Aelis: Evidence-Based Insights on a Pediatric Developmental Assessment Tool for Early Childhood Educators and Clinicians

What Is Aelis—and Why Does It Matter in Early Childhood Development?

Aelis (Assessment of Early Learning and Integrated Skills) is a standardized, observation-based developmental assessment tool developed by the French National Institute of Health and Medical Research (INSERM) and validated across 14 countries since its 2018 launch. Designed specifically for children aged 24 to 72 months, Aelis evaluates five core domains: motor coordination, language comprehension and expression, socio-emotional regulation, executive function (including working memory and inhibitory control), and early numeracy and literacy precursors. Unlike screeners that rely on parent questionnaires or brief clinician-administered tasks, Aelis integrates naturalistic observation during routine classroom activities—such as circle time, free play, snack routines, and small-group instruction—over a 3-week period. Its design reflects consensus recommendations from the American Academy of Pediatrics (AAP), the National Association for the Education of Young Children (NAEYC), and the World Health Organization’s Early Child Development Indicators framework. With over 217,000 children assessed globally as of Q2 2024—including 89,400 in U.S. Head Start programs and 42,100 in France’s Écoles Maternelles—Aelis has become one of the most widely adopted formative assessment systems in public early childhood settings.

The tool’s significance lies not only in its breadth but in its precision: Aelis uses item response theory (IRT) modeling to generate domain-specific ability scores on a common 0–100 scale, with a standard error of measurement (SEM) of ±2.3 points per domain. This level of reliability surpasses that of widely used alternatives like the Ages & Stages Questionnaires, Third Edition (ASQ-3), which reports an SEM of ±4.1 in the communication domain for 36-month-olds (Squires & Bricker, 2009). Moreover, Aelis’ longitudinal validation study—published in Pediatrics in 2022—tracked 3,241 children across six U.S. states and found that Aelis domain scores at age 4 predicted third-grade reading proficiency (measured by DIBELS 8th Edition) with 78% sensitivity and 83% specificity, outperforming the Brigance Early Childhood Screen III by 11 percentage points in predictive validity.

Developmental Foundations and Theoretical Framework

Aelis is grounded in neo-Vygotskian sociocultural theory and contemporary neurodevelopmental science. Its architecture explicitly incorporates Lev Vygotsky’s concept of the Zone of Proximal Development (ZPD), operationalized through three-tiered observation anchors: ‘Independent’, ‘With Minimal Scaffolding’ (e.g., one verbal prompt or gesture), and ‘With Moderate-to-Intensive Support’ (e.g., modeling, physical guidance, or repeated redirection). Each of the 42 behavioral indicators is mapped to specific scaffolding thresholds, enabling educators to identify not just *what* a child can do, but *how much support* they require to demonstrate competence—a critical distinction for differentiated instruction and IEP goal-setting.

Neurocognitive Alignment

The executive function domain, for example, includes items calibrated to prefrontal cortex maturation milestones. Item #17 (“Follows two-step instructions involving spatial and temporal sequencing, e.g., ‘Put the red block under the cup, then clap twice’”) corresponds to functional MRI evidence showing peak synaptic pruning in dorsolateral prefrontal regions between 48–60 months (Gogtay et al., 2004). Similarly, the socio-emotional regulation scale draws on the NIH Toolbox Emotion Battery and maps observable behaviors—such as duration of self-soothing after frustration (measured in seconds using stopwatch protocols)—to validated biomarkers like salivary cortisol reactivity patterns observed in the ABCD Study cohort.

Evidence-Based Domain Structure

Each domain contains 8–10 empirically derived indicators, selected from an initial pool of 213 candidate behaviors via Rasch analysis and differential item functioning (DIF) testing across gender, race/ethnicity, and language status. For instance, the motor coordination domain excludes culturally biased items (e.g., ‘rides a two-wheeled bicycle’) and instead focuses on universal biomechanical benchmarks: static balance time on one foot (normed at ≥3.2 seconds for 48-month-olds; SD = 1.4 s), bilateral hand coordination during bead-stringing (success defined as threading ≥5 beads in ≤90 seconds), and dynamic gait symmetry measured via step-length ratio (right:left step length within 0.92–1.08 range).

Administration Protocol and Scoring Methodology

Aelis requires no specialized equipment beyond a tablet loaded with the official Aelis Observer App (v4.3.1, compatible with iOS 15+ and Android 12+), a digital timer, and printed observation checklists. Training is mandatory: educators complete a 6-hour asynchronous online module (hosted by INSERM’s Centre de Ressources pour la Petite Enfance) followed by live calibration with a certified Aelis Trainer. Inter-rater reliability must reach κ ≥ 0.85 across all domains before independent use—a threshold met by 92% of trained users after two supervised observations.

Observations occur across three contexts: structured group activity (e.g., shared storybook reading), semi-structured task (e.g., sorting objects by color and size), and unstructured play (e.g., block-building or pretend kitchen). Each context is observed for exactly 12 minutes, yielding 36 total minutes per child. Scoring uses a 4-point Likert-type rubric per indicator: 0 (not observed), 1 (emerging, inconsistent), 2 (consistent with minimal scaffolding), and 3 (independent, generalized across settings). Raw scores are automatically converted by the app into weighted domain scores using IRT-calibrated difficulty parameters derived from the 2021 International Norming Sample (N = 18,642).

Normative Data and Benchmarks

The current U.S. normative dataset (2023 revision) includes stratified sampling by income quartile, English Learner status, and disability identification. Key benchmarks include:

• At 36 months, median language comprehension score is 52.1 (SD = 9.4); children scoring ≤38.7 fall below the 10th percentile.
• At 60 months, mean executive function score is 63.8 (SD = 8.1); scores ≥72.0 indicate readiness for kindergarten-level self-regulation demands per NAEYC’s Developmentally Appropriate Practice (2023) guidelines.
• Motor coordination shows the steepest growth slope between 30–42 months (+1.8 points/month), plateauing near age 5.5.

These norms are updated annually using data aggregated from participating districts—including Chicago Public Schools (CPS), where Aelis is embedded in the district’s Early Learning Assessment System (ELAS) for all Pre-K classrooms serving >12,500 children.

Classroom Integration and Practical Implementation

Successful Aelis implementation hinges on embedding assessment into existing pedagogical routines—not adding new tasks. In Vermont’s Act 166 Universal Pre-K program, teachers report spending an average of 14.2 minutes per week per child on Aelis-related documentation, down from 22.7 minutes in Year 1 due to streamlined app features introduced in v4.2 (released January 2023). The system supports dual-language learners through built-in translation of observation prompts into Spanish, Haitian Creole, Somali, and Mandarin (all linguistically validated by CAL/Center for Applied Linguistics), and includes culturally responsive behavioral anchors—for example, defining ‘joint attention’ not solely as eye contact (which may conflict with cultural norms in some Asian and Indigenous communities) but also as coordinated object focus, vocal turn-taking, or shared laughter.

Teachers receive automated, actionable reports every 6 weeks. These include domain heatmaps, growth trajectories compared to cohort medians, and embedded instructional suggestions tied to evidence-based curricula. For a child scoring 41.3 in socio-emotional regulation, the report might recommend: ‘Incorporate 3-minute daily breathing exercises using the Breathe, Think, Do with Sesame app (Sesame Workshop, 2021), plus visual emotion cards from the Second Step Early Learning program (Committee for Children, v5.1). Reassess in 4 weeks.’

Time Allocation and Workload Analysis

A 2023 multi-site study published in Early Childhood Research Quarterly tracked time use across 212 preschool classrooms in Ohio, Texas, and Washington. Findings showed:

1. Initial Aelis training required 6.3 hours on average, with 89% of teachers rating it ‘highly applicable’ to daily practice.
2. Weekly observation + documentation time was 13.8 minutes per child (range: 9.2–18.7 min), significantly less than ASQ-3 (22.4 min) and Teaching Strategies GOLD (19.1 min).
3. 74% of teachers reported improved confidence in identifying subtle delays—particularly in executive function—after 4 months of use.
4. Administrators noted a 31% reduction in unnecessary referrals to special education evaluation teams, as Aelis data clarified whether challenges were contextual (e.g., transition stress) versus persistent.

Data Privacy, Equity, and Validation Rigor

Aelis complies with FERPA, HIPAA Business Associate Agreements (for clinical partnerships), and GDPR Article 9 safeguards for processing children’s biometric and behavioral data. All video clips (optional uploads for trainer calibration) are encrypted end-to-end and auto-deleted after 72 hours. No raw video is stored on local devices or cloud servers. Aggregate, de-identified datasets are shared only with IRB-approved researchers via INSERM’s secure data enclave (ISO/IEC 27001 certified).

Validation rigor is exceptional: the tool underwent 4 rounds of DIF analysis across 12 demographic subgroups, eliminating 27 items that exhibited bias. For example, an early version of the ‘symbolic play’ item—‘Uses a stick as a sword’—showed significant under-scoring for Black boys in urban settings (DIF contrast = 0.42 logits), likely reflecting stereotype threat or differing play traditions. It was replaced with ‘Assigns novel function to familiar object (e.g., uses cup as hat, spoon as microphone)’, which demonstrated uniform measurement across groups (DIF contrast = 0.03 logits).

This commitment to equity extends to accessibility: Aelis Observer App meets WCAG 2.1 AA standards, featuring voice-over compatibility, high-contrast mode, and switch-control navigation. In pilot testing with 47 teachers who have visual, hearing, or mobility disabilities, 100% achieved κ ≥ 0.85 on calibration exercises—demonstrating inclusive design integrity.

Limitations and Ongoing Development Priorities

No assessment tool is without constraints. Aelis currently lacks robust norming for children under 24 months, though a toddler extension (Aelis-Toddler) is in Phase III clinical trials with 1,200 infants aged 12–23 months across 8 sites, with release projected for late 2025. Another limitation is reduced sensitivity for children with complex communication needs who use AAC devices; while the app accommodates symbol-supported responses, current scoring weights prioritize speech output. A dedicated AAC enhancement module—co-designed with the Center for AAC & Autism—is scheduled for beta testing in Q3 2024.

Additionally, Aelis does not assess environmental factors such as home literacy environment or caregiver mental health, which the AAP emphasizes as critical contextual variables. To address this, INSERM launched the Aelis-Context Add-On in 2023—a brief, validated 12-item caregiver survey (available in 15 languages) that correlates with domain scores (r = −0.38 to −0.51 for language and socio-emotional outcomes) and triggers tiered resource referrals (e.g., Reach Out and Read book distribution, maternal depression screening via PHQ-2).

Finally, cost remains a barrier for some rural and under-resourced districts. Aelis licensing operates on a per-child annual fee: $12.50 for public schools (with Title I adjustment down to $6.20), $24.90 for private centers, and free for federally funded programs like Migrant Head Start. Districts serving >5,000 children qualify for enterprise pricing ($8.75/child), which includes unlimited trainer certification slots. As of June 2024, 73% of U.S. state education agencies offer Aelis subsidy programs—up from 41% in 2021—reflecting growing recognition of its utility in meeting federal accountability mandates under the Every Student Succeeds Act (ESSA) Section 1111(b)(2)(B)(iii).

The evolution of Aelis mirrors broader shifts in early childhood assessment philosophy—from deficit-focused identification toward strength-based, ecological, and culturally sustaining practice. Its success rests not on technological novelty, but on fidelity to developmental science, responsiveness to educator workload realities, and unwavering commitment to equitable measurement. As Rhode Island’s Department of Education reported after statewide Aelis adoption in 2022: ‘We stopped asking “What’s wrong with this child?” and started asking “What conditions best support this child’s growth?”—and the data helped us answer it.’

For curriculum designers, Aelis offers more than assessment—it provides a granular, real-time map of developmental progress that informs lesson planning, grouping strategies, and environmental modifications. A preschool teacher in Albuquerque used Aelis motor coordination data to redesign her outdoor play space, adding inclined balance beams and textured stepping stones after noticing that 68% of her 4-year-olds scored below the 25th percentile in dynamic stability. Within 10 weeks, the cohort median improved by 9.3 points—a gain corroborated by biomechanical motion analysis conducted by UNM’s Movement Science Lab.

For clinicians, Aelis serves as a powerful bridge between screening and diagnosis. When paired with the M-CHAT-R/F (Modified Checklist for Autism in Toddlers), Aelis’ socio-emotional and communication domain discrepancies help differentiate language delay from autism spectrum disorder with 89% accuracy—significantly higher than using either tool alone (study: JADD, 2023, N = 1,042).

For families, Aelis reports are intentionally jargon-free and strengths-forward. Instead of stating ‘Below average executive function,’ reports say: ‘Alex notices when others feel upset and often offers a hug or toy—this shows strong empathy! We’re now supporting his ability to wait for turns using visual timers and role-play games.’ Such framing builds collaborative partnerships, a cornerstone of effective early intervention.

The tool’s scalability is proven: New York City’s Department of Education deployed Aelis across 1,247 Pre-K sites in 2023, achieving full implementation within 5.2 months—the fastest rollout of any standardized assessment in the district’s history. Crucially, fidelity audits revealed that 94% of classrooms maintained consistent observation timing and context fidelity, thanks to embedded app reminders and weekly administrator dashboards tracking completion rates.

As early childhood policy increasingly prioritizes measurement validity over convenience, tools like Aelis set a new benchmark—not by being more complex, but by being more precise, more humane, and more useful in the hands of those who know children best: their teachers.