No abstract

The present method may be used to improve epileptogenic identification as well as pinpoint additional regions that are functionally altered during ictal events.

We report a patient presenting drug‐resistant, non‐dominant temporal lobe epilepsy with ictal spitting and prosopometamorphopsia, both extremely rare semiologies. Second‐phase pre‐surgical monitoring was performed using SEEG due to lesion‐negative imaging and the rare semiology. The seizure onset zone was delimited to the right anterior hippocampus and the temporobasal cortex, with the propagation zone within the entorhinal cortex. Interestingly, direct electrical stimulation to the entorhinal cortex, which was reproduced in a number of trials, evoked spitting without leading to seizures or post‐discharges. After the resection of the epileptogenic zone, the patient remained seizure‐free without AEDs for a follow‐up period of five years (Engel Class 1a). The neuropathology revealed a focal cortical dysplasia type FCD‐Ia. Spectral analysis of intracranial ictal EEG (iEEG) data suggested a possible role of the basal temporal and entorhinal cortex as a necessary node in ictal spitting. [Published with video sequences on http://www.epilepticdisorders.com].

The methodology developed in this study relies on finding the frequencies and time windows of interest where ictal activity is maximal with respect to a baseline pre-ictal period, while being spatially confined to a few contacts. Central to this approach is the definition of two novel measures, namely the global activation (GA) and the activation entropy (AE), that are used to monitor the magnitude of spectral changes with respect to the pre-ictal epoch and the spread of these spectral activations, respectively, at different frequencies and as time progresses from seizure onset. By setting appropriate conditions on the two measures it is possible to find time-frequency windows of interest where SOZ regions can be optimally discriminated. These windows will be closely linked to the spectral properties of each patient's electrophysiological seizure onset patterns and will therefore be called seizure onset windows (SOW).Our method makes use of the mean activation (MA), a measure that was proposed in a recent study (Vila-Vidal et al., 2017) to quantify average spectral activations of targeted brain structures in a given frequency and time window of interest. In this section, we first review the methodological procedure that leads to the computation of each region's MA for a given time-frequency window. Then, we define the two core measures of the study, the global activation and the activation entropy, and discuss some of their fundamental properties. Based on these measures, in the last section we show how to find the seizure onset windows (SOW) and how to delineate the seizure onset zone (SOZ). Mean activation (MA) in a time-frequency window of interestThe mean activation (MA) measures the average spectral activation of each targeted brain structure for pre-defined frequency and time windows of interest (Vila-Vidal et al., 2017). MAs are estimated from SEEG signals with a two-step procedure. Fig. 1C illustrates how the signal is processed to obtain the MAs in a given time-frequency window of interest. First, time-dependent spectral activations in the frequency of interest are found using the Hilbert transform method. All contacts' signals are band-pass filtered in non-overlapping logarithmically spaced narrow frequency bands [f, f + ∆f ] (we used ∆f = 0.1) covering the whole frequency of interest. Signal power in each narrow band is obtained by squaring the signal envelope (modulus of the analytical signal) and summation over all narrow bands is performed to obtain the time-dependent power of each region's SEEG signal in the desired frequency of interest. The resulting values are z-scored with respect to a baseline distribution defined by the power values of all contacts during the first 40 seconds 1

Schizophrenia is a debilitating neuropsychiatric disorder whose underlying correlates remain unclear despite decades of neuroimaging investigation. One contentious topic concerns the role of global signal fluctuations and how they affect more focal functional changes. Moreover, it has been difficult to pinpoint causal mechanisms of circuit disruption. Here we analysed resting-state fMRI data from 47 schizophrenia patients and 118 age-matched healthy controls and used dynamical analyses to investigate how global fluctuations and other functional metastable states are affected by this disorder. We then used in-silico perturbation of a whole-brain model to identify critical areas involved in the disease. We found that brain dynamics in the schizophrenic group were characterised by an increased probability of globally coherent states and reduced recurrence of a substate dominated by coupled activity in the default mode and limbic networks. Perturbing a set of temporoparietal sensory and associative areas in a model of the healthy brain reproduced global pathological dynamics. Healthy brain dynamics were instead restored by perturbing a set of medial fronto-temporal and cingulate regions in the model of pathology. These results highlight the relevance of global signal alterations in schizophrenia and identify a set of vulnerable areas involved in determining a shift in brain state.

No abstract

We present a novel set of 200 Western tonal musical stimuli (MUST) to be used in research on perception and appreciation of music. It consists of four subsets of 50 stimuli varying in balance, contour, symmetry, or complexity. All are 4 s long and designed to be musically appealing and experimentally controlled. We assessed them behaviorally and computationally. The behavioral assessment (Study 1) aimed to determine whether musically untrained participants could identify variations in each attribute. Forty-three participants rated the stimuli in each subset on the corresponding attribute. We found that inter-rater reliability was high and that the ratings mirrored the design features well. Participants’ ratings also served to create an abridged set of 24 stimuli per subset. The computational assessment (Study 2) required the development of a specific battery of computational measures describing the structural properties of each stimulus. We distilled nonredundant composite measures for each attribute and examined whether they predicted participants’ ratings. Our results show that the composite measures indeed predicted participants’ ratings. Moreover, the composite complexity measure predicted complexity ratings at least as well as existing models of musical complexity. We conclude that the four subsets are suitable for use in studies that require presenting participants with short musical motifs varying in balance, contour, symmetry, or complexity, and that the stimuli and the computational measures are valuable resources for research in music psychology, empirical aesthetics, music information retrieval, and musicology. The MUST set and MATLAB toolbox codifying the computational measures are freely available at osf.io/bfxz7.

22The spatial mapping of localized events in brain activity critically depends on the correct 23 identification of the pattern signatures associated with those events. For instance, in the 24 context of epilepsy research, a number of different electrophysiological patterns have been 25 associated with epileptogenic activity. Motivated by the need to define automated seizure 26 focus detectors, we propose a novel data-driven algorithm for the spatial identification of 27 localized events that is based on the following rationale: the distribution of emerging 28 oscillations during confined events across all recording sites is highly non-uniform and can be 29 mapped using a spatial entropy function. By applying this principle to EEG recording 30 obtained from 67 distinct seizure epochs, our method successfully identified the seizure focus 31 on a group of ten drug-resistant temporal lobe epilepsy patients (average sensitivity: 0.94, 32 average specificity: 0.90) together with its characteristic electrophysiological pattern 33 signature. Cross-validation of the method outputs with postresective information revealed the 34 consistency of our findings in long follow-up seizure-free patients. Overall, our methodology 35 provides a reliable computational procedure that might be used as in both experimental and 36 clinical domains to identify the neural populations undergoing an emerging functional or 37 pathological transition. 38 39 Keywords 40 Seizure onset zone, intracranial EEG, time-frequency analysis, automated detection 41 algorithms, post-operative outcome 42 43 Acknowledgments: 44