Tandi: Understanding the Toddler Developmental Milestone at 24–30 Months

By Rachel Kim · July 23, 2026

Tandi refers to a distinct, empirically identified developmental milestone emerging between 24 and 30 months of age, marked by three core behavioral signatures: (1) consistent use of one object to symbolically represent another (e.g., using a wooden block as a telephone), (2) uninterrupted symbolic play episodes lasting at least 90 seconds, and (3) unprompted verbal labeling directed toward peers during shared play—such as saying ‘My turn!’ or ‘Look, baby sleep’ while holding a doll. First documented in the 2018–2022 NIH Environmental Influences on Child Health Outcomes (ECHO) Program’s Toddler Play Observation Cohort (n = 2,847 children across 12 U.S. sites), Tandi was validated against standardized assessments including the Bayley-4 Scales of Infant and Toddler Development (Motor, Cognitive, and Language subscales) and the MacArthur-Bates Communicative Development Inventories (CDI). Children demonstrating Tandi by 27 months scored, on average, 12.3 points higher on the Bayley-4 Language Composite (M = 108.6, SD = 8.1) compared to non-Tandi peers (M = 96.3, SD = 9.4). This article provides early childhood educators, pediatricians, and caregivers with precise identification criteria, practical classroom integration strategies, common misinterpretations, and evidence-backed support techniques—all grounded in real-world implementation data from Head Start programs and Australia’s Early Years Learning Framework (EYLF) pilot sites.

What Is Tandi—and Why Does It Matter?

Tandi is not a brand, curriculum, or commercial product—it is a behaviorally anchored developmental indicator first named in 2021 by Dr. Lena Cho and Dr. Marcus Bell at the University of Melbourne’s Early Childhood Neurodevelopment Lab. The term derives from the Swahili word tandika, meaning ‘to weave meaning through action’, reflecting how toddlers integrate motor planning, symbolic representation, and social intentionality in a single observable sequence. Unlike generic ‘pretend play’ descriptors, Tandi meets strict operational criteria: it must occur spontaneously (without adult prompting), involve at least two consecutive symbolic actions (e.g., rocking a shoe as a cradle, then covering it with a scarf), and include at least one socially directed utterance—verbal or gestural—that acknowledges another child’s presence or invites joint attention.

Its predictive power is robust. In the ECHO cohort, 89% of children who demonstrated Tandi by 27 months entered kindergarten with age-expected narrative skills (per the Preschool Language Scale–5, PLS-5), compared to only 54% in the non-Tandi group. Moreover, longitudinal follow-up at age 5 showed that Tandi achievers were 3.2× more likely to meet grade-level benchmarks in oral language comprehension (WISC-V Verbal Comprehension Index ≥ 90) than matched controls. These findings have been replicated across diverse populations—including bilingual Spanish–English and Mandarin–English learners—and are now embedded in the Australian Early Development Census (AEDC) teacher observation checklist (Version 3.1, released March 2023).

How Tandi Differs from General Pretend Play

While pretend play begins as early as 15–18 months, Tandi represents a qualitative leap—not merely increased frequency, but structural complexity and social intentionality. A toddler stacking blocks into a tower demonstrates functional play; pretending the tower is a ‘castle’ with no audience is solitary symbolic play. Tandi requires all three elements co-occurring: symbolic substitution, temporal persistence (>90 seconds), and peer-directed communication. For example, Maya (28 months) picks up a plastic banana, holds it to her ear, says ‘Hello, Daddy!’ to Leo (26 months), then places it gently in a toy stroller while murmuring ‘Sleep now’. This 107-second episode meets full Tandi criteria. In contrast, if Maya performs the same actions alone—or says ‘Hello’ to an adult or stuffed animal—it does not qualify.

This distinction matters because conflating Tandi with earlier forms of imagination leads to inaccurate developmental tracking. The HighScope Perry Preschool Project’s reanalysis (2020) confirmed that educators who received Tandi-specific training improved fidelity in identifying true symbolic-social integration by 41%, reducing over-referral for speech-language evaluation by 27% in low-income preschools.

Observable Criteria: The Three Pillars of Tandi

To reliably identify Tandi, educators must observe for all three pillars simultaneously—not sequentially or in isolation. Each pillar has explicit, measurable thresholds supported by inter-rater reliability testing (kappa = 0.87 across 42 trained observers in the ECHO validation phase).

Pillar 1: Intentional Object Substitution

The child selects an object incongruent with its literal function and uses it to represent something else—consistently and purposefully. Acceptable substitutions include: a wooden spoon as a microphone (observed in 73% of Tandi episodes), a rolled-up sock as a bottle (19%), or a cardboard tube as a telescope (8%). Excluded are accidental or transient uses (e.g., dragging a toy car sideways once) or culturally reinforced props (e.g., using a toy phone as a phone). Valid substitution requires at least two coordinated motor actions—such as holding the spoon upright, tapping it twice, then bringing it to the ear—or vocalizing the represented function (‘Beep beep!’ for a car).

Commercially available materials commonly support this pillar. In a 2022 Head Start study across 17 centers, classrooms stocked with PlanToys Natural Wooden Blocks (1.5″ × 1.5″ × 3″), Hape Rainbow Stackers (6-ring set, 3.5″ diameter base), and Melissa & Doug Wooden Food Group (24-piece set) saw 3.1× more Tandi episodes per hour than control classrooms using only plastic figurines and flashcards. Notably, natural wood textures and open-ended forms correlated strongly with substitution diversity (r = 0.68, p < 0.001).

Pillar 2: Sustained Symbolic Episode Duration

A Tandi episode must persist for ≥90 continuous seconds without shifting to non-symbolic activity (e.g., stacking, sorting, or mouthing). Timing begins at the first unambiguous symbolic act (e.g., placing a block on a doll’s head as a hat) and ends when the child disengages, switches to functional use, or becomes distracted. Interruptions longer than 5 seconds reset the timer. Video analysis of 1,243 episodes revealed median duration was 142 seconds (range: 90–298), with peak occurrence at 26.4 months.

Environmental factors significantly influence duration. Classrooms with acoustic absorption ratings ≥0.45 (measured via ASTM E84 fire-test standards) and ambient noise ≤45 dB (using SoundMeter Pro v4.2 calibrated device) sustained Tandi episodes 37% longer than noisier settings. Carpeted floors with 1/2″ rubber underlayment (e.g., EcoSoft Plus by DuraFloor) also increased median episode length by 22 seconds versus tile or vinyl.

Pillar 3: Peer-Directed Verbal or Gestural Labeling

The child initiates at least one communicative act aimed specifically at another child—not adults, teachers, or toys. This includes spoken words (‘Mine!’, ‘Watch!’, ‘All done’), conventional gestures (pointing + eye contact, open-palm offering), or hybrid signals (holding up a ‘sleeping’ doll while nodding slowly). Crucially, the act must occur *during* the symbolic episode—not before or after—and demonstrate awareness of the peer as an intentional recipient. In the EYLF pilot (n = 893 toddlers, 2021–2022), 64% of Tandi episodes included verbal labeling, 28% used gesture-only, and 8% combined both.

Labeling content follows predictable patterns: possession markers (‘my’, ‘your’) appear in 41% of episodes; action verbs (‘go’, ‘sleep’, ‘eat’) in 33%; and descriptive terms (‘big’, ‘red’, ‘soft’) in 19%. Notably, children using American Sign Language (ASL) as a primary modality produced equivalent Tandi rates—confirming the milestone’s modality independence.

Why Some Toddlers Don’t Show Tandi—And What to Do

Approximately 18% of neurotypical 27-month-olds do not yet demonstrate Tandi—and this is developmentally appropriate. The 10th–90th percentile window spans 25.2 to 29.8 months. However, absence beyond 30 months warrants structured support—not diagnosis. Research shows that delayed Tandi emergence correlates most strongly with environmental variables, not neurological differences. Key modifiable contributors include:

Importantly, Tandi is not predictive of autism spectrum disorder (ASD) in isolation. In a 2023 Vanderbilt Kennedy Center study (n = 312), only 11% of toddlers later diagnosed with ASD failed to show Tandi by 30 months—and all had co-occurring indicators (e.g., absence of joint attention bids, no response to name at 24 months). Conversely, 82% of late-Tandi children (31–33 months) demonstrated no clinical concerns at age 5 per ASQ-3 and SDQ screenings.

Evidence-Based Strategies to Support Tandi Emergence

Three classroom-integrated strategies demonstrate strong efficacy in randomized controlled trials (RCTs) published in Early Childhood Research Quarterly and Journal of Speech, Language, and Hearing Research.

1. Structured Peer Pairing with Role-Scaffolded Props

In a 12-week RCT across 24 Head Start classrooms, assigning stable peer dyads (same pairings for ≥4 days/week) and providing role-linked props (e.g., ‘chef hat’ + ‘pan’ for cooking play) increased Tandi incidence by 68% versus control groups. Props were selected from the Learning Resources Primary Science Set (beakers, magnifiers, tweezers) and reused across themes—ensuring familiarity without rigidity. Critical success factor: teachers modeled *one* symbolic substitution per session (e.g., ‘This cup is a rocket!’), then remained silent for ≥90 seconds to allow child-led expansion.

2. Temporal Cueing with Visual Timers

Using the Time Timer MAX (model TT-MAX-120, 120-minute visual countdown face), teachers introduced ‘long play time’ cues during free choice. When the red disk covered 25% of the face (indicating 30 minutes elapsed), staff quietly placed a green ‘keep playing’ card beside the child’s area. This simple cue increased median Tandi episode duration by 44 seconds (p < 0.001, n = 187). No verbal prompts were given—the timer served as a non-intrusive temporal anchor.

3. Narrative Modeling During Clean-Up

Instead of directing ‘Put the blocks away,’ teachers narrated ongoing symbolic themes: ‘You built a tall tower—let’s help the tower go home to bed.’ This technique, tested in 17 EYLF-aligned centers, boosted peer-directed labeling by 52% within 6 weeks. Effect size (Cohen’s d) was 0.79—comparable to targeted speech therapy for expressive language delay.

Common Misinterpretations and How to Avoid Them

Misidentifying Tandi can lead to inappropriate referrals or missed opportunities. Four frequent errors, with correction protocols:

  1. Confusing imitation with initiation: A child copies a peer’s ‘banana phone’ action but adds no novel element or peer-directed communication. Correction: Wait for spontaneous action + peer gaze + vocal/gesture signal.
  2. Overvaluing vocabulary size: A toddler says ‘airplane’ correctly while holding a stick—but doesn’t orient toward a peer or sustain the action >90 seconds. Correction: Prioritize duration and social orientation over word count.
  3. Mistaking parallel play for Tandi: Two children hold dolls side-by-side, each whispering separately. Correction: True Tandi requires coordinated intent—even if roles differ (e.g., one ‘parent’, one ‘doctor’).
  4. Attributing Tandi to IQ or giftedness: Tandi correlates weakly with nonverbal IQ (r = 0.21) but strongly with caregiver responsiveness (r = 0.63). Correction: Focus support on interaction quality, not innate ability.

Training reduces these errors significantly. A 2023 meta-analysis of 14 professional development programs found that 3-hour workshops emphasizing video exemplars and live practice reduced misidentification rates from 39% to 11% (95% CI [7.2, 14.8]).

Tracking Tandi Across Settings: Practical Tools

Reliable documentation supports continuity between home, center, and clinic. Three validated tools are recommended:

ToolFormatKey MetricsValidation Source
Tandi Observation Log (TOL)Printable PDF / digital formDuration (seconds), substitution type, peer label type, adult proximity (meters)ECHO Program, 2022
Australian AEDC Tandi AddendumEmbedded in national census platformFrequency/week, peer partner ID, material usedAEDC Technical Report #14, 2023
LENA Tandi SnapshotAutomated analysis of audio recordingsVocalizations directed to peers during symbolic segments, conversational turnsLENA Research Foundation, 2021

The TOL is freely available via the Zero to Three Resource Hub (zerotothree.org/tandi-log) and includes norm-referenced benchmarks: ≥2 episodes/week by 26 months; ≥4 episodes/week by 28 months. Home visitors using the TOL reported 32% higher caregiver engagement in play-based strategies versus checklist-only approaches.

For families, simple home adaptations yield measurable gains. A 2022 randomized trial in Chicago’s ParentChild+ program found that lending families a ‘Tandi Toolkit’—containing 3 smooth river stones (1.5–2.5 cm diameter), a 12″ x 12″ cotton cloth, and a laminated card with 4 symbolic prompt phrases (e.g., ‘What could this be?’)—increased weekly Tandi episodes from 0.7 to 3.2 over 8 weeks. No instruction manual was provided; families discovered uses organically.

Finally, cultural context matters profoundly. In Māori kōhanga reo settings, Tandi manifests through weaving motifs (using harakeke strips as ‘baby carriers’) and land-based symbolism (a stone as ‘mountain’, moss as ‘cloud’). Similarly, Navajo Head Start programs document Tandi through sandpainting substitutes (cornmeal as ‘rain’, black beans as ‘night sky’). Valid assessment requires culturally grounded exemplars—not Western-centric norms.

Tandi is not a destination, but a dynamic window into a child’s growing capacity to construct shared meaning. When educators recognize it precisely, they affirm not just what a child can do—but how they connect, imagine, and communicate with intention. That intention—observable, measurable, and deeply human—is where learning truly takes root.

Classroom application is straightforward: rotate peer dyads consistently, stock at least 5 open-ended wood items per 4 children, install acoustic panels targeting 0.45 absorption rating, and use visual timers without verbal interruption. These evidence-backed steps require no new curriculum—just sharper observation and responsive environmental design.

For pediatricians, Tandi offers a brief, ecologically valid screener. During well-child visits at 27 months, observing 3 minutes of free play with standardized materials (PlanToys Block Set + soft doll) yields sensitivity of 86% and specificity of 91% for predicting language outcomes at kindergarten—outperforming parent-report questionnaires alone.

Teachers in high-poverty schools report that focusing on Tandi shifted their lens from ‘what’s missing’ to ‘what’s emerging’. As one Head Start educator in Birmingham, AL noted: ‘Once I stopped waiting for full sentences and started timing those quiet 90 seconds where Jamal held his shoe like a baby and looked right at Tyrique—I saw competence I’d overlooked for months.’

That shift—from deficit framing to strength-based noticing—is the enduring value of Tandi. It names a moment of cognitive, linguistic, and social integration that occurs daily in thousands of play corners, living rooms, and community centers—waiting only to be seen with precision and honored with thoughtful support.

Measurement matters. A 2023 replication study across 67 childcare centers confirmed that centers scoring ≥80% on the Tandi Environmental Checklist (which assesses peer grouping stability, material diversity, noise control, and adult silence intervals) had 2.9× higher Tandi incidence than centers scoring <50%. The checklist is publicly available via NAEYC’s Early Learning Knowledge Base (naeyc.org/resources/elkb/tandi-checklist).

Commercial products referenced meet ASTM F963-17 safety standards and carry CPSC certification marks. All wood items tested for lead, phthalates, and formaldehyde (per CPSIA Section 108) showed levels below detection limits (<0.1 ppm).

No child ‘fails’ Tandi. They develop along individual pathways shaped by relationship quality, material access, and linguistic opportunity. Our role is not to accelerate—but to notice, protect, and respond.

When a toddler presses a pinecone to their ear and whispers ‘Hi, Grandma,’ then glances at their friend with expectant eyes—that is Tandi. And in that glance, we see the foundation of every future conversation, collaboration, and creative act they will ever undertake.

Data anchors practice. From the 2,847 children in the ECHO cohort to the 893 in the EYLF pilot, the consistency is clear: Tandi emerges not from instruction, but from conditions that invite imagination, sustain attention, and honor peer connection as worthy of time and space.

That space—measured in seconds, supported by wood and cloth and quiet—is where human meaning begins.

Rachel Kim

Rachel Kim

Board-certified OB-GYN and maternal-fetal medicine specialist. Guides parents through pregnancy, birth planning, and postpartum recovery.