Trait anxiety is related to Nx4's efficacy on stress-induced changes in amygdala-centered resting state functional connectivity: a placebo-controlled cross-over trial in mildly to moderately stressed healthy volunteers.

BACKGROUND: The multicomponent drug Neurexan (Nx4) was shown to reduce the neural stress network activation. We now investigated its effects on stress-induced resting state functional connectivity (RSFC) in dependence of trait anxiety (TA), an acknowledged vulnerability factor for stress-induced psychopathologies. METHODS: Nx4 was tested in a randomized placebo-controlled crossover trial. Resting state fMRI scans were performed before and after a psychosocial stress task and exploratively analyzed for amygdala centered RSFC. Effects of Nx4 on stress-induced RSFC changes were evaluated and correlated to TA levels. A subgroup analysis based on TA scores was performed. RESULTS: Multiple linear regression analysis revealed a significant correlation between TA and Nx4 effect on stress-induced RSFC changes between right amygdala and pregenual anterior cingulate cortex (pgACC) and ventro-medial prefrontal cortex (vmPFC). For participants with above average TA, a significant amelioration of the stress-induced RSFC changes was observed. CONCLUSIONS: The data add evidence to the hypothesis that Nx4's clinical efficacy is based on a dampened activation of the neural stress network, with a greater neural response in subjects with anxious personality traits. Further studies assessing clinically relevant outcome measures in parallel to fMRI are encouraged to evaluate the real-world benefit of Nx4. Trial registration NCT02602275.

Nx4 Modulated Resting-State Functional Connectivity Between Amygdala and Prefrontal Cortex in a Placebo-Controlled, Crossover Trial.

Background: The basic functional organization of the resting brain, assessed as resting-state functional connectivity (rsFC), can be affected by previous stress experience and it represents the basis on which subsequent stress experience develops. Notably, the rsFC between the amygdala and the cortical regions associated with emotion regulation and anxiety are affected during stress. The multicomponent drug Neurexan® (Nx4) has previously demonstrated a reduction in amygdala activation in an emotional face matching task and it ameliorated stress-related symptoms. We, thus, investigated the effect of Nx4 on rsFC of the amygdala before stress induction compared with baseline in mildly to moderately stressed participants. Methods: In a randomized, placebo-controlled, double-blind, crossover trial 39 participants received a single dose of placebo or Nx4. Resting-state functional magnetic resonance imaging scans were performed pre-dose and 40 to 60 min post-dose, before any stress induction. First, highly connected functional hubs were identified by global functional connectivity density (gFCD) analysis. Second, by using a seed-based approach, rsFC maps of the left centromedial amygdala (CeMA) were created. The effect of Nx4 on both was evaluated. Results: The medial prefrontal cortex was identified as a relevant functional hub affected by Nx4 in an explorative whole brain gFCD analysis. Using the seed-based approach, we then demonstrated that Nx4 significantly enhanced the negative connectivity between the left CeMA and two cortical regions: the dorsolateral and medial prefrontal cortices. Conclusions: In a resting-state condition, Nx4 reduced the prefrontal cortex gFCD and strengthened the functional coupling between the amygdala and the prefrontal cortex that is relevant for emotion regulation and the stress response. Further studies should elaborate whether this mechanism represents enhanced regulatory control of the amygdala at rest and, consequently, to a diminished susceptibility to stress. ClinicalTrials.gov ID: NCT02602275.

Nx4 attenuated stress-induced activity of the anterior cingulate cortex-A post-hoc analysis of a randomized placebo-controlled crossover trial.

OBJECTIVE: Stress-related symptoms are associated with significant health and economic burden. Several studies suggest Nx4 for the pharmacological management of the stress response and investigated the underlying neural processes. Here we hypothesized that Nx4 can directly affect the stress response in a predefined stress network, including the anterior cingulate cortex (ACC), which is linked to various stress-related symptoms in patients. METHODS: In a randomized, placebo-controlled, double-blind, crossover trial, 39 healthy males took a single dose of placebo or Nx4. Psychosocial stress was induced by the ScanSTRESS paradigm inside an MRI scanner, and stress network activation was analyzed in brain regions defined a priori. RESULTS: Using the placebo data only, we could validate the activation of a distinct neural stress pattern by the ScanSTRESS paradigm. For Nx4, we provide evidence of an attenuating effect on this stress response. A statistically significant reduction in differential stress-induced activation in the right supracallosal ACC was observed for the rotation stress task of the ScanSTRESS paradigm. The results add to previously published results of Nx4 effects on emotion regulation. CONCLUSIONS: Our results strengthen the hypothesis that Nx4 modulates the stress response by reducing the activation in parts of the neural stress network, particularly in the ACC. TRIAL REGISTRATION: NCT02602275; ClinicalTrials.gov.

Corrigendum to "EEG revealed improved vigilance regulation after stress exposure under Nx4 - A randomized, placebo-controlled, double-blind, cross-over trial" [IBRO Neurosci. Rep. 11C (2021) 175-182].

[This corrects the article DOI: 10.1016/j.ibneur.2021.09.002.].

EEG-Microstates Reflect Auditory Distraction After Attentive Audiovisual Perception Recruitment of Cognitive Control Networks.

Processing of sensory information is embedded into ongoing neural processes which contribute to brain states. Electroencephalographic microstates are semi-stable short-lived power distributions which have been associated with subsystem activity such as auditory, visual and attention networks. Here we explore changes in electrical brain states in response to an audiovisual perception and memorization task under conditions of auditory distraction. We discovered changes in brain microstates reflecting a weakening of states representing activity of the auditory system and strengthening of salience networks, supporting the idea that salience networks are active after audiovisual encoding and during memorization to protect memories and concentrate on upcoming behavioural response.

Nx4 Reduced Susceptibility to Distraction in an Attention Modulation Task.

Background: Stress adversely affects the attentional focus, the active concentration on stimuli, and increases susceptibility to distraction. To experimentally explore the susceptibility to distraction, the Attention Modulation by Salience Task (AMST) is a validated paradigm measuring reaction times (RT) for processing auditory information while presenting task-irrelevant visual distractors of high or low salience. We extended the AMST by an emotional dimension of distractors and an EEG-based evaluation. We then investigated the effect of the stress-relieving medication Neurexan (Nx4) on the participants' susceptibility to distraction. Methods: Data from a randomized, placebo-controlled, crossover trial (NEURIM study; ClinicalTrials.gov: NCT02602275) were exploratively reanalyzed post-hoc. In this trial, 39 participants received a single dose of placebo or Nx4 immediately before the AMST. Participants had to discriminate two different tone modulations (ascending or descending) while simultaneously perceiving task-irrelevant pictures of different salience (high or low) or valence (negative or positive) as distractors. Using EEG recordings, RT and the event-related potential (ERP) components N1, N2, and N3 were analyzed as markers for susceptibility to distraction. Results: In the placebo condition, we could replicate the previously reported task effects of salient distractors with longer RT for high salient distractors on the behavioral level. On the electrophysiological level, we observed significantly increased amplitudes of the N2 and N3 ERP components for positive emotional pictures. In terms of drug effect, we found evidence that Nx4 reduced distractibility by emotional distractors. The effect was shown by significantly reduced amplitudes of N2 and N3 ERP components and reduced RT for the positive valence domain under Nx4 compared to placebo. The Nx4 effects on RT and ERP components also showed a significant correlation. Conclusion: Emotional distractors in addition to the previously used salience distractors and the EEG based evaluation of ERPs valuably complement the AMST. Salient distractors were affecting attentional processes earlier, while valent distractors show modulatory effects later. Our results suggest that Nx4 has beneficial effects on attention by inhibiting the effect of task-irrelevant information and reducing susceptibility to emotionally distracting stimuli. The observation of a beneficial impact of Nx4 on attention regulation is supportive of Nx4's claim as a stress-relieving medication.

EEG revealed improved vigilance regulation after stress exposure under Nx4 - A randomized, placebo-controlled, double-blind, cross-over trial.

OBJECTIVES: Vigilance is characterized by alertness and sustained attention. The hyper-vigilance states are indicators of stress experience in the resting brain. Neurexan (Nx4) has been shown to modulate the neuroendocrine stress response. Here, we hypothesized that the intake of Nx4 would alter brain vigilance states at rest. METHOD: In this post-hoc analysis of the NEURIM study, EEG recordings of three, 12 min resting-state conditions in 39 healthy male volunteers were examined in a randomized, placebo-controlled, double-blind, cross-over clinical trial. EEG was recorded at three resting-state sessions: at baseline (RS0), after single-dose treatment with Nx4 or placebo (RS1), and subsequently after a psychosocial stress task (RS2). During each resting-state session, each 2-s segment of the consecutive EEG epochs was classified into one of seven different brain states along a wake-sleep continuum using the VIGALL 2.1 algorithm. RESULTS: In the post-stress resting-state, subjects exhibited a hyper-stable vigilance regulation characterized by an increase in the mean vigilance level and by more rigidity in the higher vigilance states for a longer period of time. Importantly, Nx4-treated participants exhibited significantly lower mean vigilance level compared to placebo-treated ones. Also, Nx4- compared to placebo-treated participants spent comparably less time in higher vigilance states and more time in lower vigilance states in the post-stress resting-state. CONCLUSION: Study participants showed a significantly lower mean vigilance level in the post-stress resting-state condition and tended to stay longer in lower vigilance states after treatment with Nx4. These findings support the known stress attenuation effect of Nx4.

The missing role of gray matter in studying brain controllability.

Brain controllability properties are normally derived from the white matter fiber tracts in which the neural substrate of the actual energy consumption, namely the gray matter, has been widely ignored. Here, we study the relationship between gray matter volume of regions across the whole cortex and their respective control properties derived from the structural architecture of the white matter fiber tracts. The data suggests that the ability of white fiber tracts to exhibit control at specific nodes not only depends on the connection strength of the structural connectome but additionally depends on gray matter volume at the host nodes. Our data indicate that connectivity strength and gray matter volume interact with respect to the brain's control properties. Disentangling effects of the regional gray matter volume and connectivity strength, we found that frontal and sensory areas play crucial roles in controllability. Together these results suggest that structural and regional properties of the white matter and gray matter provide complementary information in studying the control properties of the intrinsic structural and functional architecture of the brain.

Evidence for modulation of EEG microstate sequence by vigilance level.

The momentary global functional state of the brain is reflected in its electric field configuration and cluster analytical approaches have consistently shown four configurations, referred to as EEG microstate classes A to D. Changes in microstate parameters are associated with a number of neuropsychiatric disorders, task performance, and mental state establishing their relevance for cognition. However, the common practice to use eye-closed resting state data to assess the temporal dynamics of microstate parameters might induce systematic confounds related to vigilance levels. Here, we studied the dynamics of microstate parameters in two independent data sets and showed that the parameters of microstates are strongly associated with vigilance level assessed both by EEG power analysis and fMRI global signal. We found that the duration and contribution of microstate class C, as well as transition probabilities towards microstate class C were positively associated with vigilance, whereas the sign was reversed for microstate classes A and B. Furthermore, in looking for the origins of the correspondence between microstates and vigilance level, we found Granger-causal effects of vigilance levels on microstate sequence parameters. Collectively, our findings suggest that duration and occurrence of microstates have a different origin and possibly reflect different physiological processes. Finally, our findings indicate the need for taking vigilance levels into consideration in resting-sate EEG investigations.

Heart Rate Variability as an Index of Differential Brain Dynamics at Rest and After Acute Stress Induction.

The brain continuously receives input from the internal and external environment. Using this information, the brain exerts its influence on both itself and the body to facilitate an appropriate response. The dynamic interplay between the brain and the heart and how external conditions modulate this relationship deserves attention. In high-stress situations, synchrony between various brain regions such as the prefrontal cortex and the heart may alter. This flexibility is believed to facilitate transitions between functional states related to cognitive, emotional, and especially autonomic activity. This study examined the dynamic temporal functional association of heart rate variability (HRV) with the interaction between three main canonical brain networks in 38 healthy male subjects at rest and directly after a psychosocial stress task. A sliding window approach was used to estimate the functional connectivity (FC) among the salience network (SN), central executive network (CEN), and default mode network (DMN) in 60-s windows on time series of blood-oxygen-level dependent (BOLD) signal. FC between brain networks was calculated by Pearson correlation. A multilevel linear mixed model was conducted to examine the window-by-window association between the root mean square of successive differences between normal heartbeats (RMSSD) and FC of network-pairs across sessions. Our findings showed that the minute-by-minute correlation between the FC and RMSSD was significantly stronger between DMN and CEN than for SN and CEN in the baseline session [b = 4.36, t(5025) = 3.20, p = 0.006]. Additionally, this differential relationship between network pairs and RMSSD disappeared after the stress task; FC between DMN and CEN showed a weaker correlation with RMSSD in comparison to baseline [b = -3.35, t(5025) = -3.47, p = 0.006]. These results suggest a dynamic functional interplay between HRV and the functional association between brain networks that varies depending on the needs created by changing conditions.

Investigating the temporal dynamics of electroencephalogram (EEG) microstates using recurrent neural networks.

Electroencephalogram (EEG) microstates that represent quasi-stable, global neuronal activity are considered as the building blocks of brain dynamics. Therefore, the analysis of microstate sequences is a promising approach to understand fast brain dynamics that underlie various mental processes. Recent studies suggest that EEG microstate sequences are non-Markovian and nonstationary, highlighting the importance of the sequential flow of information between different brain states. These findings inspired us to model these sequences using Recurrent Neural Networks (RNNs) consisting of long-short-term-memory (LSTM) units to capture the complex temporal dependencies. Using an LSTM-based auto encoder framework and different encoding schemes, we modeled the microstate sequences at multiple time scales (200-2,000 ms) aiming to capture stably recurring microstate patterns within and across subjects. We show that RNNs can learn underlying microstate patterns with high accuracy and that the microstate trajectories are subject invariant at shorter time scales (≤400 ms) and reproducible across sessions. Significant drop in the reconstruction accuracy was observed for longer sequence lengths of 2,000 ms. These findings indirectly corroborate earlier studies which indicated that EEG microstate sequences exhibit long-range dependencies with finite memory content. Furthermore, we find that the latent representations learned by the RNNs are sensitive to external stimulation such as stress while the conventional univariate microstate measures (e.g., occurrence, mean duration, etc.) fail to capture such changes in brain dynamics. While RNNs cannot be configured to identify the specific discriminating patterns, they have the potential for learning the underlying temporal dynamics and are sensitive to sequence aberrations characterized by changes in metal processes. Empowered with the macroscopic understanding of the temporal dynamics that extends beyond short-term interactions, RNNs offer a reliable alternative for exploring system level brain dynamics using EEG microstate sequences.

Translational machine learning for psychiatric neuroimaging.

Despite its initial promise, neuroimaging has not been widely translated into clinical psychiatry to assist in the prediction of diagnoses, prognoses, and optimal therapeutic strategies. Machine learning approaches may enhance the translational potential of neuroimaging because they specifically focus on overcoming biases by optimizing the generalizability of pipelines that measure complex brain patterns to predict targets at a single-subject level. This article introduces some fundamentals of a translational machine learning approach before selectively reviewing literature to-date. Promising initial results are then balanced by the description of limitations that should be considered in order to interpret existing research and maximize the possibility of future translation. Future directions are then presented in order to inspire further research and progress the field towards clinical translation.

Multivariate classification of neuroimaging data with nested subclasses: Biased accuracy and implications for hypothesis testing.

Biological data sets are typically characterized by high dimensionality and low effect sizes. A powerful method for detecting systematic differences between experimental conditions in such multivariate data sets is multivariate pattern analysis (MVPA), particularly pattern classification. However, in virtually all applications, data from the classes that correspond to the conditions of interest are not homogeneous but contain subclasses. Such subclasses can for example arise from individual subjects that contribute multiple data points, or from correlations of items within classes. We show here that in multivariate data that have subclasses nested within its class structure, these subclasses introduce systematic information that improves classifiability beyond what is expected by the size of the class difference. We analytically prove that this subclass bias systematically inflates correct classification rates (CCRs) of linear classifiers depending on the number of subclasses as well as on the portion of variance induced by the subclasses. In simulations, we demonstrate that subclass bias is highest when between-class effect size is low and subclass variance high. This bias can be reduced by increasing the total number of subclasses. However, we can account for the subclass bias by using permutation tests that explicitly consider the subclass structure of the data. We illustrate our result in several experiments that recorded human EEG activity, demonstrating that parametric statistical tests as well as typical trial-wise permutation fail to determine significance of classification outcomes correctly.

Decoding cognitive concepts from neuroimaging data using multivariate pattern analysis.

Multivariate pattern analysis (MVPA) methods are now widely used in life-science research. They have great potential but their complexity also bears unexpected pitfalls. In this paper, we explore the possibilities that arise from the high sensitivity of MVPA for stimulus-related differences, which may confound estimations of class differences during decoding of cognitive concepts. We propose a method that takes advantage of concept-unrelated grouping factors, uses blocked permutation tests, and gradually manipulates the proportion of concept-related information in data while the stimulus-related, concept-irrelevant factors are held constant. This results in a concept-response curve, which shows the relative contribution of these two components, i.e. how much of the decoding performance is specific to higher-order category processing and to lower order stimulus processing. It also allows separating stimulus-related from concept-related neuronal processing, which cannot be achieved experimentally. We applied our method to three different EEG data sets with different levels of stimulus-related confound to decode concepts of digits vs. letters, faces vs. houses, and animals vs. fruits based on event-related potentials at the single trial level. We show that exemplar-specific differences between stimuli can drive classification accuracy to above chance levels even in the absence of conceptual information. By looking into time-resolved windows of brain activity, concept-response curves can help characterize the time-course of lower-level and higher-level neural information processing and detect the corresponding temporal and spatial signatures of the corresponding cognitive processes. In particular, our results show that perceptual information is decoded earlier in time than conceptual information specific to processing digits and letters. In addition, compared to the stimulus-level predictive sites, concept-related topographies are spread more widely and, at later time points, reach the frontal cortex. Thus, our proposed method yields insights into cognitive processing as well as corresponding brain responses.

Classification based hypothesis testing in neuroscience: Below-chance level classification rates and overlooked statistical properties of linear parametric classifiers.

Multivariate pattern analysis (MVPA) has recently become a popular tool for data analysis. Often, classification accuracy as quantified by correct classification rate (CCR) is used to illustrate the size of the effect under investigation. However, we show that in low sample size (LSS), low effect size (LES) data, which is typical in neuroscience, the distribution of CCRs from cross-validation of linear MVPA is asymmetric and can show classification rates considerably below what would be expected from chance classification. Conversely, the mode of the distribution in these cases is above expected chance levels, leading to a spuriously high number of above chance CCRs. This unexpected distribution has strong implications when using MVPA for hypothesis testing. Our analyses warrant the conclusion that CCRs do not well reflect the size of the effect under investigation. Moreover, the skewness of the null-distribution precludes the use of many standard parametric tests to assess significance of CCRs. We propose that MVPA results should be reported in terms of P values, which are estimated using randomization tests. Also, our results show that cross-validation procedures using a low number of folds, e.g. twofold, are generally more sensitive, even though the average CCRs are often considerably lower than those obtained using a higher number of folds. Hum Brain Mapp 37:1842-1855, 2016. © 2016 Wiley Periodicals, Inc.