Prediction of chronological age from resting-state EEG power in the first three years of life.

The infant brain undergoes rapid and significant developmental changes in the first three years of life. Understanding these changes through the prediction of chronological age using neuroimaging data can provide insights into typical and atypical brain development. We utilized longitudinal resting-state EEG data from 457 typically developing infants, comprising 938 recordings, to develop age prediction models. The multilayer perceptron model demonstrated the highest accuracy with an R2 of 0.82 and a mean absolute error of 92.4 days. Aperiodic offset and periodic theta, alpha, and beta power were identified as key predictors of age via Shapley values. Application of the model to EEG data from infants later diagnosed with autism spectrum disorder or Down syndrome revealed significant underestimations of chronological age. This study establishes the feasibility of using EEG to assess brain maturation in early childhood and supports its potential as a clinical tool for early identification of alterations in brain development.

Alterations in aperiodic and periodic EEG activity in young children with Down syndrome.

Down syndrome is the most common cause of intellectual disability, yet little is known about the neurobiological pathways leading to cognitive impairments. Electroencephalographic (EEG) measures are commonly used to study neurodevelopmental disorders, but few studies have focused on young children with DS. Here we assess resting state EEG data collected from toddlers/preschoolers with DS (n=29, age 13-48 months old) and compare their aperiodic and periodic EEG features with both age-matched (n=29) and cognitive-matched (n=58) comparison groups. DS participants exhibited significantly reduced aperiodic slope, increased periodic theta power, and decreased alpha peak amplitude. A majority of DS participants displayed a prominent peak in the theta range, whereas a theta peak was not present in age-matched participants. Overall, similar findings were also observed when comparing DS and cognitive-matched groups, suggesting that EEG differences are not explained by delayed cognitive ability.

Evaluating early EEG correlates of restricted and repetitive behaviors for toddlers with or without autism.

Background: Restricted and repetitive behaviors (RRB) are among the primary characteristics of autism spectrum disorder (ASD). Despite the potential impact on later developmental outcomes, our understanding of the neural underpinnings of RRBs is limited. Alterations in EEG alpha activity have been observed in ASD and implicated in RRBs, however, developmental changes within the alpha band requires careful methodological considerations when studying its role in brain-behavior relationships during infancy and early childhood. Novel approaches now enable the parameterization of the power spectrum into periodic and aperiodic components. This study aimed to characterize the neural correlates of RRBs in infancy by (1) comparing infant resting-state measures (periodic alpha and aperiodic activity) between infants who develop ASD, elevated likelihood infants without ASD, and low likelihood infants without ASD, and (2) evaluate whether these infant EEG measures are associated with frequency of RRBs measured at 24 months. Methods: Baseline non-task related EEG data were collected from 12-to-14-month-old infants with and without elevated likelihood of autism (N=160), and periodic alpha activity (periodic alpha power, individual peak alpha frequency and amplitude), and aperiodic activity measures (aperiodic exponent) were calculated. Parent-reported RRBs were obtained at 24 months using the Repetitive Behavior Scale-Revised questionnaire. Group differences in EEG measures were evaluated using ANCOVA, and multiple linear regressions were conducted to assess relationships between EEG and RRB measures. Results: No group-level differences in infant EEG measures were observed. Marginal effects analysis of linear regressions revealed significant associations within the ASD group, such that higher periodic alpha power, lower peak alpha frequency, and lower aperiodic exponent, were associated with elevated RRBs at 24 months. No significant associations were observed for non-ASD outcome groups. Limitations: The sample size for ASD (N=19) was modest for examining brain-behavior relations. Larger sample sizes are needed to increase statistical power. Conclusion: For infants with later ASD diagnoses, measures of alpha and aperiodic activity measured at 1-year of age were associated with later manifestation of RRBs at 2-years. Longitudinal studies are needed to elucidate whether the early trajectory of these EEG measures and their dynamic relations in development influence manifestations of RRBs in ASD.

Developmental trajectories of EEG aperiodic and periodic components: Implications for understanding thalamocortical development during infancy.

The development of neural circuits has long-lasting effects on brain function, yet our understanding of early circuit development in humans remains limited. Here, periodic EEG power features and aperiodic components were examined from longitudinal EEGs collected from 592 healthy 2-44 month-old infants, revealing age-dependent nonlinear changes suggestive of distinct milestones in early brain maturation. Consistent with the transient developmental progression of thalamocortical circuitry, we observe the presence and then absence of periodic alpha and high beta peaks across the three-year period, as well as the emergence of a low beta peak (12-20Hz) after six months of age. We present preliminary evidence that the emergence of the low beta peak is associated with higher thalamocortical-dependent, anesthesia-induced alpha coherence. Together, these findings suggest that early age-dependent changes in alpha and beta periodic peaks may reflect the state of thalamocortical network development.

Resting Frontal Gamma Power is Associated with Both Expressive Language and Non-verbal Cognitive Abilities in Young Autistic Children.

Previous research links resting frontal gamma power to key developmental outcomes in young neurotypical (NT) children and infants at risk for language impairment. However, it remains unclear whether gamma power is specifically associated with language or with more general cognitive abilities among young children diagnosed with autism spectrum disorder (ASD). The current study evaluates differences in resting frontal gamma power between young autistic and NT children and tests whether gamma power is uniquely associated with individual differences in expressive language, receptive language and non-verbal cognitive abilities in autistic and NT children. Participants included 48 autistic children and 58 age- and sex-matched NT children (ages 22-60 months). Baseline electroencephalography (EEG) recordings were acquired from each participant. Children also completed the Mullen Scales of Early Learning (MSEL). We found that frontal gamma power at rest did not differ between autistic and NT children. Among autistic children, reduced frontal gamma power was significantly associated with both higher expressive language skills and higher non-verbal cognitive skills, controlling for age and sex. The interaction between frontal gamma power and diagnostic status no longer explained unique variance in expressive language skills after controlling for variance associated with non-verbal cognitive skills across autistic and NT children. Together, these findings suggest that reduced gamma power is associated with both better expressive language and non-verbal cognitive skills among young autistic children. Moreover, associations between high frequency neural activity and cognition are not specific to verbal abilities but reflect neural mechanisms associated with general higher-order cognitive abilities in ASD.

Resting state EEG in young children with Tuberous Sclerosis Complex.

Background: Tuberous Sclerosis Complex (TSC) manifests behaviorally with features of autism, epilepsy, and intellectual disability. Resting state electroencephalography (EEG) offers a window into neural oscillatory activity and may serve as an intermediate biomarker between gene expression and behavioral manifestations. Such a biomarker could be useful in clinical trials as an endpoint or predictor of treatment response. However, seizures and antiepileptic medications also affect resting neural oscillatory activity and could undermine the utility of resting state EEG features as biomarkers in neurodevelopmental disorders such as TSC. Methods: This paper compares resting state EEG features in a cross-sectional cohort of young children with TSC (n=49, ages 12-37 months) to 49 age- and sex-matched typically developing controls. Within children with TSC, associations were examined between resting state EEG features, seizure severity composite score, and use of GABA agonists. Results: Compared to matched typically developing controls, children with TSC showed significantly greater alpha and beta power in permutation cluster analyses iterated across a broad frequency range (2-50Hz). Children with TSC also showed significantly greater aperiodic offset after power spectra were parameterized using SpecParam into aperiodic and periodic components. Within children with TSC, greater seizure severity was significantly related to increased periodic peak beta power. Use of GABA agonists was also independently and significantly associated with increased periodic peak beta power; the interaction between seizure severity and GABA agonist use had no significant effect on peak beta power. Conclusions: The elevated peak beta power observed in children with TSC compared to matched typically developing controls may be driven by both seizures and GABA agonist use. It is recommended to collect seizure and mediation data alongside EEG data for clinical trials. These results highlight the challenge of using resting state EEG features as biomarkers in trials with neurodevelopmental disabilities when epilepsy and anti-epileptic medication are common.

Developmental trajectories of EEG aperiodic and periodic components in children 2-44 months of age.

The development of neural circuits has long-lasting effects on brain function, yet our understanding of early circuit development in humans remains limited. Here, periodic EEG power features and aperiodic components were examined from longitudinal EEGs collected from 592 healthy 2-44 month-old infants, revealing age-dependent nonlinear changes suggestive of distinct milestones in early brain maturation. Developmental changes in periodic peaks include (1) the presence and then absence of a 9-10 Hz alpha peak between 2-6 months, (2) nonlinear changes in high beta peaks (20-30 Hz) between 4-18 months, and (3) the emergence of a low beta peak (12-20 Hz) in some infants after six months of age. We hypothesized that the emergence of the low beta peak may reflect maturation of thalamocortical network development. Infant anesthesia studies observe that GABA-modulating anesthetics do not induce thalamocortical mediated frontal alpha coherence until 10-12 months of age. Using a small cohort of infants (n = 23) with EEG before and during GABA-modulating anesthesia, we provide preliminary evidence that infants with a low beta peak have higher anesthesia-induced alpha coherence compared to those without a low beta peak.

Familial Recurrence of Autism: Updates From the Baby Siblings Research Consortium.

OBJECTIVES: Autism spectrum disorder (ASD) is estimated to be ∼10 times higher in children with versus without an autistic sibling in population-based studies. Prospective studies of infant siblings have revealed even higher familial recurrence rates. In the current prospective longitudinal study, we provide updated estimates of familial ASD recurrence using a multinational database of infants with older autistic siblings. METHODS: Data were collated across 18 sites of the Baby Siblings Research Consortium, an international network studying the earliest manifestations of ASD. A total of 1605 infants with an older autistic sibling were followed from early in life to 3 years, when they were classified as ASD or non-ASD. Hierarchical generalized linear modeling, with site as a random effect, was used to examine predictors of recurrence in families and calculate likelihood ratios. RESULTS: A total of 20.2% of siblings developed ASD, which is not significantly higher than the previously reported rate of 18.7%. Male infant sex and >1 older affected sibling were significant predictors of familial recurrence. Proband sex also influenced recurrence rates, with siblings of female probands significantly more likely to develop ASD than siblings of male probands. Race and maternal education were also associated with recurrence in families. CONCLUSIONS: The familial recurrence rate of ASD, as measured in infant sibling studies, has not changed appreciably since previous estimates were made in 2011. Younger siblings of autistic children, particularly those who are male, have an affected female sibling, multiple affected siblings, or are impacted by social inequities, should be closely monitored and promptly referred for diagnostic evaluation.

Associations between EEG trajectories, family income, and cognitive abilities over the first two years of life.

We sought to characterize developmental trajectories of EEG spectral power over the first 2 years after birth and examine whether family income or maternal education alter those trajectories. We analyzed EEGs (n = 161 infants, 534 EEGs) collected longitudinally between 2 and 24 months of age, and calculated frontal absolute power across 7 canonical frequency bands. For each frequency band, a piecewise growth curve model was fit, resulting in an estimated intercept and two slope parameters from 2 to 9 months and 9-24 months of age. Across 6/7 frequency bands, absolute power significantly increased over age, with steeper slopes in the 2-9 month period compared to 9-24 months. Increased family income, but not maternal education, was associated with higher intercept (2-3 month power) across delta-gamma bands (p range = 0.002-0.04), and reduced change in power between 2 and 9 months of age in lower frequency bands (delta-alpha, p range = 0.01-0.02). There was no significant effect of income on slope between 9 and 24 months. EEG intercept and slope measures did not mediate relationships between income and 24-month verbal and nonverbal development. These results add to growing literature concerning the role of socioeconomic factors in shaping brain trajectories.

Amblyopia treatment outcomes in patients with neurodevelopmental disorders.

PURPOSE: To compare amblyopia treatment outcomes between patients with neurodevelopmental disorders and their typically developing peers. METHODS: Of 2,311 patients diagnosed with amblyopia between 2010 and 2014 at Boston Children's Hospital, 460 met inclusion criteria (age 2-12 with anisometropic, strabismic, or mixed amblyopia [interocular difference (IOD) ≥2 lines]). Treatment and visual outcomes were analyzed according to neurodevelopmental status: neurodevelopmental delay (DD) versus typical development (TD). RESULTS: The DD group (n = 54) and TD group (n = 406) were similar in demographics, amblyogenic risk factors, baseline visual measures, prescribed therapy, and adherence (P ≥ 0.10). Between-visit follow-up time was longer for the DD group (0.65 [0.42- 0.97] years) than for the TD group (0.5 [0.36-0.82] years; P = 0.023). IOD improved similarly in each group by the last visit (DD, -0.15 logMAR [-0.31 to -0.02]; TD, -0.2 logMAR [-0.38 to -0.1]; P = 0.09). Each group reached amblyopia resolution by the last visit at similar frequencies (DD, 23/54 [43%]; TD, 211/406 [52%]; P > 0.2). DD diagnosis did not independently influence amblyopia resolution (HR, 0.77; 95% CI, 0.53-1.12; P = 0.17), but each additional month of interval time between follow-up visits reduced the likelihood of resolution by 2.7% (HR, 0.67; 95% CI, 0.51-0.87; P = 0.003). CONCLUSIONS: Patients with DD and those with TD responded similarly to amblyopia therapy; however, follow-up intervals were longer in patients with DD and correlated with the likelihood of persistent amblyopia, suggesting that greater efforts at assuring follow-up may benefit patients with DD.

Developmental trajectories of EEG aperiodic and periodic power: Implications for understanding the timing of thalamocortical development during infancy.

The development of neural circuits over the first years of life has long-lasting effects on brain function, yet our understanding of early circuit development in humans remains limited. Here, aperiodic and periodic EEG power features were examined from longitudinal EEGs collected from 592 healthy 2-44 month-old infants, revealing age-dependent nonlinear changes suggestive of distinct milestones in early brain maturation. Consistent with the transient developmental progression of thalamocortical circuitry, we observe the presence and then absence of periodic alpha and high beta peaks across the three-year period, as well as the emergence of a low beta peak (12-20Hz) after six months of age. We present preliminary evidence that the emergence of the low beta peak is associated with thalamocortical connectivity sufficient for anesthesia-induced alpha coherence. Together, these findings suggest that early age-dependent changes in alpha and beta periodic peaks may reflect the state of thalamocortical network development.

Associations Between Infant Screen Use, Electroencephalography Markers, and Cognitive Outcomes.

Importance: Research evidence is mounting for the association between infant screen use and negative cognitive outcomes related to attention and executive functions. The nature, timing, and persistence of screen time exposure on neural functions are currently unknown. Electroencephalography (EEG) permits elucidation of the neural correlates associated with cognitive impairments. Objective: To examine the associations between infant screen time, EEG markers, and school-age cognitive outcomes using mediation analysis with structural equation modeling. Design, Setting, and Participants: This prospective maternal-child dyad cohort study included participants from the population-based study Growing Up in Singapore Toward Healthy Outcomes (GUSTO). Pregnant mothers were enrolled in their first trimester from June 2009 through December 2010. A subset of children who completed neurodevelopmental visits at ages 12 months and 9 years had EEG performed at age 18 months. Data were reported from 3 time points at ages 12 months, 18 months, and 9 years. Mediation analyses were used to investigate how neural correlates were involved in the paths from infant screen time to the latent construct of attention and executive functioning. Data for this study were collected from November 2010 to March 2020 and were analyzed between October 2021 and May 2022. Exposures: Parent-reported screen time at age 12 months. Main Outcomes and Measures: Power spectral density from EEG was collected at age 18 months. Child attention and executive functions were measured with teacher-reported questionnaires and objective laboratory-based tasks at age 9 years. Results: In this sample of 437 children, the mean (SD) age at follow-up was 8.84 (0.07) years, and 227 children (51.9%) were male. The mean (SD) amount of daily screen time at age 12 months was 2.01 (1.86) hours. Screen time at age 12 months contributed to multiple 9-year attention and executive functioning measures (η2, 0.03-0.16; Cohen d, 0.35-0.87). A subset of 157 children had EEG performed at age 18 months; EEG relative theta power and theta/beta ratio at the frontocentral and parietal regions showed a graded correlation with 12-month screen use (r = 0.35-0.37). In the structural equation model accounting for household income, frontocentral and parietal theta/beta ratios partially mediated the association between infant screen time and executive functioning at school age (exposure-mediator ß, 0.41; 95% CI, 0.22 to 0.59; mediator-outcome ß, -0.38; 95% CI, -0.64 to -0.11), forming an indirect path that accounted for 39.4% of the association. Conclusions and Relevance: In this study, infant screen use was associated with altered cortical EEG activity before age 2 years; the identified EEG markers mediated the association between infant screen time and executive functions. Further efforts are urgently needed to distinguish the direct association of infant screen use compared with family factors that predispose early screen use on executive function impairments.

Neural response to repeated auditory stimuli and its association with early language ability in male children with Fragile X syndrome.

Background: Fragile X syndrome (FXS) is the most prevalent form of inherited intellectual disability and is commonly associated with autism. Previous studies have linked the structural and functional alterations in FXS with impaired sensory processing and sensory hypersensitivity, which may hinder the early development of cognitive functions such as language comprehension. In this study, we compared the P1 response of the auditory evoked potential and its habituation to repeated auditory stimuli in male children (2-7 years old) with and without FXS, and examined their association with clinical measures in these two groups. Methods: We collected high-density electroencephalography (EEG) data in an auditory oddball paradigm from 12 male children with FXS and 11 age- and sex-matched typically developing (TD) children. After standardized EEG pre-processing, we conducted a spatial principal component (PC) analysis and identified two major PCs-a frontal PC and a temporal PC. Within each PC, we compared the P1 amplitude and inter-trial phase coherence (ITPC) between the two groups, and performed a series of linear regression analysis to study the association between these EEG measures and several clinical measures, including assessment scores for language abilities, non-verbal skills, and sensory hypersensitivity. Results: At the temporal PC, both early and late standard stimuli evoked a larger P1 response in FXS compared to TD participants. For temporal ITPC, the TD group showed greater habituation than the FXS group. However, neither group showed significant habituation of the frontal or temporal P1 response. Despite lack of habituation, exploratory analysis of brain-behavior associations observed that within the FXS group, reduced frontal P1 response to late standard stimuli, and increased frontal P1 habituation were both associated with better language scores. Conclusion: We identified P1 amplitude and ITPC in the temporal region as a contrasting EEG phenotype between the FXS and the TD groups. However, only frontal P1 response and habituation were associated with language measures. Larger longitudinal studies are required to determine whether these EEG measures could be used as biomarkers for language development in patients with FXS.

Neural correlates of face processing associated with development of social communication in 12-month infants with familial risk of autism spectrum disorder.

BACKGROUND: Differences in face processing in individuals with ASD is hypothesized to impact the development of social communication skills. This study aimed to characterize the neural correlates of face processing in 12-month-old infants at familial risk of developing ASD by (1) comparing face-sensitive event-related potentials (ERP) (Nc, N290, P400) between high-familial-risk infants who develop ASD (HR-ASD), high-familial-risk infants without ASD (HR-NoASD), and low-familial-risk infants (LR), and (2) evaluating how face-sensitive ERP components are associated with development of social communication skills. METHODS: 12-month-old infants participated in a study in which they were presented with alternating images of their mother's face and the face of a stranger (LR = 45, HR-NoASD = 41, HR-ASD = 24) as EEG data were collected. Parent-reported and laboratory-observed social communication measures were obtained at 12 and 18 months. Group differences in ERP responses were evaluated using ANOVA, and multiple linear regressions were conducted with maternal education and outcome groups as covariates to assess relationships between ERP and behavioral measures. RESULTS: For each of the ERP components (Nc [negative-central], N290, and P400), the amplitude difference between mother and stranger (Mother-Stranger) trials was not statistically different between the three outcome groups (Nc p = 0.72, N290 p = 0.88, P400 p = 0.91). Marginal effects analyses found that within the LR group, a greater Nc Mother-Stranger response was associated with better expressive language skills on the Mullen Scales of Early Learning, controlling for maternal education and outcome group effects (marginal effects dy/dx = 1.15; p < 0.01). No significant associations were observed between the Nc and language or social measures in HR-NoASD or HR-ASD groups. In contrast, specific to the HR-ASD group, amplitude difference between the Mother versus Stranger P400 response was positively associated with expressive (dy/dx = 2.1, p < 0.001) and receptive language skills at 12 months (dy/dx = 1.68, p < 0.005), and negatively associated with social affect scores on the Autism Diagnostic Observation Schedule (dy/dx = - 1.22, p < 0.001) at 18 months. CONCLUSIONS: In 12-month-old infant siblings with subsequent ASD, increased P400 response to Mother over Stranger faces is positively associated with concurrent language and future social skills.

Parental Language Input Predicts Neuroscillatory Patterns Associated with Language Development in Toddlers at Risk of Autism.

In this study we investigated the impact of parental language input on language development and associated neuroscillatory patterns in toddlers at risk of Autism Spectrum Disorder (ASD). Forty-six mother-toddler dyads at either high (n = 22) or low (n = 24) familial risk of ASD completed a longitudinal, prospective study including free-play, resting electroencephalography, and standardized language assessments. Input quantity/quality at 18 months positively predicted expressive language at 24 months, and relationships were stronger for high-risk toddlers. Moderated mediations revealed that input-language relationships were explained by 24-month frontal and temporal gamma power (30-50 Hz) for high-risk toddlers who would later develop ASD. Results suggest that high-risk toddlers may be cognitively and neurally more sensitive to their language environments, which has implications for early intervention.

Increased aperiodic gamma power in young boys with Fragile X Syndrome is associated with better language ability.

BACKGROUND: The lack of robust and reliable clinical biomarkers in Fragile X Syndrome (FXS), the most common inherited form of intellectual disability, has limited the successful translation of bench-to-bedside therapeutics. While numerous drugs have shown promise in reversing synaptic and behavioral phenotypes in mouse models of FXS, none have demonstrated clinical efficacy in humans. Electroencephalographic (EEG) measures have been identified as candidate biomarkers as EEG recordings of both adults with FXS and mouse models of FXS consistently exhibit alterations in resting state and task-related activity. However, the developmental timing of these EEG differences is not known as thus far EEG studies have not focused on young children with FXS. Further, understanding how EEG differences are associated with core symptoms of FXS is crucial to successful use of EEG as a biomarker, and may improve our understanding of the disorder. METHODS: Resting-state EEG was collected from FXS boys with full mutation of Fmr1 (2.5-7 years old, n = 11) and compared with both age-matched (n = 12) and cognitive-matched (n = 12) typically developing boys. Power spectra (including aperiodic and periodic components) were compared using non-parametric cluster-based permutation testing. Associations between 30 and 50 Hz gamma power and cognitive, language, and behavioral measures were evaluated using Pearson correlation and linear regression with age as a covariate. RESULTS: FXS participants showed increased power in the beta/gamma range (~ 25-50 Hz) across multiple brain regions. Both a reduction in the aperiodic (1/f) slope and increase in beta/gamma periodic activity contributed to the significant increase in high-frequency power. Increased gamma power, driven by the aperiodic component, was associated with better language ability in the FXS group. No association was observed between gamma power and parent report measures of behavioral challenges, sensory hypersensitivities, or adaptive behaviors. LIMITATIONS: The study sample size was small, although comparable to other human studies in rare-genetic disorders. Findings are also limited to males in the age range studied. CONCLUSIONS: Resting-state EEG measures from this study in young boys with FXS identified similar increases in gamma power previously reported in adults and mouse models. The observed positive association between resting state aperiodic gamma power and language development supports hypotheses that alterations in some EEG measures may reflect ongoing compensatory mechanisms.

Sensory processing and P300 event-related potential correlates of stimulant response in children with attention-deficit/hyperactivity disorder: A critical review.

Attention-deficit/hyperactivity disorder (ADHD) is a neurodevelopmental disorder associated with considerable impairment in psychiatric and functional domains. Although stimulant medication can reduce core symptoms of inattention, hyperactivity, and impulsivity, a subgroup of patients does not respond to this intervention. A precision medicine approach has been proposed, whereby biomarkers are used to identify an effective treatment approach for a given individual. This review synthesizes the existing literature on event-related potential (ERP) correlates of stimulant response in children diagnosed with ADHD, with the goal of evaluating the potential for ERP to inform precision medicine care in this population. Forty-three articles were examined and results tentatively suggest that stimulant medications normalize the amplitude of the P300 component, and this is also associated with behavioral improvement. In contrast, results generally indicate that stimulants do not significantly alter early processing components, although there are some exceptions to this finding. Implications for research, theory, and clinical work are considered and concrete recommendations for future directions are provided. While recognizing limitations of existing literature (e.g., homogenous samples, variable methodologies), we conclude that ERP methods represent a promising approach for precision medicine care of patients with ADHD.

Prediction of autism spectrum disorder diagnosis using nonlinear measures of language-related EEG at 6 and 12 months.

BACKGROUND: Early identification of autism spectrum disorder (ASD) provides an opportunity for early intervention and improved developmental outcomes. The use of electroencephalography (EEG) in infancy has shown promise in predicting later ASD diagnoses and in identifying neural mechanisms underlying the disorder. Given the high co-morbidity with language impairment, we and others have speculated that infants who are later diagnosed with ASD have altered language learning, including phoneme discrimination. Phoneme learning occurs rapidly in infancy, so altered neural substrates during the first year of life may serve as early, accurate indicators of later autism diagnosis. METHODS: Using EEG data collected at two different ages during a passive phoneme task in infants with high familial risk for ASD, we compared the predictive accuracy of a combination of feature selection and machine learning models at 6 months (during native phoneme learning) and 12 months (after native phoneme learning), and we identified a single model with strong predictive accuracy (100%) for both ages. Samples at both ages were matched in size and diagnoses (n = 14 with later ASD; n = 40 without ASD). Features included a combination of power and nonlinear measures across the 10­20 montage electrodes and 6 frequency bands. Predictive features at each age were compared both by feature characteristics and EEG scalp location. Additional prediction analyses were performed on all EEGs collected at 12 months; this larger sample included 67 HR infants (27 HR-ASD, 40 HR-noASD). RESULTS: Using a combination of Pearson correlation feature selection and support vector machine classifier, 100% predictive diagnostic accuracy was observed at both 6 and 12 months. Predictive features differed between the models trained on 6- versus 12-month data. At 6 months, predictive features were biased to measures from central electrodes, power measures, and frequencies in the alpha range. At 12 months, predictive features were more distributed between power and nonlinear measures, and biased toward frequencies in the beta range. However, diagnosis prediction accuracy substantially decreased in the larger, more behaviorally heterogeneous 12-month sample. CONCLUSIONS: These results demonstrate that speech processing EEG measures can facilitate earlier identification of ASD but emphasize the need for age-specific predictive models with large sample sizes to develop clinically relevant classification algorithms.

Use of longitudinal EEG measures in estimating language development in infants with and without familial risk for autism spectrum disorder.

Language development in children with autism spectrum disorder (ASD) varies greatly among affected individuals and is a strong predictor of later outcomes. Younger siblings of children with ASD have increased risk of ASD, but also language delay. Identifying neural markers of language outcomes in infant siblings could facilitate earlier intervention and improved outcomes. This study aimed to determine whether EEG measures from the first 2-years of life can explain heterogeneity in language development in children at low- and high-risk for ASD, and to determine whether associations between EEG measures and language development are different depending on ASD risk status or later ASD diagnosis. In this prospective longitudinal study EEG measures collected between 3-24 months were used in a multivariate linear regression model to estimate participants' 24-month language development. Individual baseline longitudinal EEG measures included (1) the slope of EEG power across 3-12 months or 3-24 months of life for 6 canonical frequency bands, (2) estimated EEG power at age 6-months for the same frequency bands, and (3) terms representing the interaction between ASD risk status and EEG power measures. Modeled 24-month language scores using EEG data from either the first 2-years (Pearson R = 0.70, 95% CI 0.595-0.783, P=1x10-18) or the first year of life (Pearson R=0.66, 95% CI 0.540-0.761, P=2.5x10-14) were highly correlated with observed scores. All models included significant interaction effects of risk on EEG measures, suggesting that EEG-language associations are different depending on risk status, and that different brain mechanisms effect language development in low-versus high-risk infants.