Heliyon 8 (2022) e10208 Contents lists available at ScienceDirect Heliyon journal homepage: www.cell.com/heliyon Research article Prefrontal reactivity to TMS perturbation as a toy model of mental health outcomes during the COVID-19 pandemic Ruben Perellon-Alfonso a, b, María Redondo-Camo s c, d, e, Kilian Abellaneda-Perez a, b, c, Gabriele Cattaneo c, d, e, Selma Delgado-Gallen c, d, e, Goretti Espan ~a-Irla c, d, e, Javier Solana Sanchez c, d , e , Jose M. Tormos c, d , e , Alvaro Pascual-Leone c, f, g, **, 1, David Bartres-Faz a, b, c, *, 1 a Faculty of Medicine and Health Sciences, and Institute of Neurosciences, University of Barcelona, Barcelona, Spain b Institute of Biomedical Research August Pi i Sunyer (IDIBAPS), Barcelona, Spain c Institut Guttmann, Institut Universitari de Neurorehabilitacio adscrit a la UAB, Badalona, Barcelona, Spain d Universitat Autonoma de Barcelona, Bellaterra (Cerdanyola del Valles), Spain e Fundacio Institut d’Investigaci o en Ciencies de la Salut Germans Trias i Pujol, Badalona, Barcelona, Spain f Hinda and Arthur Marcus Institute for Aging Research and Deanna and Sidney Wolk Center for Memory Health, Hebrew Senior Life, Boston, MA, USA g Department of Neurology, Harvard Medical School; Boston, MA, USA A R T I C L E I N F O A B S T R A C T Keywords: Psychosocial hardships associated with the COVID-19 pandemic led many individuals to suffer adverse mental TMS-EEG health consequences, however, others show no negative effects. We hypothesized that the electroencephalo- Mental health graphic (EEG) response to transcranial magnetic stimulation (TMS) could serve as a toy-model of an individual's Resilience capacity to resist psychological stress, in this case linked to the COVID-19 pandemic. We analyzed data from 74 COVID-19 participants who underwent mental health monitoring and concurrent electroencephalography with transcranial magnetic stimulation of the left dorsolateral prefrontal cortex (L-DLPFC) and left inferior parietal lobule (L-IPL). Within the following 19 months, mental health was reassessed at three timepoints during lock-down confinement and different phases of de-escalation in Spain. Compared with participants who remained stable, those who experienced increased mental distress showed, months earlier, significantly larger late EEG responses locally after L-DLPFC stimulation (but not globally nor after L-IPL stimulation). This response, together with years of formal education, was significantly predictive of mental health status during the pandemic. These findings reveal that the effect of TMS perturbation offers a predictive toy model of psychosocial stress response, as exemplified by the COVID-19 pandemic. 1. Introduction depression and anxiety symptoms (Bueno-Notivol et al., 2021). However, whereas some individuals' mental wellbeing will be negatively impacted, The stressors associated with the coronavirus disease (COVID-19) others - on account of their ‘resilience’ - will not be affected, or even pandemic, as well as the restrictions imposed to contain the spread of the thrive in the face of adversity (Pascual-Leone and Bartres-Faz, 2021). virus, are expected to increase the global burden on mental health The concept of resilience is highly heterogeneous with various (Pfefferbaum and North, 2020; Torales et al., 2020). The World Health meanings across different fields of study (Pascual-Leone and Bartres-Faz, Organization has acknowledged this fact (Giacalone et al., 2020) and 2021). Here we use resilience to refer to the processes that enable an highlighted the importance of integrating mental health into the pre- individual to resist the development of illness, mental health problems or paredness and response plans to public health emergencies (WHO, distress when confronted with stressful events or trauma (Moore et al., 2021). Some studies estimate a 25% increase in the general prevalence of 2020; Russo et al., 2012). Conversely, psychological vulnerability is * Corresponding author. ** Corresponding author. E-mail addresses: apleone@hsl.harvard.edu (A. Pascual-Leone), dbartres@ub.edu (D. Bartres-Faz). 1 Drs. Alvaro Pascual-Leone and David Bartres-Faz contributed equally to this manuscript. https://doi.org/10.1016/j.heliyon.2022.e10208 Received 22 March 2022; Received in revised form 24 May 2022; Accepted 3 August 2022 2405-8440/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by- nc-nd/4.0/). R. Perell on-Alfonso et al. Heliyon 8 (2022) e10208 reduced ability to cope with stressors, which would constitute a risk system. In theoretical physics ‘toy-models’ refer to simple models which factor for developing psychopathology (Wright et al., 2013). Rather than nevertheless provide a quantitative explanation and reliable prediction of a dichotomy, resilience and vulnerability can be understood as opposite a given phenomenon (Georgescu, 2012; Marzuoli, 2008). Specifically, as ends of a continuum, which likely reflects the dynamic product of a illustrated in Figure 1, here we use the EEG response to TMS as a ‘toy complex interplay of individual and environmental factors (Cathomas model’ predictive of the eventual (months later) impact of the COVID et al., 2019; Rutter, 2012; Tost et al., 2015), including genetic and de- pandemic and confinement on mental health. We hypothesized that in- mographic characteristics, socio-economic status, developmental cir- dividual differences in the electrophysiological cortical response to single cumstances, access to health care, living conditions, adherence to certain pulse TMS brain perturbation of the left dorsolateral prefrontal cortex, lifestyle factors (e.g. cognitive, physical, nutritional and sleep habits), compared to another control cortical target (i.e., inferior parietal lobule), engagement in emotion-regulation practices such as meditation, social and recorded using EEG, would be predictive of psychological distress relations and support (particularly early in life), and years of education outcomes during the COVID-19 pandemic and confinement. Our findings (Campbell-Sills et al., 2009; Di Marco et al., 2014; Frankish and Horton, contribute to the understanding of biological brain mechanisms of 2017; Gelfo et al., 2018; Livingston et al., 2017). resilience processes and identify a potential target and novel strategy to Evidence from animal models and human neuroimaging studies have promote individual resilience. identified several brain regions and networks that likely play a role in the continuum of resilience and vulnerability. Converging evidence points at 2. Results the crucial roles of anterior cingulate and insular cortices and their connections within the salience network (Menon, 2015; Menon and 2.1. The dynamics of the EEG response to TMS perturbation differentiate Uddin, 2010), as well as limbic structures, such as the amygdala and the individuals eventually found to have a ‘negative’ impact on mental health ventral striatum (Holz et al., 2020). Additionally, the prefrontal cortex status, from those who remained ‘stable’, during the COVID-19 pandemic has been identified as an important structure (Maier and Watkins, 2010). Specifically, prefrontal cortical volume, activation and connectivity with Assessments of mental health using the four-item patient health limbic structures, positively correlate with resilience to traumatic events questionnaire (PHQ-4), an ultra-brief depression and anxiety screening (Bolsinger et al., 2018); and longitudinal studies using functional mag- self-report questionnaire, were obtained prior to the COVID-19 pandemic netic resonance imaging (fMRI) found that children who were better able and up to three additional times during the pandemic. If during all to regulate amygdala's emotional response through recruitment of the timepoints across the confinement, the PHQ-4 score was lower or equal frontoparietal network exhibited greater resilience to developing than before the pandemic outbreak, subjects were classified as ‘stable’ (n depressive symptoms following maltreatment (Rodman et al., 2019). ¼ 32). Conversely, if a given subject had a higher score at any timepoint Prefrontal function is also involved in the pathophysiology and treatment during the pandemic, they were classified as having a ‘negative’ impact of psychiatric conditions, most prominently, major depressive disorder (n ¼ 32). Because not all participants completed stimulation at both (Hare and Duman, 2020) and schizophrenia (Selemon and Zecevic, target locations, the subgroups used in this analysis were actually smaller 2015). Finally, the prefrontal cortex appears to not only play a central for each stimulation target (for L-DLPFC, 23 stable and 25 negative; for L- role in psychological resilience and psychiatric pathophysiology, but has IPL, 22 stable and 23 negative). To make sure that the mental health been also proposed as a hub region for cognitive resilience in normal impact of the pandemic was related to the levels of stress perceived aging (Franzmeier et al., 2017b) as well as to brain atrophy and pa- during the outbreak, we tested for correlation between the average score thology associated with Alzheimer's disease (Franzmeier et al., 2018; of the three pandemic PHQ-4 timepoints, and the scores of the 14-item Neitzel et al., 2019). Therefore, in the present study we focused on the perceived stress scale (Cohen et al., 1983), which was also completed prefrontal cortex to investigate the neural substrate of resilience to by participants during the pandemic, and found a strong positive corre- mental health impact of the COVID-19 pandemic and, argued that single lation (Rs ¼ .69; p < .001), indicating that subjects experiencing more pulse transcranial magnetic stimulation (TMS) in combination with stress during the pandemic also had a larger mental health impact. electroencephalography (EEG) could be used in human experimental Overall, participants had a low to moderate level of perceived stress designs akin to the intervention-based animal studies of the neural sub- during the pandemic (Mdn ¼ 14; range from 2 to 32). Additionally, to strate of resilience. rule out that the groups significantly differed demographically, in resting Animal studies that employ stimuli that can be precisely quantified motor threshold or pre-COVID PHQ-4 score, a multivariate ANOVA was and controlled, for example using tail-shock stress paradigms (Seligman used to compare both groups regarding four independent variables (age, et al., 1975) in which a mild electric shock is given to the tail of rats or years of education, pre-COVID PHQ4 score and resting motor threshold; mice, illustrate the power of such interventional experimental ap- see supplementary table S1 for descriptive statistics), which showed that proaches to gain mechanistic insights into the substrate of resilience. the groups did not differ in any of these variables (F (4, 63) ¼ 3.869, p ¼ Similar approaches in human research combining non-invasive brain 0.130; Wilk's lambda ¼ 0.888, partial η2 ¼ 7.111). stimulation with neuroimaging are possible. For example, Shafi and Point-by-point non-parametric permutation testing (1000 permuta- colleagues (Shafi et al., 2015) have shown that brain responses to TMS tions) with cluster correction for multiple comparisons (Cohen, 2014) on allow identifying abnormal cortical activity patterns before the mani- the TMS-EEG evoked time-series, revealed a single broad cluster (i.e., festation of clinical symptoms in some forms of epilepsy. More recently, 202–269 ms post-stimulus) surviving correction for multiple compari- Abellaneda-P erez and colleagues (Abellaneda-Perez et al., 2019) have sons, only during stimulation of the left dorsolateral prefrontal cortex shown that the default mode network's response profile to intermittent (L-DLPFC) (Figure 2, A). Inspection of the topographical distributions in theta burst stimulation of the inferior parietal lobule can be used to source space for the surviving cluster, confirms that individuals whose predict cognitive decline or maintenance after a three-year follow-up in mental health was negatively impacted had a qualitatively stronger an aging population, well over and above of what baseline neuroimaging frontal activation than those who remained stable (Figure 2, B). There data alone could predict. Furthermore, recent methodological advances were no significant clusters revealed after analysis of the responses to the (Ozdemir et al., 2020, 2021b), have revealed that single pulse TMS can left inferior parietal lobule (L-IPL) control target (Figure 2, C and D), nor be used concurrently with EEG to produce highly specific and reliable for the distributed responses to stimulation on either target (Figure S2). cortical response profiles. Finally, given that only a subsample (i.e., 50% of participants) We propose that TMS-EEG can be a ‘toy model’ of the impact of a completed stimulation of both targets, we additionally run permutation perturbation onto an individual brain and provide a quantitative obser- testing of the local TMS-EEG evoked time-series after DLPFC stimulation, vation of the effect of the controlled external perturbation on brain dy- but including only the participants that had undergone stimulation of namics that can be used to test specific predictions about a complex both targets. The results on this subsample of participants are comparable 2 R. Perell on-Alfonso et al. Heliyon 8 (2022) e10208 Figure 1. Schematic Illustration of the toy-model approach in our study design. MRI-guided single pulse TMS perturbation is used to experimentally model as the stressor, captured by evoked EEG reactivity, and measured at local and distributed (i.e., global) levels. We examined whether this ‘toy-model’ can predict the eventual impact of the COVID-19 on mental health assessed months later. We further hypothesized the stressor would be moderated by demographic and individual factors such as years of formal education. Modified from Pascual-Leone and Bartres-Faz (Pascual-Leone and Bartres-Faz, 2021). to the full sample (Figure S3), since the broad cluster surviving multiple each other, as reveled by the lack of correlation between them (R2 ¼ comparisons correction remains in the same location, albeit slightly .187, p ¼ .167). Subsequently, we tested a reduced model (F (2,53) ¼ shorter in duration (i.e., from 200 to 253 ms after the TMS pulse). 10.5, p < .001, R2adj ¼ .257) retaining only as predictors local brain reactivity (t ¼ 3.66, p < .001) and education (t ¼ -3.40, p ¼ .001). See supplementary table S2 (model “Reduced DLPFC”) for detailed results. 2.2. Local EEG response to TMS perturbation of the left dorsolateral Likelihood ratio test comparing the two models showed that the full prefrontal cortex predicts mental health during the pandemic's lockdown model did not provide a better fit than the reduced one (χ 2 (6) ¼ 4.98, p ¼ confinement .546). The lower Akaik and Bayesian information criteria (AIC and BIC, respectively) values for the reduced model further suggest a better and Without dividing the sample into two groups, a multiple linear more parsimonious fit (AICfull ¼ 79.83, AICreduced ¼ 72.82; BICfull ¼ regression model was fit to determine the potential of TMS evoked EEG 98.06, BICreduced ¼ 78.89). Figure 3 illustrates the linear relationship perturbation of the L-DLPFC to predict mental health outcomes after the between the significant predictors and the response variable in the COVID-19 pandemic outbreak and the strict lock-down confinement reduced model. Analysis of variance of the reduced model revealed that imposed to curb community transmission of the virus. The model's local L-DLPFC reactivity explained 12.79% of total variance in mental response variable was the mean of the total scores for the three PHQ-4 health during the pandemic, while education explained 15.64%. questionnaires, which were completed by participants during the lock- To confirm that our findings were specifically associated with pre- down confinement. Candidate predictors were the local and global frontal reactivity, we fitted a model replacing the predictors for local and (i.e., distributed) brain EEG reactivity to the TMS pulse — recorded global EEG reactivity with those measured when stimulating the L-IPL. before the pandemic outbreak —, as well as their interaction with the The regression model for this control stimulation target did not signifi- stimulation target definition method (i.e., functional or anatomical). cantly predict mental health during the pandemic (F (8,46) ¼ 0.4, p ¼ Additionally, we included age, gender, and years of formal education as .915, R2adj ¼ -.097). See supplementary table S2 (model “Full IPL”) for predictors, because these are demographic and individual factors detailed results. partially predictive of resilience to stress (Campbell-Sills et al., 2009). To demonstrate the specificity of the stimulation itself, a model was Finally, we included as a predictor the number of months before the fitted where we added the local baseline pre-TMS activity as an additional pandemic since each subject underwent TMS-EEG. This was included to predictor to the reduced L-DLPFC model. The resulting model, while still control for the possibility that the amount of time passed from stimula- significant (F (3,47) ¼ 7.09, p < .001, R2adj ¼ .25), revealed that baseline tion to pandemic would have an impact in the prediction. pre-TMS EEG activity did not significantly contribute to predict mental The full linear regression model for the L-DLPFC stimulation target health during the pandemic (t ¼ .67, p ¼ .507). See supplementary table significantly predicted mental health during the pandemic (F (8,47) ¼ S2 (model “Reduced þ Baseline”) for detailed results. 3.1, p ¼ .007, R2adj ¼ .234), and revealed as significant predictors the Finally, to ensure that the results are consistent, even when only local brain reactivity to TMS (t ¼ 3.27, p < .002) and years of formal considering the subsample of participants who completed stimulation of education (t ¼ -2.86, p ¼ .006). See supplementary Table S2 (model “Full both targets, a model was fitted with only this subsample and the local DLPFC”) for detailed results. Both predictors were independent from 3 R. Perell on-Alfonso et al. Heliyon 8 (2022) e10208 Figure 2. Results of permutation testing of the difference between stable and negatively impacted individuals on the TMS evoked EEG time-series. A) shows the significant differences at the local EEG time-series during L-DLPFC stimulation in grey vertical bands, while the vertical green band highlights the cluster surviving correction for multiple comparisons. B) shows the topographical distribution in source space of the response to L-DLPFC stimulation during the green shaded time- window in A, for both groups and their difference. C) depicts the results of permutation testing for the control stimulation of the L-IPL. D) shows the topographical distribution for the same time window as B, for both groups and their difference. Red and blue contours along the plot lines in A and C depict the standard error of the mean. DLPFC, dorsolateral prefrontal cortex; IPL, inferior parietal lobule. Figure 3. Scatter plots illustrating the linear relationship between the predictors and the response variable in the reduced model. Black line in each plot depicts the least squares regression line; shaded grey contours depict 95% confidence intervals. The response variable depicted here in the y-axis is Box-Cox transformed (λ ¼ -0.078). PHQ-4, four item patient health questionnaire; DLPFC, dorsolateral prefrontal cortex; AUC, area under the curve. 4 R. Perell on-Alfonso et al. Heliyon 8 (2022) e10208 response to IPL stimulation added as a predictor to the reduced model. individuals may be reflective of cortical hyperexcitability. Furthermore, The resulting model was still significant (F (3,33) ¼ 3.25; p ¼ .034) and the amplitude of late TMS-EEG responses (¼>180 ms) is related to revealed that, while local DLPFC reactivity (t ¼ 2.44, p ¼ .021) and ed- GABA-B mediated inhibition, as it is significantly reduced after long in- ucation (t ¼ -2.46, p ¼ .019) were significant, local IPL reactivity was not terval intracortical inhibition (de Goede et al., 2020). Therefore, the (t ¼ -1.03, p ¼ .311). See supplementary table S2 (model “Reduced increased amplitude in late EEG responses found in individuals that had a Subsample þ IPL) for detailed results. negative impact, when compared to the ones who remained stable, might reflect a relatively lower intracortical inhibitory capacity and point to 3. Discussion differential levels of activation of parvo-albumin positive cells and integrity of peri-neural nets – main substrates of intracortical We tested the EEG brain reactivity to TMS perturbation as a toy model excitability-inhibition balance (Favuzzi et al., 2017; Xue et al., 2014). of mental health outcomes in the face of the COVID-19 pandemic and This would predict that conditions that alter and disrupt parvo-albumin lockdown confinement. The results show that the local response to TMS positive cells and integrity of peri-neural nets, such as status post trau- perturbation of the left DLPFC — measured months before the pandemic matic brain injury, early stages of Alzheimer's disease, or schizophrenia, outbreak — offers a predictive marker of the future mental health impact would be associated with a loss of resilience and increased vulnerability of the pandemic and confinement. These results serve as a proof of to stressors. Epidemiologic data appear to support such notions (Buckley concept that understanding the TMS pulse as an external transitory insult et al., 2009; Ehrenberg et al., 2018; Hammond et al., 2019). allows quantification of the brain responses and identifies critical and Our results are novel and relevant in advancing our understanding of specific substrates of susceptibility to more complex stressors. At the core the neural mechanisms of resilience. However, this study has some lim- of this ‘toy-model’ is the assumption that single TMS stimuli can be un- itations. We had to conduct the regression analysis on each stimulation derstood as transitory perturbations or insults in themselves. This is target separately due to missing data, which may have hindered statis- supported by evidence showing that the stimuli interfere with ongoing tical power. The regression analysis results would benefit from further brain activity by suddenly injecting an amount of current into the neural validation on a separate independent sample to be able to make reliable circuitry that disrupts ongoing brain activity, resulting in phase resetting predictions of mental health outcomes based on the response to TMS and TMS-evoked perturbation of the ongoing EEG (Rocchi et al., 2018). perturbation. Moreover, due to the limited number of participants rela- Indeed, failure to suppress this perturbation can lead — in the presence of tive to the number of mental health monitoring timepoints during the pathological conditions such as stroke or epilepsy — to a cascading pandemic, we were unable to account for all the mental health trajec- synchronization of neuronal activity that in turn might lead to a seizure tories that have been described in the literature, such as people who have (Kimiskidis, 2019; Kimiskidis et al., 2017). Thus, we can interpret the a negative impact but eventually recover (e.g., Gambin et al., 2021). present results as showing that in the presence of such a brain state Instead, in the permutation analysis, we focused on dichotomizing the disruption, a more resilient brain is better able to tolerate the perturba- two main trajectories (i.e., resilient vs vulnerable). Future studies should tion. A link between the ability of the brain to withstand a targeted attack investigate the neurophysiological signature in response to stimulation of and resilience, has also been proposed by Santarnecchi and colleagues subjects who recover in the medium or long term, and that here were all employing in-silico models (Santarnecchi et al., 2015), but the present classified as vulnerable. Finally, the changes in mental health observed study is the first to offer direct experimental support on a topic of sub- during the pandemic were small overall, with most participants not stantial timely relevance. surpassing clinical screening thresholds for depression and anxiety, Our findings are specific to TMS prefrontal perturbation, because therefore, a sample with a broader range of mental health impact could neither the response to L-IPL stimulation, nor the pre-TMS baseline EEG provide a clearer picture of the neurophysiological determinants of such activity held significant predictive value. Moreover, we show that the impact. However, despite the narrow range of mental health changes, we findings are restricted locally to the stimulated area, because the are still able to show that TMS-EEG can detect a neurophysiological distributed measure of response could not differentiate stable from signature underlying the future differential impact of the pandemic on negatively impacted individuals and did not yield significant predictive mental health. value. Nevertheless, years of formal education was also found to be The presented results are not only relevant as a proof of concept for predictive of mental health during the pandemic, which is unsurprising using intervention-based designs in neuroimaging investigations of the given the well-known epidemiological-level notion that individuals with neural basis of resilience, but also add to the existing evidence of a pri- higher socioeconomic status (encompassing, among others, educational mary role of the left prefrontal cortex in resilience processes (Bolsinger attainment and income) have lower odds of being depressed (Lorant et al., 2018; Dedovic et al., 2009; Franzmeier et al., 2017a, 2017b, et al., 2003), and that education might be a protective factor both against 2017b, 2018, 2017b; Holz et al., 2020; Neitzel et al., 2019; Rodman et al., cognitive as well as emotional vulnerability, by boosting a higher effi- 2019; Stern et al., 2018). This, in turn, singles out the prefrontal cortex as ciency on top-down emotional regulatory processes (Huang et al., 2019). a promising target for interventions aiming to promote positive outcomes In this context, and while education and TMS-EEG reactivity were in- after disrupting events such as the pandemic and associated social re- dependent from one another in this analysis, it is still plausible that both strictions, including the potentially transformative possibility of using reflect prefrontal function, and is therefore, their joint inclusion that best non-invasive stimulation to promote brain resilience by modulating predicted mental health status during the pandemic. prefrontal brain activity. Several of our results and other lines of evidence Dividing the sample into participants who remained stable and those support such potentially transformative therapeutic intervention: (1) the who had a negative change in mental health during the pandemic, known protective role of higher prefrontal function to the deleterious allowed us to directly compare the dynamics of the EEG response to the mental health effects of stress and trauma (Bolsinger et al., 2018; TMS perturbation, revealing that the most discriminative time segment Rodman et al., 2019); (2) the link we have demonstrated between the after TMS perturbation of the local L-DLPFC is the late TMS evoked L-DLPFC response to a brain perturbation and the mental health out- response between 202 ms and 269 ms post-TMS. Interestingly, this occurs comes when facing the stressors associated with the pandemic; (3) our in the vicinity of the commonly found P180 EEG evoked component in finding of exaggerated response to TMS perturbation of the L-DLPFC in response to single TMS pulses of the primary motor cortex (Lioumis et al., individuals that would be negatively impacted by pandemic related 2009). This component has been found to significantly decrease after stress; and (4) the established ability of non-invasive stimulation tech- application of voltage-gated sodium channel blockers, such as lamo- niques to induce long lasting brain plastic changes (Huang et al., 2017). trigine and carbamazepine (Darmani and Ziemann, 2019; Premoli et al., Finally, the identified electrophysiological dynamics in the local DLPFC 2017), indicating that this component reflects cortical excitability. response to TMS perturbation is a potential neurophysiological marker Therefore, we propose that higher amplitudes found in vulnerable that might be useful in a preventive precision medicine framework, when 5 R. Perell on-Alfonso et al. Heliyon 8 (2022) e10208 assessing the potential risk of deleterious mental health impacts for a EEG response to left dorsolateral prefrontal stimulation would be pre- given individual, when exposed to future stressful events such as a new dictive of mental health during the pandemic. Stimulation on the left pandemic. inferior parietal lobule was included in the analysis as a control stimu- lation condition. 4. Methods 4.2. Neuronavigated TMS-EEG 4.1. Study design Transcranial magnetic stimulation was delivered over the left In the present study we analyzed existing data from participants of the dorsolateral prefrontal cortex (L-DLPFC) and the left inferior parietal longitudinal study ‘Barcelona Brain Health Initiative’, BBHI for short lobule (L-IPL). Stimulation was guided by a BrainSight neuronavigation (Cattaneo et al., 2018). In mid-March 2020, during the COVID-19 system (RogueResearch, Inc., Canada). Targets were determined for epidemic, the BBHI launched a longitudinal substudy to investigate the each individual based on either anatomy or the cortical parcellation by mental and brain health impact of societal and personal restrictions Yeo and colleagues (Thomas Yeo et al., 2011). See supplementary imposed by the pandemic (Bartres-Faz et al., 2021; Pascual-Leone et al., materials for MRI acquisition parameters and target determination 2021). For the present report, we selected those BBHI participants who procedures. Stimulation intensity was 120% of resting motor threshold, had undergone concurrent TMS-EEG between July 2018 and February determined as the minimum intensity required to elicit motor evoked 2020, before the COVID-19 pandemic outbreak, as well as mental health potentials in the first dorsal interosseous muscle of the relaxed right monitoring before and during the lockdown using the Patient Health hand, of at least 50μV peak-to-peak, in at least five out of ten trials Questionnaire for Depression and Anxiety (PHQ-4), a standardized (Rossini et al., 2015). For each target, 120 single biphasic pulses were ultra-brief tool for detecting both anxiety and depressive disorders delivered through an MCF-B65 butterfly coil, using a MagPro X100 (Kroenke et al., 2009). The scale was administered at four different stimulator (Magventure, Inc., Denmark), with a random inter-pulse timepoints; one between November 2018 and January 2020, hence interval between four and six seconds. The order of targets was ran- before a mandatory lockdown that was issued by the Spanish Govern- domized for each participant. In order to attenuate auditory evoked ment on March 14th 2020, and another three timepoints during the responses induced by the TMS coil click, participants listened to pandemic, spanning a total of 3 months during the strictest white-noise through earplug-earbuds at their maximum comfortable home-confinement and initial phases of de-escalation (Figure 4). volume. Stimulation was performed concurrently with EEG using a TMS The sample included 74 healthy adults (45 male) ranging from 42 to compatible ActiChamp 64-channel amplifier system, coupled with an 66 years (M ¼ 55.07; SD ¼ 7.1), with a range of years of formal education ActiCap Slim with active electrodes (BrainProducts, GmbH., Germany). from 8 to 28 years (M ¼ 18.01; SD ¼ 3.85). Consistent with the BBHI While the use of active electrodes is relatively novel in the context of general inclusion criteria, none of these individuals reported a medical TMS-EEG, recent research has successfully used them to evaluate TMS diagnosis of any major neuropsychiatric disorder (including mood and evoked brain reactivity (to cite some, Gamboa Arana et al., 2020; anxiety disorders) at study entrance and had normal cognitive function as Ozdemir et al., 2021b, 2021a, 2020; Redondo-Cam os et al., 2022; assessed by comprehensive neuropsychological testing (Cattaneo et al., Rocchi et al., 2021). Moreover, it has been recently shown that the TMS 2018). All participants gave written informed consent, and the local evoked potential waveforms are reliable and comparable to those ob-  ethics committee (Comite d’Etica i Investigaci o Clínica de la Uni o Cata- tained with passive electrodes (Mancuso et al., 2021), provided that lana d'Hospitals) approved the protocols here described and conformed interelectrode impedance is kept low. Therefore, we monitored elec- to the Declaration of Helsinki for research involving human subjects. trode impedance to make sure it was kept under 5kΩ for all electrodes The objective of this analysis was to evaluate the potential of using and throughout the experiment. EEG data was recorded DC to 500Hz the brain response to TMS perturbation – quantified by EEG – as a toy and digitized at a 1KHz sampling rate. While 76% of participants model of the mental health impact of a complex stressor, namely, the completed stimulation of the L-DLPFC and 74% completed stimulation COVID-19 pandemic and confinement, the impact of which was quanti- of the L-IPL, only 50% of participants completed stimulation on both fied with the PHQ-4 questionnaires. Given the known involvement of the targets. For this reason, statistical analysis was conducted separately for prefrontal cortex in various forms of resilience, we hypothesized that the each stimulation target. Figure 4. Timeline of relevant events for the study cohort. According to the Spanish Government state of alarm dictation orders and subsequent de-escalation (https ://www.boe.es/eli/es/rd/2020/03/14/463/con). 1st, 2nd and 3rd indicate the three PHQ-4 based mental health monitoring timepoints during the pandemic. PHQ-4, four item patient health questionnaire; COVID-19, coronavirus diesease 2019; TMS-EEG, transcraneal magnetic stimulation with concurrent electroencephalography. 6 R. Perell on-Alfonso et al. Heliyon 8 (2022) e10208 4.3. Mental health assessment Global response: the global mean field amplitude (GMFA; Lehmann and Skrandies, 1980) of the TMS evoked potentials (TEPs) in sensor Mental health was measured using the PHQ-4, an ultra-brief four item space, was taken as the estimated time-series of the global brain response depression and anxiety screening self-report questionnaire, that consists to the TMS pulse. This measure was computed according to the following of a 2-item depression scale (PHQ-2) and a 2-item anxiety scale (GAD-2). formula: Each subitem scores in the range of 0–6, with combined range from 0 to v(ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi )ffi u Pk u 12. On each subscale a score of 3 or greater is considered positive for t i ðVi ðtÞ–VmeanðtÞ Þ2 GMFAðtÞ ¼ screening purposes. The test was administered a total of four times in an K online format, once before the pandemic and at three timepoints during the confinement and de-escalation. All participants in this analysis Where Vi(t) is the voltage at electrode i at a certain point in time, Vmean(t) completed the pre-pandemic questionnaire and most completed the three is the mean of instantaneous TEP across electrodes, and K is the number additional ones during the pandemic (69%), however, few participants of electrodes. completed only two (23%) or one (8%) of them. For the purposes of Local response: To extract local measures, we first defined a region of quantifying mental health status during the pandemic in the regression interest (ROI) of 100 vertices around each subject's stimulation target models, we used the mean of total scores from the completed question- coordinate, which corresponds to a cortex surface area of approximately naires during the pandemic. 10 cm2. Then the TEP response at the targeted location in source space was taken as the local reactivity time-series. To allow group level sta- 4.4. EEG data preprocessing and analysis tistics, the TEP time-series in source space of each vertex within the target ROI were rectified, averaged together, and then normalized via z-score All EEG data was preprocessed using functions from the EEGLAB transformation: toolbox (Delorme and Makeig, 2004) and the TESA plugin (Mutanen et al., 2020; Rogasch et al., 2017). Source reconstruction and analysis z¼(TEP-μ)/σ was performed using Brainstorm (Tadel et al., 2019) and custom made Where μ is the average of the pre-stimulus baseline (-500 ms to -3 ms) and Matlab (The MathWorks, Inc., USA) scripts. σ is the standard deviation of the baseline. First, the data was segmented around the TMS pulse (1000 to 1000 ms For both global and local TMS response measures, the trapezoidal from the pulse) and baseline corrected (900 ms to -100 ms from the integration from 15 ms to 400 ms post-TMS stimulus was used in the pulse). Then the direct electrical pulse artifact (between -2 ms and 14 ms regression analyses as an estimate of the overall response to TMS from pulse) was zero-padded. Bad channels were then identified via visual perturbation. inspection and removed (range from 0 to 3; M ¼ 0.49, SD ¼ 0.76). Bad epochs were first tagged based on threshold voltage (>100 μV), probability and kurtosis using the inbuilt TESA plugin functions, visual inspection 4.6. Statistical analysis ensured that the epochs were correctly tagged and that no bad epochs were missed, then they were removed from further analysis (range from 0 to 53; Statistical analysis was performed in RStudio (RStudio Team, 2020) M ¼ 20.59; SD ¼ 9.66). A first round of fast independent component and Matlab 2020b. analysis (ICA) was used to reject any remains of the immediate electrical To compare the global and local TMS evoked time-series for each pulse artifact (range from 0 to 3; M ¼ 0.67; SD ¼ 0.65). The zero-padded target we conducted four permutation-based tests. In each test, we pulse artifact was then linearly interpolated, and the data was re- computed the difference of means for each data point within the time- referenced to the average of all channels. Finally, a second round of ICA series time-window of interest (from 15 ms to 400 ms after the TMS was used to reject any other remaining artifacts (e.g., muscle, eye- pulse). In each of the 1000 permutations, the labels for each group movements, heartbeat and others), as well as the somatosensory and (resilient or vulnerable) were scrambled. The resulting p-values were auditory potentials evoked by transcutaneous scalp nerve excitation and adjusted for multiple comparisons using cluster correction (Cohen, coil firing sounds, respectively (range from 21 to 23: M ¼ 28.37; SD ¼ 2014), whereby the size and magnitude of a given cluster of significant 2.77). These are commonplace preprocessing procedures for TMS-EEG data timepoints is considered to survive correction if the size and magnitude and have been described in greater detail elsewhere (Rogasch et al., 2017). of the cluster is above 95% of all cluster sizes and magnitudes discovered The cleaned preprocessed data was then used for source reconstruc- during permutation testing. To test the consistency of the significant tion in Brainstorm. For each subject a forward model was estimated via results on the subsample of participants that completed stimulation on the openMEEG algorithm (Kybic et al., 2005) using the default settings both targets, an additional permutation-based test was run on this sub- (i.e., 3 layers with 1922 vertices each; skull and scalp conductivities of 1 sample. To test the correlation between perceived stress and mental and brain conductivity of 0.0125; adaptative integration), and based on health during the pandemic a Spearman rank correlation was used. each subject's T1 and T2 weighted MRI images and digitized real elec- To investigate the predictive value of TMS reactivity at a local and trode locations, when available (the 29 subjects for which anatomical global levels we used two multiple linear regression models, one for each target determination was used, had no digitized electrode locations, stimulation target (i.e., L-DLPFC and L-IPL). The full model included as therefore, the standard 10-10 electrode locations were used instead). The predictors the local and global TMS brain reactivity measures, age, inverse solution was estimated using the minimum norm imaging gender, and years of education. Additionally, and to control for the method (Salmelin and Baillet, 2009). Sources were then computed as possibility that the target definition method influenced the candidate current density maps for constrained orientations only (i.e., normal to reactivity measures, we included the interaction between the targeting cortex). These are commonplace source reconstruction procedures for method (2 levels: anatomical or functional) and both local and global TMS-EEG data and have been described in greater detail elsewhere reactivity measures, for each model. In each regression model the (Ozdemir et al., 2020, 2021a, 2021b). response variable was each subject's mean of the completed pandemic PHQ-4 scores. In the presence of significant predictors, a reduced model 4.5. Local and distributed EEG measures of the response to TMS including only those was defined and compared against the full model. perturbation To determine the better fitting model, we used the likelihood ratio test and further confirmed the result based on the AIC and BIC values. To To quantify EEG derived brain reactivity measures to the TMS pulse assert the TMS induced specificity of the findings, we fitted an additional we computed the following global (i.e., distributed) and local reactivity model where the possible contribution of the pre-TMS stimulus baseline measures: to the prediction of mental health was tested. Finally, to confirm the 7 R. Perell on-Alfonso et al. Heliyon 8 (2022) e10208 consistency of the results on the subsample of participants that completed Acknowledgements stimulation on both targets, a model was fitted adding the local IPL reactivity as a predictor and including only the subsample participants. Authors thank the Barcelona Brain Health Initiative project partici- Lilliefors test on each model's residuals revealed that they were not pants, whose invaluable contribution made this research possible. We are normally distributed, therefore, we transformed the response variable in thankful for the contribution of Indre Pileckyte to the statistical analysis each model using Box-Cox transformation, resulting in normally in this work, as well as for her review of the manuscript. distributed residuals, therefore, the results reported in this work corre- spond to the regression models with the transformed response variable. References Assumptions of multicollinearity, autocorrelation and heteroscedasticity were met in each model. Abellaneda-Perez, K., Vaque-Alcazar, L., Vidal-Pi~neiro, D., Jannati, A., Solana, E., Bargallo, N., Santarnecchi, E., Pascual-Leone, A., Bartres-Faz, D., 2019. Age-related differences in default-mode network connectivity in response to intermittent theta- Declarations burst stimulation and its relationships with maintained cognition and brain integrity in healthy aging. Neuroimage 188, 794–806. Bartres-Faz, D., Macia, D., Cattaneo, G., Borras, R., Tarrero, C., Solana, J., Tormos, J.M., Author contribution statement Pascual-Leone, A., 2021. The paradoxical effect of COVID-19 outbreak on loneliness. BJ Psych Open 7, 1–4. Ruben Perellon-Alfonso: Conceived and designed the experiments; Bolsinger, J., Seifritz, E., Kleim, B., Manoliu, A., 2018. Neuroimaging correlates of Analyzed and interpreted the data; Contributed analysis tools; Wrote the resilience to traumatic events—a comprehensive review. Front. Psychiatr. 9. Buckley, P.F., Miller, B.J., Lehrer, D.S., Castle, D.J., 2009. Psychiatric comorbidities and paper. schizophrenia. Schizophr. Bull. 35, 383–402. María Redondo-Cam os: Conceived and designed the experiments; Bueno-Notivol, J., Gracia-García, P., Olaya, B., Lasheras, I., L opez-Anton, R., Performed the experiments; Analyzed and interpreted the data. Santabarbara, J., 2021. Prevalence of depression during the COVID-19 outbreak: a meta-analysis of community-based studies. Int. J. Clin. Health Psychol. 21. Kilian Abellaneda-Perez: Conceived and designed the experiments; Campbell-Sills, L., Forde, D.R., Stein, M.B., 2009. Demographic and childhood Interpreted the data; Wrote the paper. environmental predictors of resilience in a community sample. J. Psychiatr. Res. 43, Gabriele Cattaneo, Javier Solana Sanchez: Conceived and designed 1007–1012. Cathomas, F., Murrough, J.W., Nestler, E.J., Han, M.H., Russo, S.J., 2019. Neurobiology the experiments; Interpreted the data. of resilience: interface between mind and body. Biol. Psychiatr. 86, 410–420. Selma Delgado-Gallen, Goretti Espa~ na-Irla: Performed the experiments. Cattaneo, G., Bartres-Faz, D., Morris, T.P., Sanchez, J.S., Macia, D., Tarrero, C., Jose M. Tormos: Conceived and designed the experiments; Contrib- Tormos, J.M., Pascual-Leone, A., 2018. The Barcelona brain health initiative: a cohort study to define and promote determinants of brain health. Front. Aging uted materials and data. Neurosci. 10. Alvaro Pascual-Leone, David Bartres-Faz: Conceived and designed the Cohen, M.X., 2014. Analyzing Neural Time Series Data: Theory and Practice. MIT press. experiments; Interpreted the data; Contributed materials and data; Wrote Cohen, S., Kamarck, T., Mermelstein, R., 1983. A global measure of perceived stress. J. Health Soc. Behav. 24, 385–396. the paper. Darmani, G., Ziemann, U., 2019. Pharmacophysiology of TMS-evoked EEG potentials: a mini-review. Brain Stimul. 12, 829–831. de Goede, A.A., Cumplido-Mayoral, I., van Putten, M.J.A.M., 2020. Funding statement Spatiotemporal dynamics of single and paired pulse TMS-EEG responses. Brain Topogr. 33, 425–437. David Bartres-Faz and Kilian Abellaneda-Perez were supported by Dedovic, K., D’Aguiar, C., Pruessner, J.C., 2009. What stress does to your brain: a review Ministerio de Ciencia, Innovaci on y Universidades de Espa~ na [RTI2018- of neuroimaging studies. Can. J. Psychiatr. 54, 6–15. Delorme, A., Makeig, S., 2004. Eeglab_Jnm03.Pdf. Tech. Soc 134, 9–21. 095181-B-C21]. Di Marco, L.Y., Marzo, A., Mu~ noz-Ruiz, M., Ikram, M.A., Kivipelto, M., Ruefenacht, D., Ruben Perell on-Alfonso was supported by a fellowship from “la Venneri, A., Soininen, H., Wanke, I., Ventikos, Y.A., Frangi, A.F., 2014. Modifiable Caixa” Foundation [ID 100010434. Fellowship code: LCF/BQ/DI19/ lifestyle factors in dementia: a systematic review of longitudinal observational cohort studies. J. Alzheim. Dis. 42, 119–135. 11730050]. Ehrenberg, A.J., Suemoto, C.K., França Resende, E. de P., Petersen, C., Leite, R.E.P., Alvaro Pascual-Leone was supported by the National Institutes of Rodriguez, R.D., Ferretti-Rebustini, R.E. de L., You, M., Oh, J., Nitrini, R., Health, United States [R24AG06142 and P01AG031720]. Pasqualucci, C.A., Jacob-Filho, W., Kramer, J.H., Gatchel, J.R., Grinberg, L.T., 2018. Neuropathologic correlates of psychiatric symptoms in Alzheimer’s disease. The research leading to these results has received funding from “la J. Alzheimers. Dis. 66, 115–126. Caixa” Foundation [LCF/PR/PR16/11110004]. Favuzzi, E., Marques-Smith, A., Deogracias, R., Winterflood, C.M., Sanchez-Aguilera, A., Mantoan, L., Maeso, P., Fernandes, C., Ewers, H., Rico, B., 2017. Activity-Dependent gating of parvalbumin interneuron function by the perineuronal net protein brevican. Data availability statement Neuron 95, 639–655.e10. Frankish, H., Horton, R., 2017. Prevention and management of dementia: a priority for public health. Lancet 390, 2614–2615. The authors encourage interested investigators to reach out and we Franzmeier, N., G€  ottler, J., Grimmer, T., Drzezga, A., Araque-Caballero, M.A., Simon- will honor all reasonable and scientifically motivated requests for data Vermot, L., Taylor, A.N.W., Bürger, K., Catak, C., Janowitz, D., Müller, C., access and make the raw data available when required. Duering, M., Sorg, C., Ewers, M., 2017a. Resting-state connectivity of the left frontal cortex to the default mode and dorsal attention network supports reserve in mild cognitive impairment. Front. Aging Neurosci. 9, 1–11. Declaration of interest's statement Franzmeier, N., Hartmann, J., Taylor, A.N.W., Araque-Caballero, M., Simon-Vermot, L., Kambeitz-Ilankovic, L., Bürger, K., Catak, C., Janowitz, D., Müller, C., Ertl- Wagner, B., Stahl, R., Dichgans, M., Duering, M., Ewers, M., 2018. The left frontal The authors declare the following conflict of interests: Dr. A. Pascual- cortex supports reserve in aging by enhancing functional network efficiency Rik Leone is listed as an inventor on several issued and pending patents on Ossenkoppele. Alzheimer's Res. Ther. 10, 1–12. Franzmeier, N., Hartmann, J.C., Taylor, A.N.W., Caballero, M.A.A.,  Simon-Vermot, L., the real-time integration of noninvasive brain stimulation with electro- Buerger, K., Kambeitz-Ilankovic, L.M., Ertl-Wagner, B., Mueller, C., Catak, C., encephalography and magnetic resonance imaging. He is co-founder of Janowitz, D., Stahl, R., Dichgans, M., Duering, M., Ewers, M., 2017b. Left frontal hub Linus Health and TI Solutions AG; serves on the scientific advisory boards connectivity during memory performance supports reserve in aging and mild for Starlab Neuroscience, Magstim Inc., Radiant Hearts, and MedR- cognitive impairment. J. Alzheim. Dis. 59, 1381–1392. Gambin, M., Oleksy, T., Sȩkowski, M., Wnuk, A., Wozniak-Prus, M., Kmita, G. 1/4yna, hythms, and is an Associate Editor for Annals of Neurology. None of the Holas, P., Pisula, E., Łojek, E., Hansen, K., Gorgol, J., Kubicka, K., Huflejt- other authors reports any conflict of interest. ŁUkasik, M., Cudo, A., Łys, A.E., Szczepaniak, A., Bonanno, G.A., 2021. Pandemic trajectories of depressive and anxiety symptoms and their predictors: five-wave study during the COVID-19 pandemic in Poland. Psychol. Med. 19, 1–3. Additional information Gamboa Arana, O.L., Palmer, H., Dannhauer, M., Hile, C., Liu, S., Hamdan, R., Brito, A., Cabeza, R., Davis, S.W., Peterchev, A.V., Sommer, M.A., Appelbaum, L.G., 2020. Intensity- and timing-dependent modulation of motion perception with transcranial Supplementary content related to this article has been published magnetic stimulation of visual cortex. Neuropsychologia 147, 107581. online at https://doi.org/10.1016/j.heliyon.2022.e10208. 8 R. Perell on-Alfonso et al. Heliyon 8 (2022) e10208 Gelfo, F., Mandolesi, L., Serra, L., Sorrentino, G., Caltagirone, C., 2018. The Pfefferbaum, B., North, C.S., 2020. Mental health and the covid-19 pandemic. N. Engl. J. neuroprotective effects of experience on cognitive functions: evidence from animal Med. 383, 510–512. studies on the neurobiological bases of brain reserve. Neuroscience 370, 218–235. Premoli, I., Biondi, A., Carlesso, S., Rivolta, D., Richardson, M.P., 2017. Lamotrigine and Georgescu, I., 2012. Quantum simulation: Toy model. Nat. Phys. 8, 444. levetiracetam exert a similar modulation of TMS-evoked EEG potentials. Epilepsia 58, Giacalone, A., Rocco, G., Ruberti, E., 2020. Physical health and psychosocial 42–50. considerations during the coronavirus disease 2019 outbreak. Psychosomatics 61, Redondo-Cam os, M., Cattaneo, G., Perellon-Alfonso, R., Alviarez-Schulze, V., Morris, T.P., 851–852. Solana-Sanchez, J., Espa~ na-Irla, G., Delgado-Gallen, S., Pach on-García, C., Albu, S., Hammond, F.M., Corrigan, J.D., Ketchum, J.M., Malec, J.F., Dams-O’Connor, K., Hart, T., Zetterberg, H., Tormos, J.M., Pascual-Leone, A., Bartres-Faz, D., 2022. Local Novack, T.A., Bogner, J., Dahdah, M.N., Whiteneck, G.G., 2019. Prevalence of prefrontal cortex TMS-induced reactivity is related to working memory and reasoning medical and psychiatric comorbidities following traumatic brain injury. J. Head in middle-aged adults. Front. Psychol. 13, 1–10. Trauma Rehabil. 34, E1–E10. Rocchi, L., Di Santo, A., Brown, K., Iban ~ ez, J., Casula, E., Rawji, V., Di Lazzaro, V., Hare, B.D., Duman, R.S., 2020. Prefrontal cortex circuits in depression and anxiety: Koch, G., Rothwell, J., 2021. Disentangling EEG responses to TMS due to cortical and contribution of discrete neuronal populations and target regions. Mol. Psychiatr. 25, peripheral activations. Brain Stimul. 14, 4–18. 2742–2758. Rocchi, L., Iban~ez, J., Benussi, A., Hannah, R., Rawji, V., Casula, E., Rothwell, J., 2018. Holz, N.E., Tost, H., Meyer-Lindenberg, A., 2020. Resilience and the brain: a key role for Variability and predictors of response to continuous theta burst stimulation: a TMS- regulatory circuits linked to social stress and support. Mol. Psychiatr. 25, 379–396. EEG study. Front. Neurosci. 12, 1–11. Huang, C.M., Fan, Y.T., Lee, S.H., Liu, H.L., Chen, Y.L., Lin, C., Lee, T.M.C., 2019. Rodman, A.M., Jenness, J.L., Weissman, D.G., Pine, D.S., McLaughlin, K.A., 2019. Cognitive reserve-mediated neural modulation of emotional control and regulation in Neurobiological markers of resilience to depression following childhood people with late-life depression. Soc. Cognit. Affect Neurosci. 14, 849–860. maltreatment: the role of neural circuits supporting the cognitive control of emotion. Huang, Y.Z., Lu, M.K., Antal, A., Classen, J., Nitsche, M., Ziemann, U., Ridding, M., Biol. Psychiatr. 86, 464–473. Hamada, M., Ugawa, Y., Jaberzadeh, S., Suppa, A., Paulus, W., Rothwell, J., 2017. Rogasch, N.C., Sullivan, C., Thomson, R.H., Rose, N.S., Bailey, N.W., Fitzgerald, P.B., Plasticity induced by non-invasive transcranial brain stimulation: a position paper. Farzan, F., Hernandez-pavon, J.C., 2017. NeuroImage Analysing concurrent Clin. Neurophysiol. 128, 2318–2329. transcranial magnetic stimulation and electroencephalographic data : a review and Kimiskidis, V., 2019. The dynamics of TMS-induced seizures and epileptiform discharges. introduction to the open-source TESA software. Neuroimage 147, 934–951. Brain Stimul. 12, 484. Rossini, P.M., Burke, D., Chen, R., Cohen, L.G., Daskalakis, Z., Di Iorio, R., Di Lazzaro, V., Kimiskidis, V.K., Tsimpiris, A., Ryvlin, P., Kalviainen, R., Koutroumanidis, M., Valentin, A., Ferreri, F., Fitzgerald, P.B., George, M.S., Hallett, M., Lefaucheur, J.P., Langguth, B., Laskaris, N., Kugiumtzis, D., 2017. TMS combined with EEG in genetic generalized Matsumoto, H., Miniussi, C., Nitsche, M.A., Pascual-leone, A., Paulus, W., Rossi, S., epilepsy: a phase II diagnostic accuracy study. Clin. Neurophysiol. 128, 367–381. Rothwell, J.C., Siebner, H.R., Ugawa, Y., Walsh, V., Ziemann, U., Committee, I.F.C.N., Kroenke, K., Spitzer, R.L., Williams, J.B.W., L€ owe, B., 2009. An ultra-brief screening scale Rossini, P.M., Burke, D., Chen, R., Cohen, L.G., Daskalakis, Z., Iorio, R., Di for anxiety and depression: the PHQ–4. Psychosomatics 50, 613–621. Lazzaro, V., Di Ferreri, F., Fitzgerald, P.B., George, M.S., Hallett, M., Lefaucheur, J.P., Kybic, J., Clerc, M., Abboud, T., Faugeras, O., Keriven, R., Papadopoulo, T., 2005. Langguth, B., Matsumoto, H., Miniussi, C., Nitsche, M.A., Pascual-leone, A., A common formalism for the integral formulations of the forward EEG problem. IEEE Paulus, W., Rossi, S., Rothwell, J.C., Siebner, H.R., 2015. Clinical Neurophysiology Trans. Med. Imag. 24, 12–28. Non-invasive electrical and magnetic stimulation of the brain , spinal cord , roots and Lehmann, D., Skrandies, W., 1980. Reference-free identification of components of peripheral nerves : basic principles and procedures for routine clinical and research checkerboard-evoked multichannel potential fields. Electroencephalogr. Clin. application . An updated report from an. Clin. Neurophysiol. 126, 1071–1107. Neurophysiol. 48, 609–621. RStudio Team, 2020. RStudio. Integrated Development Environment for R. Lioumis, P., Kicic, D., Savolainen, P., M€akel€a, J.P., K€ahk€onen, S., 2009. Reproducibility of Russo, S.J., Murrough, J.W., Han, M., Charney, D.S., Nestler, E.J., 2012. Review TMS - evoked EEG responses. Hum. Brain Mapp. 30, 1387–1396. Neurobiology of resilience. Nat. Neurosci. 15, 1475–1484. Livingston, G., Sommerlad, A., Orgeta, V., Costafreda, S.G., Huntley, J., Ames, D., Rutter, M., 2012. Resilience as a dynamic concept. Dev. Psychopathol. 24, 335–344. Ballard, C.G., Banerjee, S., Burns, A., Cohen-Mansfield, J., Cooper, C., Fox, N., Salmelin, R., Baillet, S., 2009. Electromagnetic brain imaging. Hum. Brain Mapp. 30, Gitlin, L.N., Howard, R., Kales, H.C., Larson, E., Ritchie, K., Rockwood, K., 1753–1757. Sampson, E.L., Samus, Q.M., Schneider, L.S., Selbæk, G., Teri, L., Mukadam, N., 2017. Santarnecchi, E., Rossi, S., Rossi, A., 2015. The smarter, the stronger: intelligence level The lancet international commission on dementia prevention and care. Lancet 390, correlates with brain resilience to systematic insults. Cortex 64, 293–309. 2673–2734. Selemon, L.D., Zecevic, N., 2015. Schizophrenia: a tale of two critical periods for Lorant, V., Deliege, D., Eaton, W., Robert, A., Philippot, P., Ansseau, M., 2003. prefrontal cortical development. Transl. Psychiatry 5. Socioeconomic inequalities in depression: a meta-analysis. Am. J. Epidemiol. 157, Seligman, M.E., Rosellini, R.A., Kozak, M.J., 1975. Learned helplessness in the rat: time 98–112. course, immunization, and reversibility. J. Comp. Physiol. Psychol. 88, 542–547. Maier, S.F., Watkins, L.R., 2010. Role of the medial prefrontal cortex in coping and Shafi, M.M., Vernet, M., Klooster, D., Chu, C.J., Boric, K., Barnard, M.E., Romatoski, K., resilience. Brain Res. 1355, 52–60. Westover, M.B., Christodoulou, J.A., Gabrieli, J.D.E., Whitfield-Gabrieli, S., Pascual- Mancuso, M., Sveva, V., Cruciani, A., Brown, K., Iban ~ ez, J., Rawji, V., Casula, E., Leone, A., Chang, B.S., 2015. Physiological consequences of abnormal connectivity in Premoli, I., D’Ambrosio, S., Rothwell, J., Rocchi, L., 2021. Transcranial evoked a developmental epilepsy. Ann. Neurol. 77, 487–503. potentials can be reliably recorded with active electrodes. Brain Sci. 11, 1–16. Stern, Y., Arenaza-Urquijo, E.M., Bartres-Faz, D., Belleville, S., Cantilon, M., Chetelat, G., Marzuoli, A., 2008. Toy models in physics and the reasonable effectiveness of Ewers, M., Franzmeier, N., Kempermann, G., Kremen, W.S., Okonkwo, O., mathematics. In: Deduction, Computation, Experiment. Springer, pp. 49–64. Scarmeas, N., Soldan, A., Udeh-Momoh, C., Valenzuela, M., Vemuri, P., Menon, V., 2015. Salience Network, Brain Mapping: an Encyclopedic Reference. Elsevier Vuoksimaa, E., Arenaza Urquiljo, E.M., Bartres-Faz, D., Belleville, S., Cantillon, M., Inc. Chetelat, G., Clouston, S.A.P., Estanga, A., Ewers, M., Franzmeier, N., Gold, B., Menon, V., Uddin, L.Q., 2010. Saliency, switching, attention and control: a network Habeck, C., Jones, R., Kempermann, G., Kochhann, R., Kremen, W., Lim, Y.Y., model of insula function. Brain Struct. Funct. 214, 655–667. Martínez-Lage, P., Morbelli, S., Okonkwo, O., Ossenkoppele, R., Pettigrew, C., Moore, T.M., White, L.K., Barzilay, R., Calkins, M.E., Jones, J.D., Young, J.F., Gur, R.C., Rosen, A.C., Scarmeas, N., Soldan, A., Song, X., Udeh-Momoh, C., Stern, Y., Gur, R.E., 2020. Development of a scale battery for rapid assessment of risk and Valenzuela, M., Van Loenhoud, A.C., Vemuri, P., Vuoksimaa, E., 2018. Whitepaper: resilience. Psychiatr. Res. 288, 112996. defining and investigating cognitive reserve, brain reserve, and brain maintenance. Mutanen, T.P., Biabani, M., Sarvas, J., Ilmoniemi, R.J., Rogasch, N.C., 2020. Source-based Alzheimer's Dementia 1–7. artifact-rejection techniques available in TESA, an open-source TMS–EEG toolbox. Tadel, F., Bock, E., Niso, G., Mosher, J.C., Cousineau, M., Pantazis, D., Leahy, R.M., Brain Stimul. 13, 1349–1351. Baillet, S., 2019. MEG/EEG group analysis with brainstorm. Front. Neurosci. 13, Neitzel, J., Franzmeier, N., Rubinski, A., Ewers, M., 2019. Left frontal connectivity 1–21. attenuates the adverse effect of entorhinal tau pathology on memory. Neurology 93, Thomas Yeo, B.T., Krienen, F.M., Sepulcre, J., Sabuncu, M.R., Lashkari, D., E347–E357. Hollinshead, M., Roffman, J.L., Smoller, J.W., Z€ ollei, L., Polimeni, J.R., Fisch, B., Ozdemir, R.A., Boucher, P., Fried, P.J., Momi, D., Jannati, A., Pascual-Leone, A., Liu, H., Buckner, R.L., 2011. The organization of the human cerebral cortex estimated Santarnecchi, E., Shafi, M.M., 2021a. Reproducibility of cortical response modulation by intrinsic functional connectivity. J. Neurophysiol. 106, 1125–1165. induced by intermittent and continuous theta-burst stimulation of the human motor Torales, J., O’Higgins, M., Castaldelli-Maia, J.M., Ventriglio, A., 2020. The outbreak of cortex. Brain Stimul. 14, 949–964. COVID-19 coronavirus and its impact on global mental health. Int. J. Soc. Psychiatr. Ozdemir, R.A., Tadayon, E., Boucher, P., Momi, D., Karakhanyan, K.A., Fox, M.D., 66, 317–320. Halko, M.A., Pascual-Leone, A., Shafi, M.M., Santarnecchi, E., 2020. Individualized Tost, H., Champagne, F.A., Meyer-Lindenberg, A., 2015. Environmental influence in the perturbation of the human connectome reveals reproducible biomarkers of network brain, human welfare and mental health. Nat. Neurosci. 18, 4121–4131. dynamics relevant to cognition. Proc. Natl. Acad. Sci. U. S. A. 117, 8115–8125. WHO. WHO Executive Board stresses need for improved response to mental health impact Ozdemir, R.A., Tadayon, E., Boucher, P., Sun, H., Momi, D., Ganglberger, W., of public health emergencies [WWW Document]. URL. https://www.who.int/new Westover, M.B., Pascual-Leone, A., Santarnecchi, E., Shafi, M.M., 2021b. Cortical s/item/11-02-2021-who-executive-board-stresses-need-for-improved-response-to-me responses to noninvasive perturbations enable individual brain fingerprinting. Brain ntal-health-impact-of-public-health-emergencies. (Accessed 23 February 2021). Stimul. 14. Wright, M.O., Masten, A.S., Narayan, A.J., 2013. Resilience processes in development: Pascual-Leone, A., Bartres-Faz, D., 2021. Human brain resilience: a call to action. Ann. four waves of research on positive adaptation in the context of adversity. In: Neurol. 1–14. Goldstein, S., Brooks, R.B. (Eds.), Handbook of Resilience in Children. Springer US, Pascual-Leone, A., Cattaneo, G., Macia, D., Solana, J., Tormos, J.M., Bartres-Faz, D., 2021. Boston, MA, pp. 15–37. Beware of optimism bias in the context of the COVID-19 pandemic. Ann. Neurol. 11, Xue, M., Atallah, B.V., Scanziani, M., 2014. Equalizing excitation-inhibition ratios across 2019–2021. visual cortical neurons. Nature 511, 596–600. 9