Transcranial Direct Current Stimulation (tDCS) is a non-invasive neuromodulation technique with a wide variety of applications in both the clinical and cognitive psychology domains. As increasingly acknowledged, its effectiveness is subject dependent, which may lead to timely and costly treatments with ineffective results if this variability is not taken into account. We propose the usage of electroencephalography (EEG) for the analysis and prediction of individual responses to tDCS. In this context the application of machine learning can be of enormous help. We analysed resting-state EEG activity to identify subgroups of participants with an homogeneous electrophysiological profile and their response to different tDCS interventions. The study described herein, which focuses on healthy controls, was conducted within a clinical trial for the development of treatments based on tDCS for age-matched children diagnosed with Attention Deficit Hyperactivity Disorder (ADHD) and Autism Spectrum Disorder (ASD). We have studied a randomized, double-blind, sham-controlled tDCS intervention in 56 healthy children and adolescents aged 10-17, applied in 2 parallel groups over 2 target regions, namely left Dorsolateral Prefrontal Cortex (lDLPFC) and right Inferior Frontal Gyrus (rIFG). Cognitive behavioural tasks were used to both activate particular brain areas during the stimulation and to assess the impact of the intervention afterwards. We have implemented an unsupervised learning approach to stratify participants based on their resting-state EEG spectral features before the tDCS application. We have then applied a correlational analysis to identify EEG profiles associated with tDCS subject response to the specific stimulation sites and the presence or not of concurrent tasks during the intervention. In the results we found specific digital electrophysiological profiles that can be associated to a positive response, whereas subjects with other profiles respond negatively or do not respond to the intervention. Findings suggest that unsupervised machine learning procedures, when associated with proper visualization features, can be successfully used to interpret and eventually to predict responses of individuals to tDCS treatment.

Contemplative neuroscience has increasingly explored meditation using neuroimaging. However, the brain mechanisms underlying meditation remain elusive. Here, we implemented a causal mechanistic framework to explore the spatiotemporal dynamics of expert meditators during meditation and rest. We first applied a model-free approach by defining a probabilistic metastable substate (PMS) space for each state, consisting of different probabilities of occurrence from a repertoire of dynamic patterns. Different brain signatures were mainly found in the triple-network model (i.e., the executive control, salience, and default-mode networks). Moreover, we implemented a model-based approach by adjusting the PMS of the resting state to a whole-brain model, which enabled us to explorein silicoperturbations to transition to the meditation state. Consequently, we assessed the sensitivity of different brain areas regarding their perturbability and their mechanistic local-global effects. Using a synchronous protocol, we successfully transitioned from the resting state to the meditative state by shifting areas mainly from the somatomotor and dorsal attention networks. Overall, our work reveals distinct whole-brain dynamics in meditation compared to rest, and how the meditation state can be induced with localized artificial perturbations. It motivates future work regarding meditation as a practice in health and as a potential therapy for brain disorders.

No abstract

Transcranial Direct Current Stimulation (tDCS) is a non-invasive neuromodulation technique with a wide variety of clinical and research applications. As increasingly acknowledged, its effectiveness is subject dependent, which may lead to time consuming and cost ineffective treatment development phases. We propose the combination of electroencephalography (EEG) and unsupervised learning for the stratification and prediction of individual responses to tDCS. A randomized, sham-controlled, double-blind crossover study design was conducted within a clinical trial for the development of pediatric treatments based on tDCS. The tDCS stimulation (sham and active) was applied either in the left dorsolateral prefrontal cortex or in the right inferior frontal gyrus. Following the stimulation session, participants performed 3 cognitive tasks to assess the response to the intervention: the Flanker Task, N-Back Task and Continuous Performance Test (CPT). We used data from 56 healthy children and adolescents to implement an unsupervised clustering approach that stratify participants based on their resting-state EEG spectral features before the tDCS intervention. We then applied a correlational analysis to characterize the clusters of EEG profiles in terms of participant’s difference in the behavioral outcome (accuracy and response time) of the cognitive tasks when performed after a tDCS-sham or a tDCS-active session. Better behavioral performance following the active tDCS session compared to the sham tDCS session is considered a positive intervention response, whilst the reverse is considered a negative one. Optimal results in terms of validity measures was obtained for 4 clusters. These results show that specific EEG-based digital phenotypes can be associated to particular responses. While one cluster presents neurotypical EEG activity, the remaining clusters present non-typical EEG characteristics, which seem to be associated with a positive response. Findings suggest that unsupervised machine learning can be successfully used to stratify and eventually predict responses of individuals to a tDCS treatment.