EEG functional connectivity as a Riemannian mediator: An application to malnutrition and cognition.

Mediation analysis assesses whether an exposure directly produces changes in cognitive behavior or is influenced by intermediate "mediators". Electroencephalographic (EEG) spectral measurements have been previously used as effective mediators representing diverse aspects of brain function. However, it has been necessary to collapse EEG measures onto a single scalar using standard mediation methods. In this article, we overcome this limitation and examine EEG frequency-resolved functional connectivity measures as a mediator using the full EEG cross-spectral tensor (CST). Since CST samples do not exist in Euclidean space but in the Riemannian manifold of positive-definite tensors, we transform the problem, allowing for the use of classic multivariate statistics. Toward this end, we map the data from the original manifold space to the Euclidean tangent space, eliminating redundant information to conform to a "compressed CST." The resulting object is a matrix with rows corresponding to frequencies and columns to cross spectra between channels. We have developed a novel matrix mediation approach that leverages a nuclear norm regularization to determine the matrix-valued regression parameters. Furthermore, we introduced a global test for the overall CST mediation and a test to determine specific channels and frequencies driving the mediation. We validated the method through simulations and applied it to our well-studied 50+-year Barbados Nutrition Study dataset by comparing EEGs collected in school-age children (5-11 years) who were malnourished in the first year of life with those of healthy classmate controls. We hypothesized that the CST mediates the effect of malnutrition on cognitive performance. We can now explicitly pinpoint the frequencies (delta, theta, alpha, and beta bands) and regions (frontal, central, and occipital) in which functional connectivity was altered in previously malnourished children, an improvement to prior studies. Understanding the specific networks impacted by a history of postnatal malnutrition could pave the way for developing more targeted and personalized therapeutic interventions. Our methods offer a versatile framework applicable to mediation studies encompassing matrix and Hermitian 3D tensor mediators alongside scalar exposures and outcomes, facilitating comprehensive analyses across diverse research domains.

Clinical and Electrophysiological Differences between Subjects with Dysphonetic Dyslexia and Non-Specific Reading Delay.

Reading is essentially a two-channel function, requiring the integration of intact visual and auditory processes both peripheral and central. It is essential for normal reading that these component processes go forward automatically. Based on this model, Boder described three main subtypes of dyslexia: dysphonetic dyslexia (DD), dyseidetic, mixed and besides a fourth group defined non-specific reading delay (NSRD). The subtypes are identified by an algorithm that considers the reading quotient and the % of errors in the spelling test. Chiarenza and Bindelli have developed the Direct Test of Reading and Spelling (DTRS), a computerized, modified and validated version to the Italian language of the Boder test. The sample consisted of 169 subjects with DD and 36 children with NSRD. The diagnosis of dyslexia was made according to the DSM-V criteria. The DTRS was used to identify the dyslexia subtypes and the NSRD group. 2⁻5 min of artefact-free EEG (electroencephalogram), recorded at rest with eyes closed, according to 10⁻20 system were analyzed. Stability based Biomarkers identification methodology was applied to the DTRS and the quantitative EEG (QEEG). The reading quotients and the errors of the reading and spelling test were significantly different in the two groups. The DD group had significantly higher activity in delta and theta bands compared to NSRD group in the frontal, central and parietal areas bilaterally. The classification equation for the QEEG, both at the scalp and the sources levels, obtained an area under the robust Receiver Operating Curve (ROC) of 0.73. However, we obtained a discrimination equation for the DTRS items which did not participate in the Boder classification algorithm, with a specificity and sensitivity of 0.94 to discriminate DD from NSRD. These results demonstrate for the first time the existence of different neuropsychological and neurophysiological patterns between children with DD and children with NSRD. They may also provide clinicians and therapists warning signals deriving from the anamnesis and the results of the DTRS that should lead to an earlier diagnosis of reading delay, which is usually very late diagnosed and therefore, untreated until the secondary school level.

Junior temperament character inventory together with quantitative EEG discriminate children with attention deficit hyperactivity disorder combined subtype from children with attention deficit hyperactivity disorder combined subtype plus oppositional defiant disorder.

Oppositional defiant disorder (ODD) is frequently associated with Attention Deficit Hyperactivity Disorder (ADHD) but no clear neurophysiological evidence exists that distinguishes the two groups. Our aim was to identify biomarkers that distinguish children with Attention Deficit Hyperactivity Disorder combined subtype (ADHD_C) from children with ADHD_C + ODD, by combining the results of quantitative EEG (qEEG) and the Junior Temperament Character Inventory (JTCI). 28 ADHD_C and 22 ADHD_C + ODD children who met the DSMV criteria participated in the study. JTCI and EEG were analyzed. Stability based Biomarkers identification methodology was applied to the JTCI and the qEEG separately and combined. The qEEG was tested at the scalp and the sources levels. The classification power of the selected biomarkers was tested with a robust ROC technique. The best discriminant power was obtained when TCI and qEEG were analyzed together. Novelty seeking, self-directedness and cooperativeness were selected as biomarkers together with F4 and Cz in Delta; Fz and F4 in Theta and F7 and F8 in Beta, with a robust AUC of 0.95 for the ROC. At sources level: the regions were the right lateral and medial orbito-frontal cortex, cingular region, angular gyrus, right inferior occipital gyrus, occipital pole and the left insula in Theta, Alpha and Beta. The robust estimate of the total AUC was 0.91. These structures are part of extensive networks of novelty seeking, self-directedness and cooperativeness systems that seem dysregulated in these children. These methods represent an original approach to associate differences of personality and behavior to specific neuronal systems and subsystems.

Principal component analysis for reduction of ocular artefacts in event-related potentials of normal and dyslexic children.

OBJECTIVE: The aim of this study was to reduce ocular artefacts in single trial event-related potentials (ERPs) recorded in normal and in dyslexic children. METHODS: ERPs were recorded during passive and active reading of centrally presented alphabetic letters and non alphabetic symbols. EEG was recorded from 10 EEG locations using the 10-20 system. Diagonal EOG from the right eye was also recorded. Principal component analysis (PCA) was applied in order to reduce ocular artefacts: the first or the second principal component (PC) was subtracted when the correlation coefficient between the component and EOG was greater or equal to 0.9 or 0.95, respectively. Performance of the method was tested on simulated and real data, on both single and averaged trials, varying EOG amplitude and artefact transmission characteristics. RESULTS: Applying the method to real recordings from normal and dyslexic children, we obtained a significant increase in the number of useful trials. In normal children we retrieved 41.0% of the rejected trials in passive and 39.1% in active reading. In dyslexic children 36.7 and 32.2% of the rejected trials in passive and active reading could be included in the respective averages. CONCLUSIONS: The method allows an increase in the number of trials suitable for averaging, a great improvement in ERP quality and a reduction in the recording time.

The psychophysiology of reading.

Early identification of dyslexia would be fundamental to prevent the negative consequences of delayed treatment in the social, psychological and occupational domains. Movement-related potentials of dyslexic children are characterized by inadequate ability to program movements and reduced capacity to evaluate their performance and to correct their errors. Reading-related potentials recorded during different reading conditions elicit a series of positive and negative components with specific functional meaning and with a characteristic spatial-temporal pattern. These reading-related potentials, when analyzed with sLORETA, show significantly different patterns of activation when comparing self-paced reading aloud to passive viewing of single letters. Comparison of fMRI and sLORETA during both tasks showed that the cortical region with the widest inter-modality similarities is the middle-superior temporal lobe during self-paced reading aloud. Neuropsychological studies have shown the existence of clinical subtypes of dyslexia; these studies have been confirmed by the results of ICA applied to the EEG. Dyslexia can be defined as a disorder of programming and integrating ideokinetic elements, associated with a deficiency in the fast processing and integration of sensory information, with reduced efficiency of error systems analysis. Each of these phenomena occurs at different levels of the central nervous system and at different times.

Reading aloud: a psychophysiological investigation in children.

This study investigated the electrophysiological responses to single-letter reading in children (reading-related potentials) and explored the morphological differences between covert and overt reading conditions. Sixty-five healthy children (6-13 years) participated in this study. Reading-related potentials were recorded during visual stimulation with single Italian alphabetic letters. Stimuli were displayed for 5 ms either automatically at a randomly jittered time lag or upon voluntary self-paced button press by children. In the covert conditions, children had to passively look at single letters, while in the overt conditions children were required to read aloud the letters. Electromyographic activity of the forearm and lips was additionally recorded during all tasks. Superimposition of reading-related potentials with the electromyographic activity of forearm and lips during self-paced reading aloud allowed to segregate the reading-related components into four periods: preparatory, pre-lexical, lexical and post-lexical. Reading-related potentials of the preparatory period can be related to preparation/intention to read, those of the pre-lexical period to visual-perceptual processes, those of the lexical period to the external/internal reafferent activity and those of the post-lexical period to the feedback processes following task completion. Analysis of variance showed a significant interaction of reading-related components with electrode locations and task conditions in all periods. The systematic characterization of the neurophysiological correlates of the elementary association between letters and sounds is helpful to highlight the neurobiological and functional basis of reading in healthy as well as impaired readers, for possibly developing neurophysiologically grounded rehabilitation therapies and further improving the explanatory models of dyslexia.

Modulation of specific brain activity by the perceptual analysis of very subtle geometrical relationships of the Mangina-Test stimuli: a functional magnetic resonance imaging (fMRI) investigation in young healthy adults.

The Mangina-Test provides a neuropsychometric assessment of varying degrees of "Analytical-Specific Visual Perception", i.e., the ability to identify simple stimuli inserted into more complex ones according to their exact geometrical properties in a limited span of time. Perceptual analysis of stimuli dealing with the exact discrimination of size and dimension is related more to mathematical abilities (MATH), while perceptual analysis of direction and spatial orientation is related more to abilities in reading and reading comprehension (READ). Some stimuli are MIXED since they deal with both of the above features combined. We previously determined the distributed neuronal network of analytical-specific visual perceptual processes as measured by the Mangina-Test. Here, we aim at further assessing as to how brain activity is differentially modulated by the discrimination of very subtle category-specific perceptual relationships. Brain activity was measured by functional magnetic resonance imaging in 12 young healthy subjects while they performed a computer-adapted version of the Mangina-Test. Behavioral results of the present study indicate that performance was not different among stimulus categories. However, brain functional data show that analytical-specific perceptual processes for MATH, READ and MIXED stimuli rely on partially distinct brain circuits. Bilateral posterior parietal, premotor and prefrontal regions along with the anterior cingulate appeared to be more activated by MATH stimuli, while READ stimuli predominantly activated bilateral medial occipito-temporal, amygdala/parahippocampal and sensorimotor cortices, and the right inferior frontal cortex. Moreover, the posterior parahippocampal cortex showed a higher activity specifically for MIXED stimuli. Altogether, these findings demonstrate that in the Mangina-Test, the exact discrimination of very subtle perceptual relationships between geometrical stimuli distinctly modulates cerebral activity, so that category-specific brain responses can be related to identifiable cognitive abilities.

Neural correlates of "analytical-specific visual perception" and degree of task difficulty as investigated by the Mangina-Test: a functional magnetic resonance imaging (fMRI) study in young healthy adults.

The Mangina-Test is a neuropsychometric method for evaluating varying degrees of "analytical-specific perception" as they relate to learning abilities and disabilities. It consists of the identification of simple stimuli which are masked within a complex configuration according to their exact size, dimension, direction, spatial orientation, and shape within a limited span of time. This test has been successfully applied in clinical settings for the assessment of cognitive abilities and disorders in young and elderly populations. This investigation aimed to examine the neural correlates of analytical-specific visual perceptual processes as measured by the Mangina-Test. Functional Magnetic Resonance Imaging (fMRI) was recorded during the administration of a computer-adapted version of the Mangina-Test in twelve young healthy adults. Multiple linear regression analysis was applied to estimate the overall brain activation during task accomplishment. In addition, the fMRI response area was correlated with task difficulty, in order to explore the spatial distribution of brain regions modulated by increasing task demand. Results indicate that a widely distributed bilateral network of brain regions, including the ventral and dorsal occipital cortex, parietal lobule, frontal and supplementary eye field, dorsolateral prefrontal cortex, and supplementary motor area, was significantly activated during test performance. Moreover, increasing difficulty significantly enhanced the neural response of ventral and dorsal occipital regions, frontal eye field, and superior parietal sulcus bilaterally, as well as the right dorsolateral prefrontal cortex. Conversely, neural activity in the left temporo-parietal junction, inferior frontal gyrus, and bilateral middle-superior temporal cortex was inversely correlated with task difficulty. Results also indicate that performance in the Mangina-Test requires an optimal integration between the enhancement of activity in specific task-related cortical areas and suppression of interfering noise from unrelated brain regions.

Combination of event-related potentials and functional magnetic resonance imaging during single-letter reading.

This work proposes a mathematical approach for combining event-related potentials (ERPs) and functional magnetic resonance images (fMRI). Data were separately recorded during the same event-related experimental design, consisting of visually presented single letters and non-alphabetic symbols, that had to be either simply observed (passive condition) or read aloud (active condition). This protocol was useful for exploring the neural correlates of reading processes. Healthy adults participated in the experiment. Averaged ERPs were decomposed by independent component analysis; low resolution electromagnetic tomography (LORETA) was applied to estimate the current density distribution maps of each independent component. fMRI images time series were analyzed by multiple linear regression. ERP-fMRI correspondence was quantified by computing the Euclidean distance between LORETA local maxima and clusters of significantly activated fMRI voxels. During reading aloud of letters, that is clearly the task most similar to natural reading conditions, significant electrical and hemodynamic response was observed in the left medial frontal gyrus (BA 6) and left middle temporal gyrus (BA 22/39) just before articulation and in the bilateral middle superior temporal gyrus (BA 22/37) during and after verbal-motor production. These results indicate that the middle-superior temporal gyrus plays a crucial and multifunctional role in grapheme-phoneme matching.