Sleep causes major shifts in how the brain uses energy and how blood flow changes, but the detailed timing and patterns of these processes are still not fully understood. In this study, researchers combined��functional PET,��EEG, and��fMRI—three powerful brain-imaging tools collected at the same time—to track how metabolism and blood flow change as people transition from wakefulness into��non-rapid eye movement (NREM) sleep.

They found that as the brain moves into NREM sleep,��global glucose metabolism steadily decreases, and at the same time,��large hemodynamic (blood-flow–related) fluctuations��begin to appear. Both of these changes closely follow shifts in��EEG arousal levels, showing how tightly linked these processes are during the onset of sleep.

The study also identified two distinct brain network patterns unique to NREM sleep:

• Aslow (~0.02 Hz), oscillating sensorimotor network��that stays metabolically active and dynamic, meaning sensory and motor areas continue to respond even during sleep.
• Adefault-mode network (DMN)��that shows reduced hemodynamic and metabolic activity, reflecting the decreased self-awareness and internal thought typical of sleep.

These findings help explain why we lose conscious awareness during sleep but can still respond to certain sensory signals. They also highlight the complex and alternating balance of neuronal, blood-flow, and metabolic activity that shapes brain function during sleep.

Finally, this work shows how combining EEG, PET, and MRI can offer powerful new insights into the biological mechanisms behind cognition, arousal, and sleep in humans.

Fig. 1: Trimodal imaging of the electrophysiological, BOLD-fMRI, and fPET-FDG metabolic dynamics accompanying arousal-state transitions.

Study protocol for the design, implementation, and evaluation of the STRATIFY clinical decision support tool for emergency department disposition of patients with heart failure

a,��b��Top: Hypnogram of scored sleep staging and the spectrogram of an occipital EEG electrode; middle: fMRI-based hemodynamic oscillations of the visual network; bottom: fPET-based metabolic dynamics of the default-mode network. Networks were extracted using a public functional atlas. Functional PET signals were temporally detrended according to the arbitrarily chosen initial wakeful period (i.e., removal of the linear trend fitted to the data points within the shaded gray area) only in this plot to help visualize altered slopes at state transitions, with an increase/decrease of TAC slope indicating increased/decreased metabolism. Changes in electrophysiological recordings, fMRI intensities, and glucose metabolism (highlighted with green arrows) were identifiable across sleep-wake cycles (a, subject i) and within the NREM sleep (b, subject ii), mirroring arousal-state transitions (top, inferred from simultaneous EEG recordings). Note that our goal here is to highlight arousal-induced changes in the imaging signals, so we show fMRI and fPET signals from exemplar networks that exhibit strong sleep-wake differences for each modality independently (see group-level results in Fig.����below; to avoid a circular analysis, we re-ran the group-level analysis without including the two illustrative subjects shown here, the findings remained unaltered).

Risk-taking behavior is a feature of many neuropsychiatric disorders, yet effective treatments are limited because we still don’t fully understand what is happening in the brain when people make risky choices. Scientists know that certain “reward circuitry” regions—such as the amygdala, orbitofrontal cortex, insula, and anterior cingulate—are involved, but the specific electrical activity that predicts risk-taking in these areas has not been well studied in humans. Identifying local field potential (LFP) frequency patterns linked to risk-taking could help guide future therapies.

In this study, eleven patients with hard-to-treat epilepsy, who already had stereotactic EEG electrodes implanted for medical reasons, took part in an experiment measuring brain activity in these reward-related regions. Each person played a simple gambling game in which they guessed whether a hidden playing card would be lower or higher than a visible one, choosing to bet either $5 or $20. While they made these decisions, researchers recorded their local field potentials—electrical signals generated by groups of neurons. The team used statistical models to look for specific changes in oscillatory power (brainwave activity across different frequencies) related to reward prediction error, which is the difference between expected and actual outcomes. They also calculated a “risk-taking value” for each trial based on the card number and the size of the bet, and identified which oscillatory patterns were linked to riskier choices.

The results showed clear time-frequency patterns associated with reward prediction error signals in both the amygdala and the orbitofrontal cortex, with several strong clusters of activity in each region. Risky choices themselves were predicted by increased oscillatory power in the theta-to-beta frequency range in the orbitofrontal cortex during card presentation, and by increased high-beta power in the insula. Further analysis pinpointed these signals to the lateral orbitofrontal cortex and the posterior insula. Activity in one insula cluster linked to risky decisions was also connected to a theta-alpha reward prediction error signal in the orbitofrontal cortex. Additionally, an amygdala reward prediction error signal was associated with how often participants chose the higher bet, and a lateral orbitofrontal cortex signal predicted high bets specifically in risky situations.

Overall, the study identifies distinct electrical activity patterns in key reward-related brain regions that predict when a person is about to make a risky decision. These oscillatory signatures could eventually serve as biomarkers—measurable indicators that help guide new treatments, including closed-loop neuromodulation, for disorders in which risk-taking becomes harmful.

Figure 1

Gambling task and behavioural data. (A) Behavioural task events and timing between events in seconds (mean ± standard deviation). (B) Mean response time (time from bet cue presentation to patient response) for each patient card number. Error bars represent standard error. (C) Mean per cent of trials on which subjects bet high ($20) for each patient card number. Asterisk indicates��P��< 0.05. Error bars represent standard error.

Vigilance, or how alert and awake we are, constantly changes and affects our thinking and behavior. This state can be disrupted in many brain disorders. Certain areas deep in the brain, called neuromodulatory nuclei in the brainstem and basal forebrain, help regulate alertness and drive widespread brain activity and communication. However, it is not well understood how the brain’s large-scale networks change when we shift between being alert and drowsy.

In this study, we used simultaneous EEG (which measures brain electrical activity) and advanced fMRI scans to explore how these arousal centers connect with other parts of the brain depending on vigilance. We found that when people are drowsy, most of these nuclei show stronger global connections, especially to regions like the thalamus, precuneus, and sensory and motor areas. When people are more alert, the nuclei connect most strongly to networks involved in attention, internal thought, and hearing. These patterns remained consistent even after controlling for blood flow effects.

To confirm our findings, we analyzed two large brain imaging datasets and showed that these connectivity patterns are reproducible across different types of fMRI scans. Overall, this study provides new insights into how brain regions that regulate arousal influence large-scale brain activity depending on our level of alertness.

Fig 1 – Reproducible static connectivity profiles of neuromodulatory arousal centers. (a) Static functional connectivity (FC) t-maps of the locus coresuleus (LC), cuneiform/subcuneiform nucleus (CSC), and nucleus basalis of Meynert (NBM) in the VU 3T-ME, HCP 3T, and HCP 7T datasets for the mCSF/WM preprocessing pipeline. The FC t-maps were thresholded at 40% of the top t-values in the gray matter and at p < 0.05 (voxel-wise false discovery rate [FDR]-corrected over the entire gray matter volume). AFNI was used for visualization of the t-maps (@chauffeur_afni function; upper functional range set to the 98^th��percentile). (b) Spatial overlap of the thresholded static FC t-maps of the subcortical arousal regions with 16 canonical brain network templates from the FINDLAB and Melbourne atlases (Shirer et al., 2012;��Tian et al., 2020). A positive value for the spatial overlap corresponds to mostly positive correlations within the brain network template while a negative value corresponds to mostly negative correlations. (c) Spatial reproducibility (Dice similarity coefficient) of the thresholded static FC t-maps between the three fMRI datasets.

]]> /valiant/2025/06/20/functional-mri-signals-exhibit-stronger-covariation-with-peripheral-autonomic-measures-as-vigilance-decreases/ Fri, 20 Jun 2025 18:26:39 +0000 /valiant/?p=4566 Gold, Benjamin P.; Goodale, Sarah E.; Zhao, Chong; Pourmotabbed, Haatef; de Zwart, Jacco A.; Özbay, Pinar S.; Bolt, Taylor S.; Duyn, Jeff H.; Chen, Jingyuan E.; Chang, Catie. Imaging Neuroscience 2 (2024): 1-25. .��

Vigilance—our level of alertness or attention—naturally rises and falls over time. These shifts are known to affect signals seen in brain scans, such as those from functional magnetic resonance imaging (fMRI), though the exact cause of these changes isn’t fully understood. Separate studies have connected changes in vigilance not only to brain signal patterns but also to changes in physical responses controlled by the autonomic nervous system, such as breathing and heart rate. This raises the question: could some of the brain signal changes actually be caused by these bodily responses?��

To explore this, we recorded fMRI scans alongside measures of brain activity (EEG), breathing, and blood oxygen levels, while people were either resting or doing a task that required attention. We found that the link between the body’s automatic functions (like pulse and respiration) and brain signals became stronger as people became less alert. These body-related signals first showed quick positive connections with brain activity, then slower negative ones, with some later positive responses in fluid-filled spaces of the brain.��

We also saw that fluctuations in EEG (a measure of brainwave activity) depended on alertness level and were related to both brain and body signals. Additionally, the strength of communication between different brain regions (called functional connectivity) increased when people were less alert—especially during rest. But when we removed the influence of body signals from the fMRI data, this increase mostly disappeared.��

Overall, our results show that changes in alertness affect not just brain activity but also how the body and brain interact. This understanding helps scientists more accurately interpret fMRI data by highlighting the important role of physiological changes.��

��

Fig 1. ��

Comparing EEG, fMRI, autonomic, and behavioral measures across time windows. (A) Simultaneous EEG, fMRI, and autonomic data were divided into non-overlapping windows of 126 s each. This panel shows three representative, contiguous windows (the fourth, fifth, and sixth windows from rest participant 3), including their “fast” (i.e., seconds-level) and baseline (i.e., window-averaged) EEG alpha/theta ratios, for a participant in the process of falling asleep. (B) For the task scans, we compared the mean of the EEG alpha/theta power ratio within each window (which we define as “baseline vigilance”) to the mean reaction time in each window with Spearman’s rank correlations for non-normal distributions. Significant negative correlations, whether excluding trials without responses (“Responses only”) or including them as indicating arbitrarily long reaction times of 4 s (“All trials”), support the use of an EEG alpha/theta ratio as a measure of vigilance in this study. (C) The temporal variance of the percent signal change in the fMRI global signal also exhibited a negative relationship with baseline vigilance levels (shifted by 4.2 s in this case to accommodate the hemodynamic delay of the fMRI signal). This effect was significant for both resting-state and task data, indicating greater global fMRI variability as baseline vigilance decreases. Although the correlation values shown in (B–C) are based on non-parametric statistics, we include least-squares trend lines for visualization. RV = respiratory volume, HR = heart rate, PWA = pulse wave amplitude.��

��

The brain is highly sensitive to signals from the body’s internal environment, as shown by the many links between brain activity, blood flow (hemodynamics), and various physiological signals. In this study, researchers found that a major form of brain–body interaction can be described by a single spatiotemporal pattern—that is, a consistent pattern across space and time.��

Using data from several independent groups and different types of functional MRI (fMRI) scans—including both single-echo and multi-echo sequences—they observed low-frequency cofluctuations (in the 0.01–0.1 Hz range) between global fMRI signals during rest, EEG (electroencephalogram) activity, and a wide range of peripheral autonomic signals. These included signals from the cardiovascular, respiratory, exocrine, and smooth muscle systems.��

Interestingly, the same brain–body cofluctuations appeared not only at rest but also during cued deep breathing, in response to sensory stimuli, and during spontaneous EEG events in sleep. Even when researchers controlled for changes in end-tidal carbon dioxide (a measure of CO₂ in the lungs), the overall spatial pattern of global fMRI activity remained. This suggests that the source of these signals cannot be fully explained by breathing-related CO₂ changes alone.��

These results indicate that the global fMRI signal reflects a key part of the brain’s arousal system, which is regulated by the autonomic nervous system—a system that helps manage vital functions like heart rate, breathing, and alertness.��

Fig. 1: Global fMRI fluctuations are associated with systemic physiological changes.��

��

The cross-correlation between the time courses of the global fMRI signal (PC1) and multiple physiological signals.��a, The spatial weights of the PC1 (left), and the phase delay map of the CPC1 (right) from the ME-REST dataset. The phase delay map of the CPC1 encodes the time delay (in radians) between voxels within the component. Because phase delay is measured in radians (0–2π), they are displayed with a circular color map.��b, Cross-correlation plots of each physiological signal with the global fMRI time course (PC1). The cross-correlation is defined as the correlation coefficient between��xt + i��and��yt, where��t��is the time index,��i��is ±30 s (that is, the index along the��x��axis of the plots) and��x��is the global fMRI signal and��y��is the physiological signal. Strong correlation at a positive time lag (that is, positive��i��index) indicates that the global fMRI signal lags or follows the physiological signal, while a strong correlation at a negative time lag (that is, negative��i��index) indicates that the global fMRI signal leads the physiological signal. Within a dataset, all participant-level cross-correlations are displayed in a lighter, more transparent color, while the mean cross-correlation across participants is displayed in a darker color. Each dataset is displayed in a separate shade of the same color. Cross-correlations between global fMRI signals and wavelet-filtered EEG power signals (Morlet wavelet; number of cycles = 15) for the ME-REST and NATVIEW-REST datasets are displayed as a heat map in the top right. The figure was created with��.��

Functional magnetic resonance imaging (fMRI) is a vital tool in neuroscience for observing brain activity, but it has limitations such as hemodynamic blurring, high cost, immobility, and incompatibility with metal implants. Electroencephalography (EEG), while offering high temporal resolution by directly recording cortical electrical activity, has limited spatial resolution and cannot capture data from deep brain structures. Combining the strengths of both methods by deriving fMRI information from EEG could provide a cost-effective way to image a broader range of brain regions and enhance the interpretation of fMRI signals.

To address the challenge of modeling fMRI from EEG, given the high dimensionality and artifact-prone nature of both modalities, a novel architecture is proposed. This model predicts fMRI signals directly from multi-channel EEG without the need for explicit feature engineering. It employs a Sinusoidal Representation Network (SIREN) to learn frequency information in brain dynamics from EEG data, which is then input to an encoder-decoder system to reconstruct the fMRI signal for specific brain regions.

The model was tested using a simultaneous EEG-fMRI dataset from eight subjects, focusing on its ability to predict subcortical fMRI signals. Results show that this new model outperforms recent state-of-the-art models, demonstrating the potential of using periodic activation functions in deep neural networks for functional neuroimaging data. This approach could significantly expand the capabilities of EEG and improve the cost-effectiveness and accessibility of brain imaging.

The study investigates auditory issues in individuals with mild traumatic brain injury (mTBI), a group often reporting hearing problems despite normal hearing tests. Researchers used a comprehensive battery of tests assessing both peripheral and central auditory system functions, including pure-tone detection, word and sentence understanding, and various auditory evoked potentials (AEPs), alongside MRI scans. Findings revealed that mTBI patients showed notable changes in several auditory tests, such as reduced otoacoustic emissions, altered middle-ear reflex thresholds, and variations in AEPs indicating auditory processing difficulties. Particularly, those with combined hearing difficulty and noise sensitivity displayed more significant auditory processing deficits and structural brain changes in regions linked to auditory processing, such as the transverse temporal gyrus and planum polare. These results underscore the complexity of auditory issues in mTBI and the need for tailored diagnostic and treatment approaches that account for the nuanced effects of brain injuries on hearing.