SUVRc and DVR (mean bias at 30-45 min: 0.02 for SUVRc, 0.10 for SUVR; mean bias at 45-60 min: 0.01 for SUVRc, 0.17 for SUVR). Longitudinal DVR changes were more accurately represented by SUVRc, compared to uncorrected SUVR. Conclusions: SUVRc for [11C]PiB PET enables more accurate quantification of A{beta} burden than SUVR in cross-sectional and longitudinal studies (relative to SRTM DVR), while enabling imaging at earlier and shorter time-windows. The improved accuracy would be beneficial in better quantifying amyloid re-emergence post anti-amyloid therapy and could be used for kinetic harmonization across time-windows and radiotracers.

As neuroimaging analysis shifts toward large-scale, multi-site studies, managing the unwanted variability introduced by combining heterogeneous datasets has become a critical challenge. Although tools such as ComBat and its neuroimaging extensions are widely used to address this variability, they only permit the modeling of categorical site effects and cannot account for continuous sources of confounding, such as image quality, head motion, and acquisition parameters. We introduce ComCat, an extension of the ComBat framework that preserves biologically relevant covariates while removing the effects of categorical site indicators and continuous nuisance variables. The latter are modeled as smooth nonlinear functions via B-spline basis expansion. ComCat is applicable to a broad range of brain analysis tasks, including voxel- and surface-based morphometry, normative modeling, and machine learning-based prediction. To demonstrate its capabilities, we evaluated ComCat on brain age prediction across five datasets covering complementary multi-site harmonization scenarios: ON-Harmony (10 subjects x 6 scanners; n = 80); the Buchert traveling-phantom dataset (1 subject x 116 scanners; n = 531); the Tohoku single-scanner, varying-acquisition dataset (n = 121); MR-ART (148 subjects with varying motion levels); and an ABIDE subset comprising 229 control subjects and 208 individuals with autism spectrum disorder across 14 scanners. Using image quality measures derived from CAT12 as continuous nuisance variables, ComCat reduced the mean absolute error (MAE) in brain age prediction relative to ComBat-GAM in all five datasets, including the two scenarios where site information was unavailable or uninformative. In the ABIDE dataset, ComCat improved harmonization while preserving the difference between the control and ASD groups, demonstrating that scanner-related variance can be removed without affecting biologically meaningful signals. ComCat can operate with or without site labels and is agnostic to the source of image quality metrics.

Aging affects individuals at varying biological rates, prompting the development of the Brain Age Index (BAI) to quantify neurobiological health relative to chronological age and disease risk. While structural MRI has dominated brain age prediction, its high cost, immobility, and low temporal resolution restrict its clinical scalability and responsiveness to transient neurophysiological changes. Electroencephalography (EEG) offers a highly scalable, portable, and temporally precise alternative capable of capturing dynamic brain states. However, the transition of EEG-based models to clinical biomarkers is impeded by methodological limitations, including small or biased datasets, inconsistent preprocessing pipelines, and a distinct lack of interpretable machine learning approaches. To address these persistent challenges, this paper presents a comprehensive, open-source, end-to-end pipeline for large-scale EEG-based brain age modeling. Developed using the Temple University Hospital EEG Corpus (TUEG) the largest publicly available resting-state EEG dataset. The pipeline encompasses rigorous data engineering, reproducible preprocessing, and robust feature extraction. Following quality control and subject-level dataset partitioning to definitively prevent data leakage, exactly 41,181 recordings were successfully retained. Two independent feature sets were extracted: the Catch22 time-series characteristics and a comprehensive set of spectral, aperiodic, and non-linear dynamics from the CCS toolbox. The methodology evaluates seven regression models, optimized via Optuna for hyperparameter tuning, and integrates SHAP (SHapley Additive exPlanations) for transparent feature importance analysis. By making this infrastructure publicly available, this work lowers the barrier to entry for large-cohort studies, fostering reproducible development and clinical validation of dynamic brain age biomarkers.

Subject-independent affect regression from physiological signals remains challenging because emotional responses vary substantially across individuals, widely used datasets provide only coarse trial-level annotations, and heterogeneous physiological modalities may not contribute reliably when treated as if they were interchangeable predictors. We have developed AffectRoute, a protocol-conditioned subject-independent affect regression that is conditioned on the protocol and assigns separate predictive functions to information from the population, electroencephalography (EEG), and peripheral physiological signals (PPS). First, a source-population prior establishes a trial-level affective anchor using only the data from source participants. TrajBridge then combines an EEG representation that is supervised by REFED for participant-specific adjustments with temporal structure obtained from the continuous REFED annotations in order to create a weakly supervised segment-resolved pseudo-trajectory and to establish a frozen trial-level baseline. PhysioRoute next reduces the remaining error by breaking down the residual correction into a source-derived direction, which is estimated from the out-of-fold residuals within the source group, and a channel-specific magnitude derived from the PPS. When evaluated on DEAP and DREAMER using a leave-one-subject-out approach at the participant level, AffectRoute showed consistent step-by-step improvements in both the mean absolute error and the concordance correlation coefficient. A method that relied solely on the source data was clearly worse than PhysioRoute, showing that the final improvement cannot be accounted for by transferable source residual regularity alone. Conventional alternatives to fusing the PPS were also found to be consistently less effective, although analyses at the channel level and with a leave-one-channel-out design showed that the peripheral contributions are axis-dependent yet distributed across channels. These results indicate that structured residual inference is an effective alternative to unrestricted multimodal fusion for subject-independent affect regression.

Beta oscillations are a fundamental feature of brain activity, linked to long-range connectivity within canonical networks, inhibition of sensorimotor cortices and the maintenance of a stable sensorimotor state in situations where the external world is predictable. The importance of beta oscillations in brain function is underscored by observations of their perturbation in neurological and psychiatric disorders. However, the precise role played by beta activity, particularly in mediating complex or skilful movements, remains incompletely understood. Here, we used a newly developed wearable optically pumped magnetometer-based magnetoencephalography (OPM-MEG) system to measure beta dynamics as participants learned to play a musical instrument. Twenty-two novice players took part in a study in which OPM-MEG data were recorded during two scanning sessions, while participants attempted to play a tune on a violin. Between the two sessions, participants received a violin lesson from an expert teacher. Results showed that robust data could be acquired during this naturalistic task, with beta oscillations decreasing in amplitude during movement and increasing upon movement cessation, as expected. Moreover, beta power whilst playing, in the motor and pre-motor areas, was significantly elevated after the lesson compared to before, and the movement-related modulation of beta amplitude was more pronounced after the lesson. These findings align with predictive coding models which suggest that beta amplitude should increase when individuals have greater certainty over the movements they carry out. Our study adds to an expanding literature on the role of beta oscillations and provides further evidence for the utility of OPM-MEG in naturalistic neuroscience.

Understanding the relationship between neural dynamics and underlying brain structure, and in particular the impact of the latter on conduction delays, remains a core question in neuroscience. In humans, we primarily access these phenomena at the macroscale through neurophysiological measures of propagation. At this scale, signal propagation is reflected in measurable delays arising from both white matter (WM) axonal conduction and grey matter (GM) synaptic and integrative processes. While specific WM features have been explicitly linked to single-axon conduction delays, no clear models link GM microstructural properties to large-scale propagation. To this aim, we combined advanced MRI microstructural modelling with resting-state MEG to link brain structure to signal propagation in a sample of 94 healthy controls. WM metrics included axonal diameter, myelination (g-ratio, myelin water fraction), and tract length, while GM was characterised using the SANDI model and cortical thickness. Propagation delays were quantified using the neuronal avalanches framework. Our findings indicated significant associations between all of WM and GM metrics and the propagation delays. When attempting to use our W/GM metrics we were able to predict up to [~]26% of the observed delay using linear regression modelling. We also found associations between frequency-specific propagation dynamics and tract length. Finally, using a canonical correlation analysis we demonstrated multivariate coupling between our W/GM metrics and the MEG frequency-specific propagation delays. Our findings provide evidence for the link between tissue structure and large-scale neural dynamics, supporting the development of biologically grounded models of signal propagation.

Identifying informative voxels is a critical, yet challenging step in functional magnetic resonance imaging (fMRI), particularly for multivariate analyses involving multiple related conditions. Existing approaches often rely on predefined regions of interest (ROIs) or activation-based criteria, which may be insufficient for capturing fine-grained representational differences. This challenge becomes particularly relevant in experimental settings and interventions such as neurofeedback training, where voxels are not only measured as neural responses but also used as targets for intervention based on their previously observed activity patterns. In this study, we propose a subject-level searchlight optimization framework that integrates voxel-wise general linear model (GLM)-based univariate analysis with representational similarity analysis (RSA)-based multivariate refinement to identify voxels that are both task-relevant and condition-sensitive. To enhance practical applicability, the framework further incorporates a data-driven hyperparameter tuning step based on Bayesian optimization, enabling efficient identification of high-performing configurations from small pilot datasets, with consistent performance when applied to larger samples. The proposed framework was evaluated using an emotion imagery fMRI dataset with four affective conditions. Results demonstrate that the multivariate refinement improves alignment between empirical and target representational structures compared with univariate selection alone. Compared with a classifier-based voxel selection approach, the RSA-based approach better preserves the representational geometry of emotional states while maintaining discriminative capacity. These findings highlight the effectiveness, efficiency, and robustness of the proposed RSA framework, providing a practical solution for identifying condition-sensitive voxels and supporting more precise multivariate investigation of affective brain states in multi-condition fMRI studies.