Dynamic topographies of intrinsic neural timescales: a key role for consciousness.

Resting state EEG in patients with disorders of consciousness (DOC) is characterized by an increase of power in the delta frequency band and a concurrent decrease in the alpha range, equivalent to a weakening or disappearance of the alpha peak. Prolongation of Intrinsic Neural Timescales (INTs) is also associated with DOCs. Together, this raises the question whether the decreased alpha peak relates to the prolonged INTs and, importantly, how that can be used for diagnosing the state of consciousness in DOC individuals. Analyzing resting state EEG recordings from both healthy subjects and DOC patients, we measure INTs through autocorrelation window (ACW) and utilize peak analysis to quantify the weakening of the alpha peak. First, we replicate previous findings of prolonged ACW in DOC patients. We then identify significantly lower alpha peak measures in DOC compared to controls. Interestingly, spectral peaks shift from the alpha to the theta range in several DOC subjects while such change is absent in healthy controls. Next, our study reveals a close relationship between ACW and alpha peak in both healthy and DOC subjects, a correlation that holds for theta peaks in DOC. Further, the prolonged ACW correlates with the state of consciousness, as quantified by the Coma Recovery Scale-Revised (CRS-R), and mediates the relationship between theta peak and CRS-R. Finally, through split analyses and machine learning, we show that ACW and alpha peak measures conjointly distinguish healthy controls and DOC patients with high accuracy (95.5%). In conclusion, we demonstrate that the prolongation of ACW, together with spectral peak measures, holds promise to serve as additional EEG biomarkers for diagnosing the state of consciousness in DOC subjects.

Chapter 9 - The brain's inner time—intrinsic neural timescales

Recent neuroimaging studies demonstrate a heterogeneity of timescales prevalent in the brain’s ongoing spontaneous activity, labeled intrinsic neural timescales (INT). At the same time, neural timescales also reflect stimulus- or task-related activity. The relationship of the INT during the brain’s spontaneous activity with their involvement in task states including behavior remains unclear. To address this question, we combined calcium imaging data of spontaneously behaving mice and human electroencephalography (EEG) during rest and task states with computational modeling. We obtained four primary findings: (i) the distinct behavioral states can be accurately predicted from INT, (ii) INT become longer during behavioral states compared to rest, (iii) INT change from rest to task is correlated negatively with the variability of INT during rest, (iv) neural mass modeling shows a key role of recurrent connections in mediating the rest-task change of INT. Extending current findings, our results show the dynamic nature of the brain’s INT in reflecting continuous behavior through their flexible rest-task modulation possibly mediated by recurrent connections.

Input processing in the brain is mediated by phase synchronization and intrinsic neural timescales, both of which have been implicated in schizophrenia. Their relationship remains unclear, though. Recruiting a schizophrenia EEG sample from the B-SNIP consortium dataset (n = 134, 70 schizophrenia patients, 64 controls), we investigate phase synchronization, as measured by intertrial phase coherence (ITPC), and intrinsic neural timescales, as measured by the autocorrelation window (ACW) during both the rest and oddball-task states. The main goal of our paper was to investigate whether reported shifts from shorter to longer timescales are related to decreased ITPC. Our findings show (i) decreases in both theta and alpha ITPC in response to both standard and deviant tones; and (iii) a negative correlation of ITPC and ACW in healthy subjects while such correlation is no longer present in SCZ participants. Together, we demonstrate evidence of abnormally long intrinsic neural timescales (ACW) in resting-state EEG of schizophrenia as well as their dissociation from phase synchronization (ITPC). Our data suggest that, during input processing, the resting state’s abnormally long intrinsic neural timescales tilt the balance of temporal segregation and integration towards the latter. That results in temporal imprecision with decreased phase synchronization in response to inputs. Our findings provide further evidence for a basic temporal disturbance in schizophrenia on the different timescales (longer ACW and shorter ITPC), which, in the future, might be able to explain common symptoms related to the temporal experience in schizophrenia, for example temporal fragmentation.

The relationship of the brain's intrinsic neural timescales (INTs) during the resting state with event-related activity in response to external stimuli remains poorly understood. Here, we bridge this gap by combining computational modeling with human magnetoencephalography (MEG, resting state: N=64, 45 female; task state: N=58, 41 female) data to investigate the relation of intrinsic neuronal timescales (INT) with task-related activity, e.g., event-related fields (ERFs). Using the Jansen-Rit model, we first show that intracolumnar (and thus intra-regional) excitatory and inhibitory connections (rather than inter-regional feedback, feedforward and lateral connections between the columns of different regions) strongly influence both resting state INTs and task-related ERFs. Secondly, our results demonstrate a positive relationship between the magnitude of event-related fields (mERFs) and INTs, observed in both model simulations and empirical MEG data collected during an emotional face recognition task. Thirdly, modeling shows that the positive relationship of mERF and INT depends on intracolumnar connections through observing that the correlation between them disappears for fixed values of intracolumnar connections. Together, these findings highlight the importance of intracolumnar connections as a shared biological mechanism underlying both the resting-state's INTs and the task-state's event-related activity including their interplay.Significance Statement Intrinsic neural timescales (INTs) reflect the temporal persistence of neural activity and are increasingly recognized as a fundamental property of brain dynamics. How INTs relate to event-related neural responses remains poorly understood. Bridging this gap would give us a wider perspective on rest-task relationship in the brain. In this study, we combine modeling and magnetoencephalography (MEG) data to investigate this relationship. Modeling shows that intracolumnar connectivity simultaneously modulates both resting-state INTs and task-state event-related activity, leading to a positive correlation. The positive correlation is confirmed in MEG data. By bridging the gap between resting-state dynamics and task-state event-related responses, our work advances the understanding of how spontaneous and stimulus-driven brain processes are fundamentally intertwined.

Decreased intrinsic neural timescale in treatment-naïve adolescent depression

Intrinsic neural timescales (INT) reflect the time window of neural integration within a brain region and can be measured via resting-state functional magnetic resonance imaging (rs-fMRI). Despite the potential relevance of INT to cognition, brain organization, and neuropsychiatric illness, the influences of physiological artifacts on rs-fMRI INT have not been systematically considered. Two artifacts, head motion and respiration, pose serious issues in rs-fMRI studies. Here, we described their impact on INT estimation and tested the ability of two denoising strategies for mitigating these artifacts, high-motion frame censoring and global signal regression (GSR). We used a subset of the Human Connectome Project Young Adult (HCP-YA) dataset with runs annotated for breathing patterns (Lynch et al., 2020) and at least one “clean” (reference) run that had minimal head motion and no respiration artifacts; other runs from the same participants (n= 46) were labeled as “non-clean.” We found that non-clean runs exhibited brain-wide increases in INT compared with their respective clean runs and that the magnitude of error in INT between non-clean and clean runs correlated with the amount of head motion. Importantly, effect sizes were comparable with INT effects reported in the clinical literature. GSR and high-motion frame censoring improved the similarity between INT maps from non-clean runs and their respective clean run. Using a pseudo-random frame-censoring approach, we uncovered a relationship between the number of censored frames and both the mean INT and mean error, suggesting that frame censoring itself biases INT estimation. A group-level correction procedure reduced this bias and improved similarity between non-clean runs and their respective clean run. Based on our findings, we offer recommendations for rs-fMRI INT studies, which include implementing GSR and high-motion frame censoring with Lomb–Scargle interpolation of censored frames, and performing group-level correction of the bias introduced by frame censoring.

Intrinsic neural timescales relate to the dynamics of infraslow neural waves

Schizophrenia is hypothesized to involve a disturbance in the temporal dynamics of self-processing, specifically within the interoceptive, exteroceptive, and cognitive layers of the self. This study aimed to investigate the intrinsic neural timescales (INTs) within these self-processing layers among people with schizophrenia. We conducted a functional magnetic resonance imaging (fMRI) study to investigate INTs, as measured by the autocorrelation window, among people with schizophrenia and healthy controls during both resting-state and task (memory encoding and retrieval) conditions. We obtained data from the UCLA Consortium for Neuropsychiatric Phenomics data set and preprocessed using fMRIPrep. We included 45 people with schizophrenia and 65 healthy controls. Compared with controls, participants with schizophrenia exhibited significantly shorter INTs across all 3 self-processing layers during rest (p < 0.05). In addition, those with schizophrenia showed less INT shortening during task states, leading to reduced rest-task differences in INT across all self-processing layers (p < 0.05). We observed similar patterns of shortened INTs in primary sensory and motor regions. We included people with schizophrenia taking medication, which may influence INTs; our study was also limited by the relatively slow temporal resolution of the fMRI data and the higher variability of the autocorrelation function in the schizophrenia group, compared with the control group. Our findings suggest that schizophrenia is characterized by a global temporal disturbance of the self, manifesting as shorter and inflexible INTs across self-processing and sensorimotor regions. These results support the hypothesis that schizophrenia involves a fundamental disruption in the temporal integration of neural signals, contributing to the core self-disturbance observed in the disorder.

Intrinsic neural timescales, which reflect the duration of neural information storage within local brain regions and capacity for information integration, are typically measured using autocorrelation windows (ACWs). Extraction of intrinsic neural timescales from resting-state brain activity has been extensively applied in psychiatric disease research. Given the potential of intrinsic neural timescales as a neural marker for psychiatric disorders, investigating their reliability is crucial. This study, using an open-source database, aimed to evaluate the test-retest reliability of ACW-0 and ACW-50 under both eyes-open and eyes-closed conditions across three sessions. The intraclass correlation coefficients (ICCs) were employed to quantify the reliability of the intrinsic neural timescales. Our results showed that intrinsic neural timescales exhibited good reliability (ICC > 0.6) at the whole-brain level across different index types and eye states. Spatially, except for the right temporal region in the eyes-open condition, all other regions showed moderate-to-high ICCs. Over 60% of the electrodes demonstrated moderate-to-high intrinsic neural timescale ICCs under both eyes-open and eyes-closed conditions, with ACW-0 being more stable than ACW-50. Moreover, in the new dataset, the above results were consistently reproduced. The present study comprehensively assessed the reliability of intrinsic neural timescale under various conditions, providing robust evidence for their stability in neuroscience and psychiatry.

Intrinsic neural timescales (INT) reflect the duration for which brain areas store information. A posterior–anterior hierarchy of increasingly longer INT has been revealed in both typically developed individuals (TD), as well as persons diagnosed with autism spectrum disorder (ASD) and schizophrenia (SZ), though INT are, overall, shorter in both patient groups. In the present study, we aimed to replicate previously reported group differences by comparing INT of TD to ASD and SZ. We partially replicated the previously reported result, showing reduced INT in the left lateral occipital gyrus and the right post-central gyrus in SZ compared to TD. We also directly compared the INT of the two patient groups and found that these same two areas show significantly reduced INT in SZ compared to ASD. Previously reported correlations between INT and symptom severity were not replicated in the current project. Our findings serve to circumscribe the brain areas that can potentially play a determinant role in observed sensory peculiarities in ASD and SZ.

During rest, intrinsic neural dynamics manifest at multiple timescales, which progressively increase along visual and somatosensory hierarchies. Theoretically, intrinsic timescales are thought to facilitate processing of external stimuli at multiple stages. However, direct links between timescales at rest and sensory processing, as well as translation to the auditory system are lacking. Here, we measured intracranial electroencephalography in 11 human patients with epilepsy (4 women), while listening to pure tones. We show that in the auditory network, intrinsic neural timescales progressively increase, while the spectral exponent flattens, from temporal to entorhinal cortex, hippocampus, and amygdala. Within the neocortex, intrinsic timescales exhibit spatial gradients that follow the temporal lobe anatomy. Crucially, intrinsic timescales at baseline can explain the latency of auditory responses: as intrinsic timescales increase, so do the single-electrode response onset and peak latencies. Our results suggest that the human auditory network exhibits a repertoire of intrinsic neural dynamics, which manifest in cortical gradients with millimeter resolution and may provide a variety of temporal windows to support auditory processing.SIGNIFICANCE STATEMENT:Endogenous neural dynamics are often characterized by their intrinsic timescales. These are thought to facilitate processing of external stimuli. However, a direct link between intrinsic timing at rest and sensory processing is missing. Here, with intracranial electroencephalography (iEEG), we show that intrinsic timescales progressively increase from temporal to entorhinal cortex, hippocampus, and amygdala. Intrinsic timescales at baseline can explain the variability in the timing of iEEG responses to sounds: cortical electrodes with fast timescales also show fast and short-lasting responses to auditory stimuli, which progressively increase in the hippocampus and amygdala. Our results suggest that a hierarchy of neural dynamics in the temporal lobe manifests across cortical and limbic structures and can explain the temporal richness of auditory responses.

Are intrinsic neural timescales related to sensory processing? Evidence from abnormal behavioral states

Integrative gray matter volume and molecular analyses of altered intrinsic neural timescale in internet gaming disorder.

Temporal lobe epilepsy (TLE) is the most common pharmacoresistant epilepsy in adults. Here we profiled local neural function in TLE in vivo, building on prior evidence that has identified widespread structural alterations. Using resting-state functional magnetic resonance imaging (rs-fMRI), we mapped the whole-brain intrinsic neural timescales (INT), which reflect temporal hierarchies of neural processing. Parallel analysis of structural and diffusion MRI data examined associations with TLE-related structural compromise. Finally, we evaluated the clinical utility of INT. We studied 46 patients with TLE and 44 healthy controls from two independent sites, and mapped INT changes in patients relative to controls across hippocampal, subcortical, and neocortical regions. We examined region-specific associations to structural alterations and explored the effects of age and epilepsy duration. Supervised machine learning assessed the utility of INT for identifying patients with TLE vs controls and left- vs right-sided seizure onset. Relative to controls, TLE showed marked INT reductions across multiple regions bilaterally, indexing faster changing resting activity, with strongest effects in the ipsilateral medial and lateral temporal regions, and bilateral sensorimotor cortices as well as thalamus and hippocampus. Findings were similar, albeit with reduced effect sizes, when correcting for structural alterations. INT reductions in TLE increased with advancing disease duration, yet findings differed from the aging effects seen in controls. INT-derived classifiers discriminated patients vs controls (balanced accuracy, 5-fold: 76% ± 2.65%; cross-site, 72%-83%) and lateralized the focus in TLE (balanced accuracy, 5-fold: 96% ± 2.10%; cross-site, 95%-97%), with high accuracy and cross-site generalizability. Findings were consistent across both acquisition sites and robust when controlling for motion and several methodological confounds. Our findings demonstrate atypical macroscale function in TLE in a topography that extends beyond mesiotemporal epicenters. INT measurements can assist in TLE diagnosis, seizure focus lateralization, and monitoring of disease progression, which emphasizes promising clinical utility.