Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

EEG signatures of cognitive and social development of preschool children–a systematic review

  • Supriya Bhavnani ,

    Roles Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Visualization, Writing – original draft, Writing – review & editing

    ‡ These authors share first authorship on this work.

    Affiliation Sangath, New Delhi, India

  • Georgia Lockwood Estrin ,

    Roles Conceptualization, Formal analysis, Investigation, Methodology, Project administration, Visualization, Writing – original draft, Writing – review & editing

    ‡ These authors share first authorship on this work.

    Affiliation Centre for Brain and Cognitive Development, Birkbeck, University of London, London, United Kingdom

  • Rianne Haartsen,

    Roles Formal analysis, Writing – review & editing

    Affiliation Centre for Brain and Cognitive Development, Birkbeck, University of London, London, United Kingdom

  • Sarah K. G. Jensen,

    Roles Conceptualization, Formal analysis, Methodology, Writing – review & editing

    Affiliation Boston Children’s Hospital, Harvard Medical School, Boston, MA, United States of America

  • Teodora Gliga,

    Roles Supervision, Writing – review & editing

    Affiliation School of Psychology, University of East Anglia, Norwich, United Kingdom

  • Vikram Patel,

    Roles Conceptualization, Supervision, Writing – review & editing

    Affiliations Sangath, New Delhi, India, Harvard T.H. Chan School of Public Health, Boston, MA, United States of America

  • Mark H. Johnson

    Roles Conceptualization, Resources, Supervision, Writing – review & editing

    Affiliations Centre for Brain and Cognitive Development, Birkbeck, University of London, London, United Kingdom, Department of Psychology, University of Cambridge, Cambridge, United Kingdom



Early identification of preschool children who are at risk of faltering in their development is essential to ensuring that all children attain their full potential. Electroencephalography (EEG) has been used to measure neural correlates of cognitive and social development in children for decades. Effective portable and low-cost EEG devices increase the potential of its use to assess neurodevelopment in children at scale and particularly in low-resource settings. We conducted a systematic review aimed to synthesise EEG measures of cognitive and social development in 2-5-year old children. Our secondary aim was to identify how these measures differ across a) the course of development within this age range, b) gender and c) socioeconomic status (SES).

Methods and findings

A systematic literature search identified 51 studies for inclusion in this review. Data relevant to the primary and secondary aims was extracted from these studies and an assessment for risk of bias was done, which highlighted the need for harmonisation of EEG data collection and analysis methods across research groups and more detailed reporting of participant characteristics. Studies reported on the domains of executive function (n = 22 papers), selective auditory attention (n = 9), learning and memory (n = 5), processing of faces (n = 7) and emotional stimuli (n = 8). For papers investigating executive function and selective auditory attention, the most commonly reported measures were alpha power and the amplitude and latency of positive (P1, P2, P3) and negative (N1, N2) deflections of event related potential (ERPs) components. The N170 and P1 ERP components were the most commonly reported neural responses to face and emotional faces stimuli. A mid-latency negative component and positive slow wave were used to index learning and memory, and late positive potential in response to emotional non-face stimuli. While almost half the studies described changes in EEG measures across age, only eight studies disaggregated results based on gender, and six included children from low income households to assess the impact of SES on neurodevelopment. No studies were conducted in low- and middle-income countries.


This review has identified power across the EEG spectrum and ERP components to be the measures most commonly reported in studies in which preschool children engage in tasks indexing cognitive and social development. It has also highlighted the need for additional research into their changes across age and based on gender and SES.


The importance of making a concerted global effort towards optimising early child development is rapidly being recognised, particularly as child survival increases due to the successful reduction in infant and child mortality rates across the world. Using stunting and poverty as indicators, Lu and colleagues demonstrated that over 200 million children in low and middle income countries (LMICs) are at risk of suboptimal development [1]. Another study using care-giver report data from 35 LMICs suggests that one in every three preschool-age children are failing to meet expected cognitive or social developmental milestones [2]. Cognitive abilities include learning and memory, selective visual and auditory discrimination and executive function; social abilities can be indexed by how children process facial and emotional stimuli. These domains of development lay the foundation for learning and therefore readiness for school, with delayed or suboptimal development of these abilities negatively impacting academic performance [3,4].

Key to ensuring that all children thrive is the early identification of those not following a typical developmental trajectory, and their subsequent timely referral to interventions. The most widely used approach to assess neurodevelopment is behavioural observations by specialists. Given the scarcity of clinical professionals in LMICs, it is essential to create and validate efficient methods that are objective, amenable for administration by trained non-specialist workers and therefore scalable in multiple low resource settings [5,6]. Neurophysiological methods like electroencephalography (EEG) offer complementary methods to assess brain development in children as it is a non-invasive, direct measure of brain activity with high temporal resolution. In addition to laboratory grade equipment used in most EEG studies, low-cost, portable EEG devices have recently become available on the market. Some companies also offer cloud-based analysis of the data, removing the need for expertise on site. These advances present an opportunity to examine the potential use of EEG at scale in the future [7].

Given that brain plasticity (ability to adapt to environmental circumstances) is at its peak in early childhood, interventions to optimise child development implemented during preschool years are known to be most effective in improving developmental outcomes [810]. Intervening in the early years also provides the highest return on investments, further strengthening the argument in favour of early identification of children at risk for not attaining their full developmental potential [11].

A significant amount of research has been done to establish trajectories of cognitive and social development. However, due to disparities in research funding, these studies, which require considerable sample sizes and longitudinal follow up, have largely been restricted to children from high-income countries (HICs). This is despite the fact that a disproportionately greater number of children at risk of not attaining their full developmental potential reside in LMICs. Given emerging evidence that signatures of brain development differ across cultures, there is an urgent need to capture a broader range of developmental trajectories globally including underserved populations [12]. This is an essential first step to identification of children who are developing sub-optimally and improvement of their individual prospects, resulting, in the long term, in lifting people out of poverty to break the vicious cycle of intergenerational transmission of disadvantage [13].

To this end, it is valuable to synthesise the existing knowledge that EEG studies, which have been used to assess neural correlates of cognitive and social developmental processes such as visual attention and memory for decades albeit in HICs, has generated [1417]. A range of measures have been developed to examine: a) the timing (latency) and amplitude of event-related potentials (ERPs), time-locked brain activity in response to a stimulus [18]; or b) continuous brain activity, either during a task or at rest (called resting state), examining the synchronisation of oscillations via spectral power and connectivity [19]. Accumulating evidence highlights the potential of EEG recorded during resting state to identify children faltering in their development [16,20] or those with neurodevelopmental disorders such as autism spectrum disorders (ASD), attention deficit hyperactivity disorder (ADHD) or learning-disability [2127]. Recent systematic reviews have also focused on establishing the prognostic accuracy of resting-state EEG recorded in preterm infants in predicting neurodevelopmental outcomes [28,29]. Some efforts have also been made to synthesise the vast body of EEG literature to isolate resting-state EEG measures that can serve as signatures of cognitive and social development in preschool children [30].

However, there are limited reviews consolidating measures that are derived from EEG recordings done while preschool children, aged 2–5 years, are engaged in tasks designed to measure cognitive and social development [31], perhaps due to the challenges in assessing children of this age. Therefore, in an effort to identify neural correlates that may reflect the developmental status of key cognitive and social abilities in preschool children, we conducted a systematic review of the EEG literature to synthesise existing knowledge across studies. Our primary aim was to identify task-related EEG measures that indexed cognitive and social development in children aged 2–5 years.

Furthermore, EEG measures have been demonstrated to change over the course of development [14,32], with evidence of non-linear brain development emerging from early seminal EEG studies in the 1980s-90s conducted by Thatcher and colleagues [3335]. Interestingly, the evidence of differences based on gender is mixed [36,37]. Some studies also indicate differences in EEG measures based on socioeconomic status of children [38,39]. Gaining clarity on EEG differences based on gender and SES are particularly relevant in the context of the children from LMICs who are more likely to grow up in poverty, with girls often receiving a disproportionately low share of scant resources [40]. Our secondary aim was thus to identify how such EEG measures differ across a) the course of development within this age range, b) gender and c) socioeconomic status.



This systematic review was conducted following PRISMA guidelines ( and the results are summarised in a PRISMA flowchart in Fig 1. A comprehensive literature search was conducted on five databases—Embase, Medline and Psycinfo (searched on 12/01/2018, updated on 15/05/2020) yielded 27858 records, Scopus (searched on 16/02/2018, updated on 15/05/2020) yielded 14900 records and Cinahl (searched on 23/02/2018, updated on 15/05/2020) yielded 284 records. The search terms included EEG, cognition, attention, social development and child development and are detailed in S1 Table. These 43042 records contained 15581 duplicates which were removed. Reviewers (SB and GLE) split the remaining 27461 records equally. All 27461 records were screened first at the title level and any undecided papers were discussed between reviewers. Abstract screening was completed on 4121 records, and inter-rater reliability was established between reviewers SB and GLE at this stage. Shared reference manager tool (Endnote and Rayyan) libraries were used to consult each other about inclusion or exclusion of papers throughout the process of review. Each reviewer classified approximately 10% of the other’s papers while being blinded to the other’s decisions. Cohen’s kappa was determined to be 0.63 (95% CI: 0.52, 0.74) representing ‘good’ reliability. All conflicting decisions were discussed and resolved, and any unresolved papers were discussed with co-author RH until consensus was reached. Based on the screening of abstracts, 547 records made it to the full text review and, based on the criteria below, 48 were categorised into ‘included’ while 499 were ‘excluded’. In order to ensure comprehensiveness of the database search, bibliographies of recent included studies were reviewed for relevance to our research question. Three relevant records were added through this process resulting in final inclusion of 51 papers.

Fig 1. PRISMA flowchart—summary of the results of the search conducted in this review.

Inclusion/Exclusion criteria

Articles were excluded if they 1) were not published within 3 decades prior to the search date, 2) were not published in English language or peer-reviewed journals, 3) were case or series of case studies (number of participants ≤10), 4) reported results of interventions, 5) did not contain participants within the target age range of 2–5 years i.e. 24–72 months, 6) included participants beyond the target age range and did not disaggregate results by age, 7) included participants with atypical development including any diagnosed mental and neurodevelopmental disorders or cognitive delays, physical disabilities or genetic disorders, 8) included participants at risk of atypical development due to known risk factors such as preterm birth, pre- and perinatal infections and maternal conditions like diabetes or depression, 9) were conducted while participants were in resting state, asleep/unconscious, anxious, in fear, pain, or experiencing a headache and 10) assessed lower level sensory processing such as vision and hearing, other domains of development such as language, and motor, or academic skills like mathematic/arithmetic, reading/comprehension. Studies reporting assessment of cognitive and social domains of executive function, selective auditory attention, learning and memory, processing of faces and emotional stimuli as defined by the papers, were included.

Data extraction

A data extraction table was created for the following features of included papers: a) publication details such as income level of the country in which the study was conducted and sample size; b) participant characteristics such as age, demographic information including parental education and income, recruitment strategy, inclusion/exclusion criteria and reasons for loss of participants; c) EEG device characteristics such as brand, number of electrodes and sampling rate; d) study procedure details like the type of cognitive domain assessed, task used and setup information; e) data pre-processing steps of filtering, artefact identification and rejection, segmentation of EEG signal, regions and time windows of interest; f) data analysis methods, including statistical methods, significant results and conclusions relevant to the primary and secondary aims of this review, and finally g) limitations of the study acknowledged by the authors. Reviewers SB and GLE split the included records equally and extracted data independently. The data that has been extracted was then synthesised by them together in close consultation with co-author RH.

Assessment of risk of bias

The KMet quality appraisal checklist [41], which was created in response to the need for standardised quality assessment criteria applicable for evaluating primary research studies from a variety of fields, was piloted and 11 of the 14 questions were adapted in consultation with co-author SJ for use in this study, along with two additional questions (see S2 Table for list of 13 quality appraisal questions). Reviewers SB and GLE appraised all studies together by achieving consensus through discussions. Studies could score either 2 for ‘yes’ (high quality), 1 for ‘partial’ or 0 for ‘no’ (low quality) for each quality appraisal question. For the question on appropriate sample size (question 8), studies were scored ‘yes’ if they included 20 or more participants in each analysis group (e.g. age groups), ‘partial’ if total participant number was 20 or more but analysis group number was less than 20 and ‘no’ if data from less than 20 participants was analysed. Percentage of yes, partial and no were calculated to provide a graphical summary of the appraisal of all included studies.


Characteristics of included studies

Study participants.

The 51 studies included in this review represent a total of 2123 participants within the target age range of 2–5 years. Most of the participants of included studies were aged between 4–5 years (S1 Fig and Table 1) with fewer studies (9/44) conducted with younger (2-3-year old) children. The participants of these studies covered an age range of over 1.5 years. Almost half (24/51) of the included studies report developmental changes in EEG measures by comparing cross-sectional data from preschool children with other ages ranging from infants to adults. All studies specified the distribution of gender amongst their participants, however only eight report on the impact of gender on their results.

Table 1. Details of included papers: Sample size, mean age, task name, EEG metric and results in relation to the cognitive ability, age, gender and socioeconomic status.

A total of 48 of 51 studies were conducted in high-income countries with the majority (34/48) being from USA. Many studies (19/48) did not specify the ethnicity of their participants and in those that did, a lack of diversity was evident, with the average proportion of Caucasians being 79.2% (SD: 16.4%) (S3 Table). Three studies were published from China, representing an upper-middle-income country. None of the studies identified in this review were conducted in LMICs and only six studies sampled children from low income families to analyse the impact of socioeconomic status (SES) on brain activity. All of the remaining studies that mention the SES of their participants (15/48 did not specify it) report that they are from middle to high income families with high levels of education (S3 Table).

EEG data collection and pre-processing procedures.

There was a large diversity in the equipment used to collect EEG data in these studies (S4 Table). The number of electrodes ranged from 5 to 128 (low to high density), with most studies using the 10/20 array. Sampling rate ranged from 100 to 2000 Hz. Data pre-processing techniques used in these studies also differed greatly making it hard to compare findings across studies (S5 Table). The data was either band-pass filtered from 0.1 or 1 to 30 or 40 Hz, or a stepped approach was taken applying a band-pass filter of 0.1 to 100 Hz first and then followed by a low-pass filter at 30 Hz (typically for ERPs). Only 3 studies reported filtering for line noise at 50 or 60 Hz. Most often the vertex (Cz electrode) was used as a reference during recording (27/51), while use of mastoids as references was less frequent (18/51). The data was then re-referenced offline to the average of all electrodes in 28 studies, of which 15 had a relatively low electrode density (<60 electrodes). To further clean the data, thresholds were used for automated identification of artefacts, for example, a peak-to-peak criterion of greater than 100 μV and 200 μV for eye blinks or movements and gross motor movements, respectively. This was typically followed by visual inspection of the data and subsequent removal of artefacts in almost all studies. EEG data was segmented and, depending on the task used in the study, corrected for baseline, which ranged from 50 to 600ms (with the majority of the studies using 100 to 200ms) before stimulus onset. In ERP studies, the time-window analysed depended on the ERP components of interest. However, the range of the time-window differed between studies examining the same component, as some studies defined time-windows a-priori while others used a data-driven approach based on grand average ERP of their sample. Similar differences between studies were found for the definition of regions or electrodes of interest, which could either be predetermined or automatically identified with for example Principal Component Analysis (PCA).

Assessment of risk of bias.

The results of the quality appraisal of all included studies are summarised in Fig 2 (see S2 Table for questions). Methodological quality of most studies was high (93–100% studies scored ‘yes’) when appraised for their research question, study design, definition of outcome (EEG) measures and reporting of analytic methods, results and conclusions. However, only 61% of studies included an adequate number of participants (N ≥ 20) in each analysed group. About half of the studies received ‘partial’ or ‘no’ scores on criteria assessing the method of recruitment and description of participants such as sociodemographic details. Interestingly, a large number of the studies either did not at all (24%), or did only partially (41%), report on the method employed to assess whether participating children were developing typically, for example by assessing children’s development using validated scales such as Mullen’s Scale of Early Learning [92] and Wechsler Intelligence Scale for Children [93]. Most studies mentioned the number of participants excluded from their analysis, along with sufficient details on the reasons for exclusion, however, they did not disaggregate this loss of participants across their analysis groups like age. 41% of the included studies did not report limitations of their study. Those that did highlighted technical limitations in the EEG equipment or data collection techniques, challenges of interpreting child performance on their tasks, small sample sizes and the use of homogenous populations.

Fig 2. Assessment of risk of bias—percentage of papers scoring ‘Yes’, ‘Partial’ and ‘No’ in response to quality appraisal questions.

EEG signatures of cognitive abilities

The data on age range and sample size of participants, tasks used, EEG measure analysed and significant findings extracted from the 51 included studies are summarized in Table 1 and described below. Most commonly reported measures were the EEG power spectrum or the amplitude and latency of ERPs. The former refers to relative fraction of the power spectrum of defined frequency bands (e.g. 6–10 Hz for alpha power). Identification of ERP components depends on the task being used (see S2 Fig for a sample trace in response to faces) and these are defined in two ways: a) based on the order of positive and negative deflections, for instance P1 is the first positive deflection or b) based on the latency with which they occur, such as N170, which is a negative peak around 170ms after stimulus onset. We categorise the domains of cognitive and social development reported in these studies into executive function, selective auditory attention, learning and memory, and processing of faces and emotional stimuli.

Executive function (EF): Visual attention, working memory and inhibitory control.

Fig 3 shows that 22 of 51 studies identified in this review assessed executive function, using multiple tasks: changes in EEG spectral power during Stroop and Yes-No tasks (8/22) was most commonly studied, followed by ERPs in Go/No-Go (6), Flanker (3) and set-shifting tasks (3) like Dimensional Change Card Sort (DCCS) or Ocean Sort tasks (tasks described below). The results in this section begin with studies measuring ERPs followed by spectral power, and start with the most commonly used tasks.

Fig 3. Distribution of papers across domains of cognitive and socio-emotional development.

Go/No-Go task. Six studies used the Go/No-Go task for response inhibition in which participants have to respond in the majority of the task trials (Go condition), but withhold their response when a particular stimulus appears (No-Go condition). Four of these studies were conducted with 5-year old children and the other two included younger age groups. Five studies reported on the amplitude and latency of the frontal negative (specifically including N2) ERP component while two reported on the positive (P1 and P3) ERP components. One study demonstrated left lateralisation of frontal negativity in both Go and No-Go trials [43] and another reported that relative to Go trials, No-Go trials elicited longer N2 latencies [43,44]. NoGo trials also elicited larger negative (N2) and positive (P3b) amplitudes compared to Go trials [43,45,47]. A study used source localisation analytic techniques to demonstrate an asymmetrical pattern of scalp lateralization of N2: right-lateralized in No-go trials and left-lateralized in Go trials [42]. One study investigated change across gender and found no significant association [47]. Brooker et al found no change across age from 3.5-4-year olds but found that the ability of Error-Related Negativity (ERN) at 3-years age to predict ERN at 4-years age was dependent on SES [46].

Flanker task. In the Flanker task participants are instructed to respond to a central relevant stimulus which is ‘flanked’ on either side by irrelevant stimuli that can either be congruent or incongruent with the central stimulus, and is primarily a measure of response inhibition. All three studies using this task in children aged 4.3–5.5 years reported that the negative amplitude of the N components over frontal electrodes (N1 and N2 ERP components) were modulated by the congruency of the trials [4850]. Additionally, the amplitude of the frontal positive ERP component, P3 was larger and with a longer latency in incongruent compared to congruent trials [48]. The P3 amplitude was found to be negatively correlated with performance on the task [50]. Rueda and colleagues demonstrated age-related differences in both negative and positive ERP components (N1, N2 and P3) between children and adults: N1 and N2 amplitudes and N1, N2 and P3 latencies decreased significantly with age [48]. They also demonstrated a change in lateralisation of the larger P3 amplitude and longer latency from the right to left hemisphere as age increased from 4-years age to adulthood. One study investigated the impact of SES on N2 and P3 amplitudes and found no significant association [50].

Set-shifting tasks. Three studies assessed cognitive flexibility through the use of set-shifting tasks like DCCS or Ocean Sort tasks in which participants are expected to start the task by sorting objects based on a particular dimension (like colour) and switch to another dimension (like shape) in the middle of the task. One reported larger P3 amplitude and shorter P2 latency at the right fronto-central and left parieto-central electrode clusters respectively in switch than stay trails in 5-year olds [55]. Another showed that fronto-central N2 amplitude, but not latency was smaller in 3.5-year old children who switched flexibly between sets [52]. The third study found that medium-to-high frontal alpha power moderated the positive relationship between child performance on EF and SFT tasks [53].

Stroop/Yes-No task. Eight papers used the Stroop task, some in conjunction with the Yes-No task, to index executive function in which participants have to remember to give a response which is opposite to the stimulus which they process (for example, saying ‘day’ when they see a black card with a moon, or saying ‘no’ when the experimenter nods their head). All eight studies measured power changes in the alpha frequency band in frontal regions, mostly defined as 6–9 Hz and two of them also measured intra-hemispheric connectivity using inter-channel coherence. Increased baseline to task alpha power in the left medial frontal region, along with language and temperament, were found to predict the performance of 3.5- and 4-year olds on the Stroop task [56,58,59] and, these factors, together with maternal education, also predicted child performance on the Hand Game task which follows principles similar to the Stroop and Yes-No tasks [60]. This group also demonstrated that alpha power increased with increasing executive demands in 4-year old children [62]. A study showed that high performers in EF tasks had increased medial frontal alpha power as compared to low performers in 3.5–4.5-year olds, and this effect was mediated by shyness [61]. Moreover, increased alpha power during EF tasks were demonstrated to be dependent on age decreasing from infancy [57] and 3.5 years [58] to 4.5 years. Baseline to task increase in alpha power and decrease in coherence across electrode pairs, also became more localised across this age range moving from being observed over the entire scalp in infants to more localised scalp regions in the older children [57,58].

One study investigated the impact of gender on alpha power during EF tasks in 4-year olds, and found that girls exhibited higher overall power which was more localised, when compared to boys [62]. Interestingly, another study using the Crayon/Marker Delay task to assess inhibitory control in younger children (2-year olds) also demonstrated the association between frontal alpha power and child performance. They however did not find any differences based on gender [51]. Finally, one study used a battery of six EF tasks including Stroop, Go-NoGo and Working Memory Span to derive a single EF performance score, in 4-year olds and demonstrated that frontal alpha power was unrelated to child performance while a decrease from baseline to task coherence was associated with performance [63].

In summary, 18 studies assessed EF using a variety of tasks and reported on varied metrics including alpha power and the amplitude and latency of ERP components showing significant associations with behavioural performance. Most notably, in 3.5–4.5-year old children, an increase in alpha power from baseline to task was associated with EF as measured by Stroop and Yes-No tasks [57,58,61,62].

Selective auditory attention.

Nine studies measured selective attention using auditory tasks (See Fig 3) in which stimuli such as stories or environmental sounds were presented from two audio sources. Children had to selectively attend to the stimulus from one channel and inhibit their attention from the other. Studies with 5-yr old children showed a broad positivity across all electrode sites that peaked around 100ms after stimulus onset, rather than the positive-negative-positive (P1-N1 complex) ERP waveform characteristic of adults [65,67,69], which does not emerge until early adolescence [69]. While pre-schoolers did not have the ability to differentially attend to stories played in two channels [68], they could selectively attend to environmental sounds [67] as evidenced by the absence [68] and presence [67] of a negative oscillation in the ERP waveform respectively.

Larger P3a amplitude in the fontal and central electrodes to novel attention-catching sounds was found to be associated with temperamental traits like higher effortful control, higher negative emotionality (such as sadness and fearfulness and feelings of discomfort) and more synchronous parent-child interaction in 2-year old children [66]. A larger mean amplitude difference between 100-200ms post stimulus onset in response to attended as compared to unattended stimuli (attention effect) in anterior and central electrode locations was also found to be correlated with superior non-verbal IQ scores in older 4-year old children recruited from low-income households [70]. Interestingly, this group subsequently showed that children from low SES were delayed by one year in development of this attention effect as compared to those from high SES as demonstrated by a 1-year follow up of the low SES sub-group [71]. Larger positive amplitudes were also elicited by distracting sounds in children with lower SES [72].

Learning and memory.

In order to evaluate neural bases of learning and memory, five studies identified in this section compared ERP components in response to familiar or previously viewed objects compared to novel ones. However, the paradigms differed in duration with two investigating immediate recall, one each investigating recall delayed by 5-minutes, one or seven days. ERPs in response to correctly versus incorrectly recalled images were found to be more positive i.e. they had less pronounced negativity, in 2.5-, 3- and 4-year old children [14,73,76]. A study showed that the amplitude of the positive slow wave (PSW) in frontocentral sites decreased with increasing recall of temporal order in which items were presented to 3-4-year old children and PSW amplitude and latency decreased across these ages [73]. Yet another study demonstrated age-related changes in a mid-latency negative component, Nc, and PSW in response to items recalled with contextual details, which increased between 3 and 4 years, and items recalled without contextual details, which were greatest in 5- year-old children. However the location of this effect differed between the age groups, moving from frontal parietal leads in 3-year olds to the left hemisphere leads in 4 year olds [74]. However, another study found no association between Nc in the frontal and central electrodes and age when comparing children younger than 5.8-years with older ones [75].

Face processing.

Faces represent a very important visual stimulus, particularly in social development. This review identified seven studies that measured neural specialisation towards processing of facial information, and these included tasks in which images of upright and inverted, familiar and unfamiliar, and scrambled faces were presented to participants. None of these studies investigated the impact of socio-economic status on face processing. One group reported the response of the power spectrum to faces in 5-year old children [79,80]. All other studies reported ERP components and the results for this section are segregated based on these ERP components.

P1. P1 in the occipital region is a commonly studied ERP component in children that occurs early during the visual processing of faces, or indeed any complex visual stimulus. It is thought to reflect early stage cortical processing of patterns and objects. Its amplitude was larger in response to older than younger faces [78,82] and increased across childhood [77,81,84]. P1 amplitude was also larger on the midline compared to right and left hemispheres in 3.5-year olds [78], but was right lateralised in 4-6-year olds [81]. P1 latency to inverted faces decreased with age from early childhood to adolescence and adulthood in three studies [77,78,84], but not in a fourth which had a more limited age range (5-10-year old children) [81]; and was found to be shorter for inverted compared to upright faces in 3.5-year old [78] and with the reverse being demonstrated in 5-year old children [82].

N170. The N170 is an ERP component recorded over temporal lobe channels that was measured as a strong negative response for faces compared to other objects in adults. From the second year of life, a strong negative inflection occurs over temporal sites in response to faces and its latency gradually decreases to be adult like in late adolescence. Given the continuity in functional properties this is referred to as the N170 through childhood, despite the longer latencies at which it occurs [94,95]. Its amplitude was larger, with shorter latencies, to eyes than faces, in 4-5-year olds and larger over the right hemisphere in all age groups [77]. N170 amplitude was also larger to face than non-face images and to adult than infant faces [77,78]. Melinder and colleagues (2010) showed that viewing faces of children elicited the largest N170 amplitude compared to viewing faces of older people [82]. N170 amplitude decreased from 4 to 8 years of age in one study [81], and from 4–5 years until 12–13 years of age in another [84]. Taylor and colleagues [77] found this pattern of change only in response to eyes. Lateralisation of the N170 response to inverted and upright faces to the left hemisphere was stronger in 4-5-year olds as opposed to adolescents and adults [77] and stronger in the right hemisphere for girls than boys [84]. N170 latency in response to faces decreased with age from 4–10 years [81] and also when 4-5-year old children were compared with adults [77,82,84]. N170 latency and amplitude showed no difference based on gender in younger children (5-10-year olds) [77,81] but was found to be faster and larger in girls than boys in older age groups [77].

Other ERP components. Other than P1 and N170, a few studies reported significant findings in relation to face processing in other ERP components, namely the P2 and the P400 in the occipital region. Larger P2 amplitudes with longer latencies for inverted than upright faces have been shown in 5-year old children [82] and in addition, P2 amplitude and latency in response to faces decrease from 4 to 10 years of age [81]. P400 response is larger to inverted than upright faces in 3.5-year old children [78] and to familiar compared to unfamiliar toys in 2-5-year olds [83].

To summarise, most studies assessing children’s neural processing of faces measured the P1 and N170 ERP components. The majority of these studies reported on changes in these components across age. For instance, P1 latency to inverted faces and the N170 amplitude and latency in response to faces were demonstrated to decrease with age [77,78,81,82,84].

Emotional stimuli processing–faces.

This review identified three studies that investigated the response of face processing ERP components described above to neutral, positive (happy, surprised) or negative (fearful, sad) emotional facial expressions. One study found that the P1 amplitude was higher when observing fearful compared to neutral faces in 3-year olds [85] while another study found no effect of emotion in 4-year olds [84]. The P1 latency was shown to be sensitive to variations in emotional expression in perceived faces in children approximately 5-years old, such that the P1 occurred early in response to neutral and positive emotions, and later for negative ones like fear and disgust, and to gender of participants, with longer latencies to all emotions except sad and angry faces in boys than girls [84]. The impact of emotions on N170 amplitude were demonstrated to be larger for neutral compared to fearful faces in 3-year olds [85] but no effect of emotion on the N170 was found in a study of participants ranging in age from 4–15 years [84]. Finally, a recent study of spectral power found evidence of higher synchronization in the theta band in response to negative emotional expressions in the subset of children for whom negative emotional content was associated with improved cognitive efficiency [86].

Emotional stimuli processing–non-faces.

Of the five studies examining the processing of emotional stimuli other than faces such as cute compared to fierce animals or disaster pictures, the EEG measure studied most often (3/5 studies) was the late positive potential (LPP) in posterior, central and frontal regions. LPP was found to be larger to negative and unpleasant stimuli than neutral and pleasant ones [8890]. Changes in LPP amplitude across development remain to be established. One study reported no effect of age on LPP amplitude in response to negative or neutral interpretations of images in 4 to 5-year olds [90]. On the other hand, using images depicting painful and non-painful situations, LPP amplitude has been demonstrated to increase from 4 to 9 years and adulthood, with no differences found between boys and girls [88]. One study used a unique prize guessing game in which children attached a value of ‘good’ or ‘bad’ to objects and found larger amplitude and longer latency for the P1 in the parietal and the positive slow wave (PSW) in the central parietal areas, but not in feedback-related negativity (FRN) elicited in response to ‘good’ as compared to ‘bad’ objects [91]. The final study in this category was unique in that it analysed alpha (6–8 Hz) power in response to video-clips designed to induce emotions such as sadness, happiness, anger and fear in shy as compared to non-shy children, and found greater frontal asymmetry in shy children in response to fear [87].


This systematic review presents a comprehensive synthesis of studies conducted over the last three decades that have used electroencephalography to measure neural correlates of cognitive and social development in 2-5-year old children. Even across the target age range of this review, participants were found to be unequally distributed with more studies focusing on older 4-5-year old children, an easier age group to collect EEG data from, with significantly lower attrition rates due to cap refusal and less movement artefacts, when compared to younger toddlers [19]. Optimising protocols to engage young children while applying EEG electrodes to reduce attrition rates, and designing age-appropriate tasks, while at the same time keeping gross motor movements to a minimum, presents a challenge that is yet to be completely overcome in this field of study [19].

The heterogeneity of the studies identified in this review is evident from an analysis of the equipment and methods used to collect EEG data and subsequently pre-process and analyse it. Studies differ with respect to the tasks that they use to measure the same cognitive domains and even within the same task used at the same age, use of different stimuli or analysis of different EEG metrics often limits the ability to synthesise their findings and limits their reproducibility [52,55]. A future way forward is to develop a common EEG platform with associated software that allows for compatible data collections across multiple sites and populations [9698].

The assessment of risk of bias conducted in this review demonstrates the strength of included studies in reporting methods and results in relation to outcomes of EEG measures, but also revealed that studies were limited in their reporting of participant details such as recruitment criteria, sociodemographic profile and methods employed to test whether children included in the study were developing appropriate to their age. This field of research would additionally benefit from reporting more details of their study population, and making greater efforts to increase their diversity to allow for generalisability and replicability of their results [99].

Another common limitation of these studies, often acknowledged by authors themselves, is the small sample size, often amplified by the loss of participants either during data collection or analysis. This limitation is, of course, closely linked to the challenge of collecting data on young children as discussed above. This challenge of extensive loss of data highlight the need to take these high attrition rates into account while sampling and use data processing methods that generate robust EEG signatures derived from ‘imperfect’ datasets which would result in more studies with larger sample sizes. Again, harmonization of technology and methods can help with this issue [9698].

Emerging evidence for neural biomarkers of cognitive and social development

Bearing in mind the heterogeneity of the identified studies, this review provides a synthesis of some key insights into neural correlates of cognitive and social development in preschool children. Two studies identified in this review have demonstrated that No-Go N2 amplitudes are larger than Go trials in 2–5 year old children [44,45], a result that has also been found in another systematic review and meta-analysis on 2-12-year old children [31]. Four studies, identified in this review, conducted on 3.5–4.5-year old children demonstrated an increase in alpha power from baseline to task in the medial frontal region as being associated with executive function as measured by Stroop and Yes-No tasks [57,58,61,62]. This finding is unsurprising considering that research has consistently shown significant frontal activation during EF tasks in typically developing infants and children [100102]. In three ERP studies on 3.5–5 year olds investigating selective attention using auditory tasks, a broad positivity has been observed 100ms post stimulus presentation as opposed to the distinct P1-N1 complex which is routinely seen in adults [65,67,69].

The studies described in this review have highlighted the importance of the mid-latency negative component (Nc) and the positive slow wave (PSW) in tasks involving learning and memory. The amplitude of PSW is greater and Nc is smaller (has less pronounced negativity so it is also more positive) in correctly compared to incorrectly recalled items [14,76]. Similar to this finding, the Nc component has been demonstrated in infants to be larger in amplitude to images that capture infants’ attention the most for instance favourite over novel toys and mother’s face over a stranger [103].

Consistent with previously published reviews, the P1 and N170 ERP components have emerged as the most commonly studied responses during processing faces. Taylor and colleagues demonstrated shorter latencies for the P1 in response to upright compared with inverted faces in 4-15-year old children [95]. This is consistent with another study included in this review in 5-year olds [82], while longer P1 latencies to upright than inverted faces were demonstrated in a study with younger participants (3.5-year olds) [78]. This seems to imply that an important developmental change occurs between these ages and warrants further investigation. One study found that the N170 amplitude was larger (more negative) to facial than non-facial stimuli in 4–5 year old children [77]. Many published studies on both children and adults concur with this finding, suggesting that this response can also be interpreted as the N170 being face-sensitive rather than face-selective, such that it is larger to objects of visual expertise, with faces being one of the objects for which most people are experts [95,104]. The N170 was found to be larger in response to fearful than neutral faces in 3-year olds [85]. This is in contrast to a study in 7–13 year old children [105], casting light on the complexity of the changes in this measure across childhood and adolescence.

A total of three studies included in this review examined the late positive potential (LPP) elicited in response to stimuli with emotional valence. More specifically, the LPP was shown to be larger in response to negative and unpleasant stimuli than positive and neutral ones in 4-5-year old children [89,90]. This finding is consistent with results from older 5-7-year old children [106]. Interestingly, the LPP measured in middle childhood has been shown to be predictive of later emotional regulation capacity [107], suggesting its importance as a potential neurophysiological marker of typical emotional development in preschool children.

Changes in EEG measures across development, gender and socioeconomic status

The secondary aims of this review included elucidating changes in EEG measures across age, gender and socioeconomic status. Despite a narrow age-range included in this review, some subtle differences in EEG measures of cognitive and social development across ages have been identified. In particular, the reduction in alpha power in the medial frontal region during executive function tasks increases with age between infancy and 4.5 years [58,61]. Another observation of the review is that N170 amplitude and latency and P1 latency in response to faces decrease with increasing age [77,81,82], consistent with prior reports [95,108]. The decrease across age in P1 and N1 latency and N2 amplitude in response to the Flanker task demonstrated in a study in this review [48] has also recently emerged as a finding of a study conducted after this review search investigating developmental changes of selective attention [109].

Only 8 of the 51 included studies either disaggregated their data by gender or included gender as a variable in the analysis. This is intriguing given there is some evidence that neurophysiological processes differ based on gender. For instance, it has been found that girls have a higher level of synchronisation for all frequency bands than boys during resting state at 5–7 years [110]. This points to a need for a concerted effort by the EEG research community to further understand the scale and nature of brain differences between genders.

Only six studies examined the impact of SES of which five demonstrated its effect on EEG metrics of Go/NoGo and selective auditory attention with larger amplitudes of positive and negative ERP components in children from lower SES [47,71,72]. This larger amplitude manifested as a delay in the development of selective auditory attention in low-SES children when compared to their high-SES peers [71]. However, one study found no impact of SES on attention and inhibitory control as measured by the Flanker Task [50]. The importance of SES highlighted by these few and recently published studies suggests that more research needs to be conducted to investigate risk factors that are known to influence trajectories of neural development and functioning in children [13]. There is some literature, outside the scope of this review, on the impact of low SES on resting state and task-based neural activity using fMRI and EEG; this research shows delayed maturation of both neuronal markers of basic sensory processing, as well as higher order processes such as brain oscillations in frontal regions that index inhibitory control [38,111,112]. However, the majority of this research has focused either on infants or older children and needs to be expanded to include the crucial preschool years. Of note, while some neurophysiological studies have begun emerging from LMICs, this review did not identify any studies using cognitive or social tasks within the target age-range that were conducted in LMICs, highlighting a large research gap.

Limitations of this review

While this review provides some unique insights into the state of developmental EEG research in the context of preschool children, one of its key limitations lies in the relatively narrow age range of 2–5 years, which might not be sufficient to capture developmental changes in some neural markers of emerging cognitive domains. However, we took the view that focusing on this age range brings to light the limited research in this age group compared to the large amount of research done in younger and older children. A second limitation of this review is the focus only on EEG studies assessing cognitive and social domains of development. A review of EEG studies assessing neural responses related to the sensory, language and motor domains were defined as being outside of the scope of this study, yet such functions are of course integral to healthy growth and development of children. A third limitation relates to publication bias as a) only studies published in English language were included in this review, b) only studies with significant results are likely to have been successfully published in peer reviewed journals and, c) within these studies, results of exploratory analyses not attaining significance might not have been reported.

Implications and recommendations for future research

In order to realise the potential that EEG has to be used at scale to measure neurocognitive development in low resource settings, which are home to a disproportionately large number of children at risk of sub-optimal development, there is an urgent need for this field of research to a) identify measures that are robust enough to offer a good signal-to-noise ratio even with lower quality portable EEG systems with lower density arrays, b) move away from the use of a small and homogenous samples to allow for greater generalisability of results, c) to standardise methods and establish best practices for task and stimulus presentation, EEG data collection and pre-processing techniques that are adopted by diverse research groups to allow comparison across studies, d) move outside of the highly controlled laboratory settings in which they are currently being conducted and into community settings or households, where such technology is likely to be implemented at scale. Previous EEG research has shed great light into neurophysiological markers of cognitive function and social-emotional processing in infants at risk of developmental delays (through studies on preterm infants), and disorders like ASD and ADHD (through studies on children at familial risk such as siblings of children diagnosed with these disorders) which have been demonstrated to be predictive of their later manifestation [28,113117]. These findings reinforce the urgent need for more research into the patterns with which these neural processes develop in neurotypical children to enable early identification of those who are at risk of faltering in their development, intervention selection and monitoring of the effectiveness of these interventions [118].

Supporting information

S1 Fig. Age range of participants in included studies.


S2 Fig.

Sample ERP traces–ERPs named based on a) the order of positive and negative deflections (e.g. P1 (highlighted in blue), N1) or b) the latency with which they occur (e.g. N170 (highlighted in red), P300).


S3 Table. Participant details of included studies.


S4 Table. Data acquisition setup of included studies.


S5 Table. Data pre-processing details of included studies.



  1. 1. Lu C, Black MM, Richter LM. Risk of poor development in young children in low-income and middle-income countries: an estimation and analysis at the global, regional, and country level. The Lancet Global Health. 2016 Dec 1;4(12):e916–22. pmid:27717632
  2. 2. McCoy DC, Peet ED, Ezzati M, Danaei G, Black MM, Sudfeld CR, et al. Early Childhood Developmental Status in Low- and Middle-Income Countries: National, Regional, and Global Prevalence Estimates Using Predictive Modeling. PLOS Medicine. 2016 Jun 7;13(6):e1002034. pmid:27270467
  3. 3. Noble KG, Tottenham N, Casey BJ. Neuroscience perspectives on disparities in school readiness and cognitive achievement. Future Child. 2005;15(1):71–89. pmid:16130542
  4. 4. Linan-Thompson S. Screening of School Readiness Skills: A Review of the Literature. 2014 May.
  5. 5. Dasgupta J, Bhavnani S, Estrin GL, Mukherjee D, Banerjee A, Belmonte MK, et al. Translating neuroscience to the front lines: point-of-care detection of neuropsychiatric disorders. Lancet Psychiatry. 2016 Oct;3(10):915–7. pmid:27692259
  6. 6. Storrs C. How poverty affects the brain. Nature News. 2017 Jul 13;547(7662):150. pmid:28703208
  7. 7. Lau-Zhu A, Lau MPH, McLoughlin G. Mobile EEG in research on neurodevelopmental disorders: Opportunities and challenges. Developmental Cognitive Neuroscience [Internet]. 2019 Apr [cited 2019 Nov 15];36. Available from: pmid:30877927
  8. 8. Rahman A, Hamdani SU, Awan NR, Bryant RA, Dawson KS, Khan MF, et al. Effect of a Multicomponent Behavioral Intervention in Adults Impaired by Psychological Distress in a Conflict-Affected Area of Pakistan: A Randomized Clinical Trial. JAMA. 2016 27;316(24):2609–17. pmid:27837602
  9. 9. Green J, Charman T, Pickles A, Wan MW, Elsabbagh M, Slonims V, et al. Parent-mediated intervention versus no intervention for infants at high risk of autism: a parallel, single-blind, randomised trial. Lancet Psychiatry. 2015 Feb;2(2):133–40. pmid:26359749
  10. 10. Britto PR, Lye SJ, Proulx K, Yousafzai AK, Matthews SG, Vaivada T, et al. Nurturing care: promoting early childhood development. The Lancet. 2017 Jan;389(10064):91–102. pmid:27717615
  11. 11. Heckman JJ. Skill Formation and the Economics of Investing in Disadvantaged Children. Science. 2006 Jun 30;312(5782):1900–2. pmid:16809525
  12. 12. Alahmadi N, Evdokimov SA, Kropotov Y (Juri), Müller AM, Jäncke L. Different Resting State EEG Features in Children from Switzerland and Saudi Arabia. Front Hum Neurosci [Internet]. 2016 Nov 2 [cited 2021 Jan 13];10. Available from: pmid:27853430
  13. 13. Alderman H. The economic cost of a poor start to life. Journal of Developmental Origins of Health and Disease. 2010 Feb;1(1):19–25. pmid:25142929
  14. 14. Marshall PJ, Bar-Haim Y, Fox NA. Development of the EEG from 5 months to 4 years of age. Clin Neurophysiol. 2002 Aug;113(8):1199–208. pmid:12139998
  15. 15. Saby JN, Marshall PJ. The utility of EEG band power analysis in the study of infancy and early childhood. Dev Neuropsychol. 2012;37(3):253–73. pmid:22545661
  16. 16. Perone S, Palanisamy J, Carlson SM. Age-related change in brain rhythms from early to middle childhood: Links to executive function. Dev Sci. 2018;21(6):e12691. pmid:29863816
  17. 17. Cuevas K, Bell MA. EEG and ECG from 5 to 10 Months of Age: Developmental Changes in Baseline Activation and Cognitive Processing During a Working Memory Task. Int J Psychophysiol. 2011 May;80(2):119–28. pmid:21338632
  18. 18. Luck SJ. An Introduction to the Event-Related Potential Technique, Second Edition [Internet]. The MIT Press. [cited 2019 Nov 14]. Available from:
  19. 19. Bell MA, Cuevas K. Using EEG to Study Cognitive Development: Issues and Practices. J Cogn Dev. 2012 Jul 10;13(3):281–94. pmid:23144592
  20. 20. Tarullo AR, Obradović J, Keehn B, Rasheed MA, Siyal S, Nelson CA, et al. Gamma power in rural Pakistani children: Links to executive function and verbal ability. Dev Cogn Neurosci. 2017;26:1–8. pmid:28436831
  21. 21. Johnson MH, Gliga T, Jones E, Charman T. Annual research review: Infant development, autism, and ADHD—early pathways to emerging disorders. J Child Psychol Psychiatry. 2015 Mar;56(3):228–47. pmid:25266278
  22. 22. Jones EJH, Gliga T, Bedford R, Charman T, Johnson MH. Developmental pathways to autism: a review of prospective studies of infants at risk. Neurosci Biobehav Rev. 2014 Feb;39:1–33. pmid:24361967
  23. 23. Bosl W, Tierney A, Tager-Flusberg H, Nelson C. EEG complexity as a biomarker for autism spectrum disorder risk. BMC Med. 2011 Feb 22;9:18. pmid:21342500
  24. 24. Bosl WJ, Tager-Flusberg H, Nelson CA. EEG Analytics for Early Detection of Autism Spectrum Disorder: A data-driven approach. Sci Rep. 2018 01;8(1):6828. pmid:29717196
  25. 25. Gurau O, Bosl WJ, Newton CR. How Useful Is Electroencephalography in the Diagnosis of Autism Spectrum Disorders and the Delineation of Subtypes: A Systematic Review. Front Psychiatry. 2017;8:121. pmid:28747892
  26. 26. Marosi E, Harmony T, Sánchez L, Becker J, Bernal J, Reyes A, et al. Maturation of the coherence of EEG activity in normal and learning-disabled children. Electroencephalogr Clin Neurophysiol. 1992 Dec;83(6):350–7. pmid:1281080
  27. 27. Gabard-Durnam LJ, Wilkinson C, Kapur K, Tager-Flusberg H, Levin AR, Nelson CA. Longitudinal EEG power in the first postnatal year differentiates autism outcomes. Nature Communications. 2019 Sep 13;10(1):1–12. pmid:30602773
  28. 28. Fogtmann EP, Plomgaard AM, Greisen G, Gluud C. Prognostic Accuracy of Electroencephalograms in Preterm Infants: A Systematic Review. Pediatrics. 2017 Feb;139(2). pmid:28143915
  29. 29. Kong AHT, Lai MM, Finnigan S, Ware RS, Boyd RN, Colditz PB. Background EEG features and prediction of cognitive outcomes in very preterm infants: A systematic review. Early Hum Dev. 2018;127:74–84. pmid:30340071
  30. 30. Anderson AJ, Perone S. Developmental change in the resting state electroencephalogram: Insights into cognition and the brain. Brain Cogn. 2018;126:40–52. pmid:30144749
  31. 31. Hoyniak C. Changes in the NoGo N2 Event-Related Potential Component Across Childhood: A Systematic Review and Meta-Analysis. Dev Neuropsychol. 2017;42(1):1–24. pmid:28059561
  32. 32. Hileman CM, Henderson H, Mundy P, Newell L, Jaime M. Developmental and individual differences on the P1 and N170 ERP components in children with and without autism. Dev Neuropsychol. 2011;36(2):214–36. pmid:21347922
  33. 33. Thatcher RW. Cyclic cortical reorganization during early childhood. Brain Cogn. 1992 Sep;20(1):24–50. pmid:1389121
  34. 34. Thatcher RW, Walker RA, Giudice S. Human cerebral hemispheres develop at different rates and ages. Science. 1987 May 29;236(4805):1110–3. pmid:3576224
  35. 35. Johnson MH, Haan M de. Developmental Cognitive Neuroscience: An Introduction. 4 edition. Chichester, West Sussex, UK; Malden, MA, USA: Wiley-Blackwell; 2015. 332 p.
  36. 36. Hanlon HW, Thatcher RW, Cline MJ. Gender Differences in the Development of EEG Coherence in Normal Children. Developmental Neuropsychology. 1999 Dec 1;16(3):479–506.
  37. 37. Etchell A, Adhikari A, Weinberg LS, Choo AL, Garnett EO, Chow HM, et al. A Systematic Literature Review of Sex Differences inChildhood Language and Brain Development. Neuropsychologia. 2018 Jun;114:19–31. pmid:29654881
  38. 38. Pavlakis AE, Noble K, Pavlakis SG, Ali N, Frank Y. Brain imaging and electrophysiology biomarkers: is there a role in poverty and education outcome research? Pediatr Neurol. 2015 Apr;52(4):383–8. pmid:25682481
  39. 39. Maguire MJ, Schneider JM. Socioeconomic status related differences in resting state EEG activity correspond to differences in vocabulary and working memory in grade school. Brain and Cognition. 2019 Dec 1;137:103619. pmid:31655309
  40. 40. United Nations Development Programme. Human development report 2019: beyond income, beyond averages, beyond today: inequalities in human development in the 21st century. 2019.
  41. 41. Kmet LM, Lee RC, Cook L. Standard Quality Assessment Criteria for Evaluating Primary Research Papers from a Variety of Fields. In 2004.
  42. 42. Lahat A, Todd RM, Mahy CEV, Lau K, Zelazo PD. Neurophysiological correlates of executive function: a comparison of European-canadian and chinese-canadian 5-year-old children. Front Hum Neurosci. 2009;3:72. pmid:20161697
  43. 43. Chevalier N, Kelsey KM, Wiebe SA, Espy KA. The temporal dynamic of response inhibition in early childhood: an ERP study of partial and successful inhibition. Dev Neuropsychol. 2014;39(8):585–99. pmid:25470223
  44. 44. Abdul Rahman A, Carroll DJ, Espy KA, Wiebe SA. Neural Correlates of Response Inhibition in Early Childhood: Evidence From a Go/No-Go Task. Dev Neuropsychol. 2017;42(5):336–50. pmid:28857635
  45. 45. Hoyniak CP, Petersen IT, Bates JE, Molfese DL. The neural correlates of temperamental inhibitory control in toddlers. Philos Trans R Soc Lond, B, Biol Sci. 2018 19;373(1744). pmid:29483345
  46. 46. Brooker RJ. Maternal Behavior and Socioeconomic Status Predict Longitudinal Changes in Error-Related Negativity in Preschoolers. Child Dev. 2018;89(3):725–33. pmid:29611867
  47. 47. St. John AM, Finch K, Tarullo AR. Socioeconomic status and neural processing of a go/no-go task in preschoolers: An assessment of the P3b. Developmental Cognitive Neuroscience. 2019 Aug 1;38:100677. pmid:31255904
  48. 48. Rueda MR, Posner MI, Rothbart MK, Davis-Stober CP. Development of the time course for processing conflict: an event-related potentials study with 4 year olds and adults. BMC Neurosci. 2004 Oct 22;5:39. pmid:15500693
  49. 49. Begnoche JP, Brooker RJ, Vess M. EEG Asymmetry and ERN: Behavioral Outcomes in Preschoolers. PLoS ONE. 2016;11(5):e0155713. pmid:27223612
  50. 50. Ruberry EJ, Lengua LJ, Crocker LH, Bruce J, Upshaw MB, Sommerville JA. Income, neural executive processes, and preschool children’s executive control. Development and Psychopathology. 2017 Feb;29(1):143–54. pmid:26817409
  51. 51. Morasch KC, Bell MA. The Role of Inhibitory Control in Behavioral and Physiological Expressions of Toddler Executive Function. J Exp Child Psychol. 2011 Mar;108(3):593–606. pmid:20719337
  52. 52. Espinet SD, Anderson JE, Zelazo PD. N2 amplitude as a neural marker of executive function in young children: an ERP study of children who switch versus perseverate on the Dimensional Change Card Sort. Dev Cogn Neurosci. 2012 Feb 15;2 Suppl 1:S49–58. pmid:22682910
  53. 53. Blankenship TL, Broomell APR, Bell MA. Semantic future thinking and executive functions at age 4: The moderating role of frontal brain electrical activity. Developmental Psychobiology. 2018;60(5):608–14. pmid:29785731
  54. 54. Lo Y-H, Liang W-K, Lee H-W, Wang C-H, Tzeng OJL, Hung DL, et al. The neural development of response inhibition in 5- and 6-year-old preschoolers: an ERP and EEG study. Dev Neuropsychol. 2013;38(5):301–16. pmid:23862634
  55. 55. Elke S, Wiebe SA. Proactive control in early and middle childhood: An ERP study. Developmental Cognitive Neuroscience. 2017 Aug 1;26:28–38. pmid:28436833
  56. 56. Wolfe CD, Bell MA. Working memory and inhibitory control in early childhood: Contributions from physiology, temperament, and language. Dev Psychobiol. 2004 Jan;44(1):68–83. pmid:14704991
  57. 57. Bell MA, Wolfe CD. Changes in brain functioning from infancy to early childhood: evidence from EEG power and coherence working memory tasks. Dev Neuropsychol. 2007;31(1):21–38. pmid:17305436
  58. 58. Wolfe CD, Bell MA. Sources of variability in working memory in early childhood: A consideration of age, temperament, language, and brain electrical activity. Cognitive Development. 2007 Oct 1;22(4):431–55.
  59. 59. Wolfe CD, Bell MA. The integration of cognition and emotion during infancy and early childhood: regulatory processes associated with the development of working memory. Brain Cogn. 2007 Oct;65(1):3–13. pmid:17630061
  60. 60. Watson AJ, Bell MA. Individual differences in inhibitory control skills at three years of age. Dev Neuropsychol. 2013;38(1):1–21. pmid:23311312
  61. 61. Wolfe CD, Bell MA. Brain Electrical Activity of Shy and Non-Shy Preschool-Aged Children during Executive Function Tasks. Infant Child Dev. 2014 May 1;23(3):259–72. pmid:24944544
  62. 62. Cuevas K, Calkins SD, Bell MA. To Stroop or Not to Stroop: Sex-Related Differences in Brain-Behavior Associations During Early Childhood. Psychophysiology. 2016 Jan;53(1):30–40. pmid:26681615
  63. 63. Swingler MM, Willoughby MT, Calkins SD. EEG power and coherence during preschoolers’ performance of an executive function battery. Dev Psychobiol. 2011 Dec;53(8):771–84. pmid:21780084
  64. 64. Bartgis J, Lilly AR, Thomas DG. Event-related potential and behavioral measures of attention in 5-, 7-, and 9-year-olds. J Gen Psychol. 2003 Jul;130(3):311–35. pmid:12926516
  65. 65. Sanders LD, Stevens C, Coch D, Neville HJ. Selective auditory attention in 3- to 5-year-old children: an event-related potential study. Neuropsychologia. 2006;44(11):2126–38. pmid:16289144
  66. 66. Pesonen A-K, Huotilainen M, Heinonen K, Komsi N, Putkinen V, Kivikoski L, et al. Brain responses to surprising sounds are related to temperament and parent-child dyadic synchrony in young children. Dev Psychobiol. 2010 Sep;52(6):513–23. pmid:20806324
  67. 67. Sanders LD, Zobel BH. Nonverbal spatially selective attention in 4- and 5-year-old children. Developmental Cognitive Neuroscience. 2012 Jul 1;2(3):317–28. pmid:22516369
  68. 68. Strait DL, Slater J, Abecassis V, Kraus N. Cortical response variability as a developmental index of selective auditory attention. Dev Sci. 2014 Mar;17(2):175–86. pmid:24267508
  69. 69. Karns CM, Isbell E, Giuliano RJ, Neville HJ. Auditory attention in childhood and adolescence: An event-related potential study of spatial selective attention to one of two simultaneous stories. Dev Cogn Neurosci. 2015 Jun;13:53–67. pmid:26002721
  70. 70. Isbell E, Wray AH, Neville HJ. Individual differences in neural mechanisms of selective auditory attention in preschoolers from lower socioeconomic status backgrounds: an event-related potentials study. Dev Sci. 2016;19(6):865–80. pmid:26234822
  71. 71. Hampton Wray A, Stevens C, Pakulak E, Isbell E, Bell T, Neville H. Development of selective attention in preschool-age children from lower socioeconomic status backgrounds. Dev Cogn Neurosci. 2017;26:101–11. pmid:28735165
  72. 72. Giuliano RJ, Karns CM, Roos LE, Bell TA, Petersen S, Skowron EA, et al. Effects of early adversity on neural mechanisms of distractor suppression are mediated by sympathetic nervous system activity in preschool-aged children. Dev Psychol. 2018 Sep;54(9):1674–86. pmid:30148395
  73. 73. Riggins T, Miller NC, Bauer PJ, Georgieff MK, Nelson CA. Electrophysiological indices of memory for temporal order in early childhood: implications for the development of recollection. Dev Sci. 2009 Mar;12(2):209–19. pmid:19143795
  74. 74. Riggins T, Rollins L. Developmental Differences in Memory During Early Childhood: Insights From Event-Related Potentials. Child Dev. 2015 May;86(3):889–902. pmid:25677124
  75. 75. Canada KL, Geng F, Riggins T. Age- and performance-related differences in source memory retrieval during early childhood: Insights from event-related potentials. Developmental Psychobiology [Internet]. [cited 2020 Jun 12];n/a(n/a). Available from: pmid:31876294
  76. 76. Meyer M, Bekkering H, Janssen DJC, de Bruijn ERA, Hunnius S. Neural correlates of feedback processing in toddlers. J Cogn Neurosci. 2014 Jul;26(7):1519–27. pmid:24392905
  77. 77. Taylor MJ, Edmonds GE, McCarthy G, Allison T. Eyes first! Eye processing develops before face processing in children. Neuroreport. 2001 Jun 13;12(8):1671–6. pmid:11409737
  78. 78. Peykarjou S, Westerlund A, Macchi Cassia V, Kuefner D, Nelson CA. The neural correlates of processing newborn and adult faces in 3-year-old children. Dev Sci. 2013 Nov;16(6):905–14. pmid:24118716
  79. 79. Lochy A, de Heering A, Rossion B. The non-linear development of the right hemispheric specialization for human face perception. Neuropsychologia. 2019 18;126:10–9. pmid:28655606
  80. 80. Lochy A, Schiltz C, Rossion B. The right hemispheric dominance for face perception in preschool children depends on the visual discrimination level. Developmental Science. 2020;23(3):e12914. pmid:31618490
  81. 81. Meaux E, Hernandez N, Carteau-Martin I, Martineau J, Barthélémy C, Bonnet-Brilhault F, et al. Event-related potential and eye tracking evidence of the developmental dynamics of face processing. Eur J Neurosci. 2014 Apr;39(8):1349–62. pmid:24517386
  82. 82. Melinder A, Gredebäck G, Westerlund A, Nelson CA. ERP Evidence for an Own-Age Bias. Dev Sci. 2010 Jul;13(4):588–98. pmid:20590723
  83. 83. Carver LJ, Dawson G, Panagiotides H, Meltzoff AN, McPartland J, Gray J, et al. Age-related differences in neural correlates of face recognition during the toddler and preschool years. Dev Psychobiol. 2003 Mar;42(2):148–59. pmid:12555279
  84. 84. Batty M, Taylor MJ. The development of emotional face processing during childhood. Dev Sci. 2006 Mar;9(2):207–20. pmid:16472321
  85. 85. Vlamings PHJM, Jonkman LM, Kemner C. An eye for detail: an event-related potential study of the rapid processing of fearful facial expressions in children. Child Dev. 2010 Aug;81(4):1304–19. pmid:20636697
  86. 86. Jiang Z, Waters AC, Liu Y, Li W, Yang L. Event-related theta oscillatory substrates for facilitation and interference effects of negative emotion on children’s cognition. International Journal of Psychophysiology. 2017 Jun 1;116:26–31. pmid:28237693
  87. 87. Theall-Honey LA, Schmidt LA. Do temperamentally shy children process emotion differently than nonshy children? Behavioral, psychophysiological, and gender differences in reticent preschoolers. Dev Psychobiol. 2006 Apr;48(3):187–96. pmid:16568410
  88. 88. Cheng Y, Chen C, Decety J. An EEG/ERP investigation of the development of empathy in early and middle childhood. Dev Cogn Neurosci. 2014 Oct;10:160–9. pmid:25261920
  89. 89. Hua M, Han ZR, Chen S, Yang M, Zhou R, Hu S. Late positive potential (LPP) modulation during affective picture processing in preschoolers. Biol Psychol. 2014 Sep;101:77–81. pmid:25025638
  90. 90. Hua M, Han ZR, Zhou R. Cognitive Reappraisal in Preschoolers: Neuropsychological Evidence of Emotion Regulation From an ERP Study. Dev Neuropsychol. 2015;40(5):279–90. pmid:26230638
  91. 91. Mai X, Tardif T, Doan SN, Liu C, Gehring WJ, Luo Y-J. Brain Activity Elicited by Positive and Negative Feedback in Preschool-Aged Children. PLOS ONE. 2011 Apr 19;6(4):e18774. pmid:21526189
  92. 92. Shank L. Mullen Scales of Early Learning. In: Kreutzer JS, DeLuca J, Caplan B, editors. Encyclopedia of Clinical Neuropsychology [Internet]. New York, NY: Springer New York; 2011. p. 1669–71. Available from:
  93. 93. Wechsler D. Wechsler Preschool and Primary Scale of IntelligenceTM—Fourth Edition (WPPSITM—IV). Pearson Clinical; 2012.
  94. 94. de Haan M, Johnson MH, Halit H. Development of face-sensitive event-related potentials during infancy: a review. Int J Psychophysiol. 2003 Dec;51(1):45–58. pmid:14629922
  95. 95. Taylor MJ, Batty M, Itier RJ. The faces of development: a review of early face processing over childhood. J Cogn Neurosci. 2004 Oct;16(8):1426–42. pmid:15509388
  96. 96. Gabard-Durnam LJ, Mendez Leal AS, Wilkinson CL, Levin AR. The Harvard Automated Processing Pipeline for Electroencephalography (HAPPE): Standardized Processing Software for Developmental and High-Artifact Data. Front Neurosci [Internet]. 2018 [cited 2020 Feb 5];12. Available from: pmid:29535597
  97. 97. Jones EJH, Mason L, Begum Ali J, van den Boomen C, Braukmann R, Cauvet E, et al. Eurosibs: Towards robust measurement of infant neurocognitive predictors of autism across Europe. Infant Behav Dev [Internet]. 2019 Nov [cited 2020 Feb 5];57. Available from: pmid:31128517
  98. 98. Levin AR, Méndez Leal AS, Gabard-Durnam LJ, O’Leary HM. BEAPP: The Batch Electroencephalography Automated Processing Platform. Front Neurosci [Internet]. 2018 [cited 2020 Feb 7];12. Available from: pmid:30131667
  99. 99. Makel MC, Plucker JA, Hegarty B. Replications in Psychology Research: How Often Do They Really Occur? Perspect Psychol Sci. 2012 Nov;7(6):537–42. pmid:26168110
  100. 100. Orekhova EV, Stroganova TA, Posikera IN. Alpha activity as an index of cortical inhibition during sustained internally controlled attention in infants. Clin Neurophysiol. 2001 May;112(5):740–9. pmid:11336888
  101. 101. Bell MA, Fox NA. The relations between frontal brain electrical activity and cognitive development during infancy. Child Dev. 1992 Oct;63(5):1142–63. pmid:1446545
  102. 102. Moriguchi Y, Hiraki K. Prefrontal cortex and executive function in young children: a review of NIRS studies. Front Hum Neurosci [Internet]. 2013 Dec 17 [cited 2020 Feb 7];7. Available from: pmid:24381551
  103. 103. Reynolds GD, Romano AC. The Development of Attention Systems and Working Memory in Infancy. Front Syst Neurosci [Internet]. 2016 [cited 2020 Feb 7];10. Available from: pmid:26973473
  104. 104. Feuerriegel D, Churches O, Hofmann J, Keage HAD. The N170 and face perception in psychiatric and neurological disorders: A systematic review. Clin Neurophysiol. 2015 Jun;126(6):1141–58. pmid:25306210
  105. 105. de Jong MC, van Engeland H, Kemner C. Attentional effects of gaze shifts are influenced by emotion and spatial frequency, but not in autism. J Am Acad Child Adolesc Psychiatry. 2008 Apr;47(4):443–54. pmid:18356706
  106. 106. Solomon B, DeCicco JM, Dennis TA. Emotional picture processing in children: An ERP study. Developmental Cognitive Neuroscience. 2012;2(1):110–9. pmid:22163263
  107. 107. Babkirk S, Rios V, Dennis TA. The Late Positive Potential Predicts Emotion Regulation Strategy Use in School-Aged Children Concurrently and Two Years Later. Dev Sci. 2015 Sep;18(5):832–41. pmid:25438825
  108. 108. Kuefner D, de Heering A, Jacques C, Palmero-Soler E, Rossion B. Early Visually Evoked Electrophysiological Responses Over the Human Brain (P1, N170) Show Stable Patterns of Face-Sensitivity from 4 years to Adulthood. Front Hum Neurosci. 2010;3:67. pmid:20130759
  109. 109. Reuter E-M, Vieluf S, Koutsandreou F, Hübner L, Budde H, Godde B, et al. A Non-linear Relationship Between Selective Attention and Associated ERP Markers Across the Lifespan. Front Psychol. 2019;10:30. pmid:30745886
  110. 110. Boersma M, Smit DJA, de Bie HMA, Van Baal GCM, Boomsma DI, de Geus EJC, et al. Network analysis of resting state EEG in the developing young brain: structure comes with maturation. Hum Brain Mapp. 2011 Mar;32(3):413–25. pmid:20589941
  111. 111. Hackman DA, Farah MJ. Socioeconomic status and the developing brain. Trends Cogn Sci (Regul Ed). 2009 Feb;13(2):65–73. pmid:19135405
  112. 112. Pietto ML, Kamienkowski JE, Lipina SJ. Electrophysiological Approaches in the Study of the Influence of Childhood Poverty on Cognition. In: Ibáñez A, Sedeño L, García AM, editors. Neuroscience and Social Science: The Missing Link [Internet]. Cham: Springer International Publishing; 2017 [cited 2020 Feb 7]. p. 349–81. Available from:
  113. 113. Black MM, Walker SP, Fernald LCH, Andersen CT, DiGirolamo AM, Lu C, et al. Early childhood development coming of age: science through the life course. Lancet. 2017 07;389(10064):77–90. pmid:27717614
  114. 114. Haartsen R, Jones EJH, Orekhova EV, Charman T, Johnson MH. Functional EEG connectivity in infants associates with later restricted and repetitive behaviours in autism; a replication study. Translational Psychiatry. 2019 Feb 4;9(1):1–14. pmid:30664621
  115. 115. Orekhova EV, Elsabbagh M, Jones EJ, Dawson G, Charman T, Johnson MH, et al. EEG hyper-connectivity in high-risk infants is associated with later autism. J Neurodev Disord. 2014;6(1):40. pmid:25400705
  116. 116. Schiavone G, Linkenkaer-Hansen K, Maurits NM, Plakas A, Maassen BAM, Mansvelder HD, et al. Preliteracy signatures of poor-reading abilities in resting-state EEG. Front Hum Neurosci. 2014;8:735. pmid:25285075
  117. 117. Elsabbagh M, Mercure E, Hudry K, Chandler S, Pasco G, Charman T, et al. Infant neural sensitivity to dynamic eye gaze is associated with later emerging autism. Curr Biol. 2012 Feb 21;22(4):338–42. pmid:22285033
  118. 118. Jones EJH, Dawson G, Kelly J, Estes A, Webb SJ. Parent-delivered early intervention in infants at risk for ASD: Effects on electrophysiological and habituation measures of social attention. Autism Res. 2017;10(5):961–72. pmid:28244271