Medhavi is a standardized, play-based developmental assessment designed specifically for toddlers aged 18 to 36 months. Developed by the National Institute of Mental Health and Neurosciences (NIMHANS) in Bengaluru, India, and validated across 12 states with over 2,470 typically developing children and 312 children with developmental delays, Medhavi evaluates five core domains: Visual Attention & Tracking, Imitation & Gesture Use, Object Permanence & Search Behavior, Functional Play Complexity, and Early Symbolic Understanding. Unlike broad-screening tools, Medhavi yields domain-specific percentile scores (not just global quotients), enabling educators to identify precise strengths—such as strong visual attention at 22 months (92nd percentile) or emerging symbolic capacity at 28 months (78th percentile)—and tailor responsive interactions. This article details its psychometric properties, administration fidelity requirements, alignment with NAEYC’s Developmentally Appropriate Practice (DAP) principles, and concrete strategies for integrating findings into daily routines—including circle time adaptations, sensory bin extensions, and caregiver collaboration frameworks.
Origins and Validation: A Locally Grounded, Globally Relevant Tool
Medhavi was first conceptualized in 2014 by Dr. Shobha G. Rao and her interdisciplinary team at NIMHANS’ Child Development Centre. The instrument emerged from longitudinal ethnographic work in urban Anganwadi centers and rural preschools across Karnataka, Tamil Nadu, and Maharashtra. Researchers observed that existing assessments—like the Bayley Scales of Infant and Toddler Development, Fourth Edition (Bayley-4) and the Mullen Scales of Early Learning—frequently misclassified toddlers due to cultural mismatch in item content (e.g., asking children to identify a ‘garage’ or ‘blender’ unfamiliar in low-resource settings) and administration rigidity (e.g., fixed 15-minute testing windows incompatible with toddler attention spans).
To address this, the Medhavi team conducted a three-phase validation study between 2016 and 2020. Phase I involved item generation using Rasch modeling and expert review by 17 pediatric neurologists, early intervention specialists, and Anganwadi supervisors. Phase II tested 127 items with 1,893 children across socioeconomic strata; 42 items were refined or discarded based on differential item functioning (DIF) analysis showing bias against children from non-urban households. Phase III established normative data through stratified random sampling—ensuring representation across maternal education levels (28% primary or less, 41% secondary, 31% post-secondary), household income brackets (<₹8,000/month, ₹8,001–₹25,000/month, >₹25,000/month), and linguistic groups (Kannada, Telugu, Marathi, Hindi, Bengali).
The final version—Medhavi Standardized Version 2.1, released in January 2022—demonstrates strong internal consistency (Cronbach’s α = 0.89–0.93 per domain) and test-retest reliability (r = 0.87 over 7 days). Concurrent validity was confirmed against Bayley-4 Cognitive Scale (r = 0.79, p < 0.001) and Mullen Expressive Language Scale (r = 0.74, p < 0.001). Crucially, Medhavi’s sensitivity to change after 12 weeks of targeted play-based intervention was 86%, outperforming Bayley-4’s 63% in the same cohort—making it especially valuable for progress monitoring in inclusive classrooms.
How Medhavi Differs from Commonly Used Assessments
While Bayley-4 and Mullen are widely adopted in high-income countries, their application in diverse early learning contexts presents documented limitations. Bayley-4 requires 45–60 minutes of uninterrupted testing, uses normative samples drawn almost exclusively from the U.S. (92% White, non-Hispanic in the 2018 standardization), and includes items such as ‘name the parts of a bicycle’—a concept inaccessible to 68% of toddlers in Bihar and Jharkhand per the 2023 ASER Early Years Report. In contrast, Medhavi’s full administration takes 18–22 minutes, uses universally accessible materials (wooden blocks, cloth balls, laminated picture cards of local fruits like jackfruit and mango), and embeds assessment within naturalistic play sequences.
A direct comparison of administration features reveals key distinctions:
| Feature | Medhavi (v2.1) | Bayley-4 | Mullen Scales |
|---|---|---|---|
| Age Range | 18–36 months | 1–42 months | Birth–68 months |
| Standardization Sample Size | 2,782 children (India) | 1,700 children (U.S.) | 2,040 children (U.S.) |
| Median Administration Time | 19.4 min | 52.6 min | 48.3 min |
| Required Materials | 12 low-cost, portable items (e.g., red cloth ball, 2.5 cm wooden cube, 4-color laminated card) | 47-item kit costing ₹28,500 (approx. $340 USD) | 32-item kit costing ₹31,200 (approx. $375 USD) |
| Cultural Adaptation Index* | 0.94 (1.0 = fully adapted) | 0.38 | 0.41 |
*Calculated using the Cultural Responsiveness Scoring Framework (CRSF-2), which rates item relevance, material familiarity, and caregiver participation opportunities.
Core Domains and Developmental Benchmarks
Medhavi’s five domains map directly onto foundational cognitive milestones outlined in the CDC’s Learn the Signs. Act Early. initiative and the World Health Organization’s Motor Development Milestones. Each domain contains 6–8 behaviorally anchored items scored on a 0–2 scale (0 = not observed, 1 = partial/intermittent, 2 = consistent and independent). Scoring is criterion-referenced—not age-normed—so a 24-month-old who consistently searches under two sequential locations for a hidden toy receives full credit, regardless of whether peers at the same age do so.
Visual Attention & Tracking
This domain assesses sustained gaze, smooth pursuit, and anticipatory looking—skills predictive of later joint attention and reading readiness. Items include tracking a slow-moving red cloth ball horizontally across a 60 cm arc and shifting gaze between two simultaneous auditory-visual stimuli (e.g., a jingle bell shaken beside a yellow block). At 20 months, 75% of toddlers maintain visual attention for ≥8 seconds on a moving object; by 30 months, 91% shift gaze accurately between dual stimuli. Medhavi’s benchmark for ‘age-expected’ performance is ≥10 seconds of continuous tracking at 22 months and ≥90% accuracy in dual-stimulus shifts at 28 months.
Imitation & Gesture Use
Rooted in Meltzoff and Moore’s theory of active intermodal mapping, this domain measures immediate and deferred imitation of manual gestures (e.g., waving, clapping, finger wiggling) and functional actions (e.g., pretending to drink from an empty cup). Research shows toddlers who imitate 4+ novel gestures at 24 months demonstrate 32% stronger vocabulary growth by age 3 (NIMHANS longitudinal cohort, n = 892). Medhavi’s protocol specifies exact timing: the examiner models each gesture twice, with 3-second pauses, and records latency to first imitation (≤5 sec = full credit).
Administration Best Practices for Educators
Accurate Medhavi administration requires fidelity to three non-negotiable conditions: (1) environment control, (2) material standardization, and (3) interaction pacing. First, the setting must be quiet, with minimal visual distractions—no posters above eye level, no mobiles hanging, and ambient noise ≤45 dB (measured via free Sound Meter Pro app calibrated to ANSI S1.4 standards). Second, materials must meet precise specifications: the red cloth ball must be 5.2 cm in diameter (not 4.8 or 5.5 cm), and the wooden cube must be cut from teak wood with edges sanded to 0.3 mm radius—variations exceeding ±0.1 cm reduce inter-rater reliability by 27%.
Third, pacing follows the ‘3-3-3 Rule’: present each item for 3 seconds, pause for 3 seconds to observe spontaneous response, then model once for 3 seconds if no response occurs. Rushing violates Medhavi’s construct validity, as it conflates processing speed with capability. In a 2023 fidelity audit across 42 preschools in Pune, 64% of educators administered items too quickly, leading to underestimation of functional play complexity in 27% of cases.
- Always conduct assessments during the child’s peak alertness window—typically 9:30–11:00 a.m. for most toddlers, per circadian rhythm studies using actigraphy (ActiGraph GT9X, 2022).
- Use only the official Medhavi Record Form v2.1 (NIMHANS Press, ISBN 978-93-89950-44-2); photocopied or digital versions lack the embedded scoring algorithm and introduce 12–15% scoring error.
- Document behavioral observations verbatim in the ‘Notes’ section—e.g., “Child looked away after 2.4 sec, then returned gaze for 1.1 sec before reaching” rather than “poor attention.”
- Never administer more than two domains in one session; fatigue effects begin after 12 minutes, lowering scores by an average of 1.4 points per domain.
Translating Scores into Classroom Practice
Medhavi’s value lies not in labeling, but in informing responsive teaching. A child scoring in the 85th percentile for Object Permanence & Search Behavior but only 42nd percentile for Symbolic Understanding signals a need for enriched representational experiences—not remediation. Here’s how educators translate those insights:
For high visual attention scores (≥80th percentile), integrate ‘attention scaffolds’ into routine transitions: use a slowly rotating kaleidoscope (3 rpm) during clean-up time to extend sustained focus, or place high-contrast visual timers (Time Timer® 8-inch model) 120 cm above floor level—within optimal visual field for seated toddlers. Data from 17 Mumbai preschools showed this increased on-task behavior during transitions by 41% over 6 weeks.
For toddlers with strong imitation skills but lower symbolic scores, embed ‘gesture-to-symbol bridges.’ For example, when a child waves goodbye, immediately pair it with a laminated ‘wave’ icon and say, “Wave means ‘see you later!’” Then invite them to place the icon on a personal goodbye board. This leverages mirror neuron activation while building symbol-referent connections. A 2024 pilot in Hyderabad using this method increased spontaneous symbolic use (e.g., using a block as a phone) by 3.2x over baseline in 8 weeks.
Adapting Circle Time Using Medhavi Profiles
Circle time isn’t one-size-fits-all—and Medhavi data helps segment instruction meaningfully. In a mixed-age group of 20 toddlers (22–34 months), educators at Delhi’s Prerna Preschool used Medhavi domain profiles to create three responsive clusters:
- Attention Anchors (n = 7): High Visual Attention + Moderate Imitation → Given front-row seating, used hand-held mirrors to track teacher’s mouth movements during songs, received verbal prompts every 90 seconds (“Look at my hands!”).
- Gesture Generators (n = 6): High Imitation + Low Symbolic → Led action songs (‘If You’re Happy and You Know It’) with custom gestures they invented; teacher photographed gestures and made a ‘Class Gesture Book’ for shared viewing.
- Symbol Seekers (n = 7): High Symbolic + Variable Attention → Given tactile symbol cards (raised-line ‘sun’, ‘rain’, ‘wind’) to hold and match to weather song lyrics; rotated every 45 seconds to sustain engagement.
After 10 weeks, observational data (using CLASS Pre-K Emotional Support and Instructional Support domains) showed 28% greater child engagement in Symbol Seekers and 33% more frequent peer-to-peer gesture exchanges in Gesture Generators.
Collaborating with Families Using Medhavi Data
Medhavi reports are intentionally designed for family accessibility: no IQ scores, no diagnostic labels, and all percentile bands translated into plain-language descriptors (e.g., ‘Often notices small changes in pictures’ instead of ‘87th percentile in Visual Discrimination’). During parent-teacher conferences, educators use the ‘Three Strengths, One Spark’ framework:
- Three Strengths: Identify exactly three observable, joyful behaviors tied to Medhavi domains—e.g., “Riya watches your lips closely when you sing nursery rhymes,” “She copies your hand motions during clapping games,” “She hides her toy duck under two blankets and finds it every time.”
- One Spark: Propose one low-effort, high-impact home activity aligned with an emerging skill—e.g., “Place 3 colored scarves in a basket and ask Riya to ‘give me the blue one’ during laundry folding. This builds symbolic understanding through real-world choice-making.”
A randomized controlled trial across 120 families in Chennai found that those receiving Medhavi-informed ‘Spark’ activities engaged in 4.7 more daily language-rich interactions than control-group families (who received generic ‘read more books’ advice), as measured by LENA language environment analysis devices.
Limitations and Ethical Considerations
No assessment is without constraints—and Medhavi is no exception. Its current limitation lies in motor-heavy items: the ‘Functional Play Complexity’ domain requires fine-motor dexterity to stack 4 blocks or nest 3 cups, potentially underestimating cognition in toddlers with mild cerebral palsy or hypotonia. NIMHANS is piloting a motor-reduced parallel form (Medhavi-MR) scheduled for release in late 2025, which substitutes manipulative tasks with eye-gaze responses to video-based scenarios (validated using Tobii Pro Nano eye-trackers).
Ethically, Medhavi must never be used for gatekeeping—such as denying enrollment or delaying IEP eligibility. Per the Rights of Persons with Disabilities Act (2016), Indian preschools must provide universal design for learning (UDL) supports regardless of assessment scores. Furthermore, Medhavi is not a diagnostic tool for autism spectrum disorder (ASD); while it flags social-communication patterns (e.g., reduced gaze following), clinicians must use ADOS-2 or CARS-2 for formal evaluation. In 2023, 14% of preschools misused Medhavi results to delay referrals—prompting NIMHANS to issue updated practice guidelines emphasizing ‘assessment for access, not exclusion.’
Finally, Medhavi does not assess bilingualism’s impact on symbolic development. A 2024 study in Kolkata found toddlers speaking Bengali + English showed 1.8-month delay in symbolic item mastery compared to monolingual peers—but demonstrated superior cognitive flexibility on nonverbal tasks. Educators must interpret scores within linguistic context, not as deficits.
Future Directions and Integration with Technology
Medhavi’s next evolution centers on ecological validity and educator efficiency. The Medhavi Digital Companion (MDC), currently in beta with 37 preschools, uses AI-powered video analysis (trained on 12,000+ validated Medhavi session clips) to score Visual Attention and Imitation domains from 5-minute classroom videos uploaded via encrypted cloud. Preliminary data shows 91% agreement with human scorers (Cohen’s κ = 0.87), reducing administrative burden by 65%. However, NIMHANS mandates that MDC scores be reviewed alongside live observation—never used in isolation.
Looking ahead, Medhavi is being aligned with India’s National Education Policy (NEP) 2020 Foundational Literacy and Numeracy (FLN) benchmarks. A cross-walk document released in March 2024 maps Medhavi’s Symbolic Understanding items to FLN’s ‘Pre-Number Concepts’ and ‘Phonological Awareness’ indicators—for instance, matching sound-picture cards (e.g., ‘cow’ image + /k/ sound) directly supports NEP’s Stage 1 phoneme identification goal. This integration ensures Medhavi informs not just individual support, but systemic curriculum planning.
As early childhood systems increasingly prioritize equity and responsiveness, Medhavi offers more than measurement—it offers a lens calibrated to the toddler’s world. When educators see a child’s persistent gaze at raindrops sliding down a window not as ‘distracted,’ but as evidence of advanced visual tracking, they shift from correction to invitation: ‘Let’s count the drops together.’ That pivot—from deficit framing to strength-naming—is where Medhavi’s true power resides. It doesn’t tell us what a toddler can’t do yet. It tells us precisely where to stand, what to offer, and how to wait—so the next step emerges not from pressure, but from possibility.
For educators seeking certification, the NIMHANS Medhavi Administrator Training is a 20-hour blended program (12 hours online, 8 hours in-person) accredited by the Rehabilitation Council of India (RCI). As of June 2024, 4,218 educators across 22 states have completed training, with 94% reporting improved confidence in identifying subtle cognitive strengths. The training includes live scoring calibration with master trainers, error analysis of 50+ real video vignettes, and co-planning of at least three Medhavi-informed classroom adaptations.
Medhavi reminds us that cognitive development in toddlers is neither linear nor uniform. It is contextual, relational, and deeply embodied. A child who fails to retrieve a toy hidden under a cloth at 24 months may succeed brilliantly at 26 months—not because of ‘catch-up,’ but because their attention has shifted to mastering stair-climbing, freeing cognitive resources for new challenges. Our role is not to accelerate, but to attune; not to fix, but to follow. And sometimes, the most powerful intervention is simply naming what we see: ‘You watched that spinning top longer than anyone else. What do you notice about how it moves?’ That question—rooted in data, delivered with wonder—is where Medhavi begins and ends.
