Push-the-Sound Boxes — Elkonin Segmentation

A small grid of 2 or 3 connected squares drawn on paper. The adult says a word slowly; the child pushes one token (button, coin, block) into a box for each phoneme heard. Originated by Soviet psychologist Daniil Elkonin (1963), Elkonin boxes are one of the most rigorously evidence-based phonemic-awareness scaffolds — the National Reading Panel found PA instruction paired with manipulatives produced d ≈ 1.37 on PA outcomes (Ehri et al., 2001), roughly double the audio-only version.

  1. Draw 2 connected squares on paper (later 3, then 4). Place 4 tokens above the boxes (buttons, coins, pebbles, blocks).
  2. Adult says a 2-phoneme word stretched — /aaa-t/. Child pushes one token into each box left-to-right as they hear each sound.
  3. After pushing, child slides each token back as they re-say each sound. Bidirectional segmentation reinforces the mapping.
  4. Move to 3-phoneme CVC words once 2-phoneme is easy — cat, sun, mom, dog, hop, big.
  5. Use connected phonation first (/mmmaaattt/), then fully segmented (/m/ /a/ /t/).
  6. Pause at the end and count. “How many sounds did cat have?” — phoneme counting is a separate, slightly harder skill that consolidates the segmentation.

Variation: Sound-Letter Boxes — for children already learning letter sounds, write the letter on each token after pushing. This is the higher-leverage version (NRP found letter-linked PA d ≈ 1.37). Magnet Boxes — draw squares on a baking tray; use letter magnets. Sand-Tray Boxes — draw three squares in a tray of sand; child writes the sound with a finger after each push. Tile-Push — alphabet tiles instead of generic tokens.

Requirements

  • Space: Small table or floor area; tray helps contain tokens
  • Surface: Flat surface (table, floor, tray) where tokens can slide
  • Materials: Paper with 2–3 connected squares drawn (or printable Elkonin templates from Reading Rockets); 4 small tokens per child (buttons, coins, pebbles, dried beans, blocks); optional letter tiles or magnets for the advanced version; pencil
  • Participants: 1 adult + 1 child; siblings can each have their own grid
  • Supervision: Adult-led; choose the words and model the first 2–3

Rationale & Objective

The Elkonin method (Elkonin, 1963; Downing’s Comparative Reading, 1973) externalises the abstract act of phoneme segmentation as a physical push — a Vygotskian scaffold that lets the body do what the mind cannot yet hold. Cunningham’s (1990) RCT showed sound-box-style segmentation training caused reading gains; McCarthy’s (2008) systematic protocol in The Reading Teacher documented effective classroom delivery. Joseph’s (1998, 2000) “Word Boxes” studies with at-risk kindergartners produced gains in phonemic awareness, letter-sound correspondences, and spelling. The decisive finding comes from the NRP meta-analysis (Ehri et al., 2001): PA instruction paired with letter manipulatives yielded d ≈ 1.37 on PA outcomes — roughly double the audio-only version. Kilpatrick (2015) frames sound boxes as a bridge between phoneme awareness and phonics. Honest caveats — most rigorous evidence is for K–1 children at risk and for CVC words; the strategy assumes the child can already isolate initial sounds (so do Tongue Twister Theater and Robot Talk first). Pure auditory use (no letters on the tokens) is meaningfully less effective than the letter-linked version — once the child has stable phoneme awareness, transitioning to letter tiles is where the literacy multiplier kicks in.

Progress Indicators

  • Early: pushes tokens randomly without matching to sounds; pushes too many or too few; can echo a CVC word but not segment it
  • Developing: pushes one token per sound for 2-phoneme words with adult modelling; needs help with 3-phoneme words; uses connected phonation
  • Proficient: independently segments 3-phoneme CVC words; tolerates fully segmented phonation; counts phonemes back (“cat has 3 sounds”); slides tokens back as they re-say each sound
  • Advanced: segments 4-phoneme words with consonant blends (stop, sand, flag); pairs letters with tokens for known sounds; spontaneously segments to spell (“how do I write dog? /d/ /o/ /g/”)

Safety Notes

  • Use tokens too large to swallow — coins, large buttons, blocks; avoid dried beans, beads, or small magnets if a younger sibling is present
  • Magnetic letter tiles can be a hazard if a child mouths them; check magnets are securely embedded and the tile is age-rated
  • Don’t start with 4-phoneme words; failing here builds avoidance of the whole activity
  • For children with fine-motor difficulty, use larger tokens that are easier to grip and slide
  • Stop after 5–7 minutes; segmentation is cognitively demanding and tires attention fast
  • If a child consistently miscounts phonemes after 6–8 weeks of practice, mention to a paediatrician or SLP — persistent segmentation difficulty is a meaningful indicator for reading risk

Hints

  • Playfulness: themed tokens (pirate doubloons, dragon scales, dinosaur eggs); pushing “into the cave/pond/mouth” rather than “into the box”; let the child decorate the grid
  • Sustain interest: vary the tokens weekly; sand-tray version on warm days; advance to letter tiles once 3-phoneme is solid (the letter-linked version is where the research effect roughly doubles)
  • Common mistake: starting with 4-phoneme words too soon; using letter names (“B”) instead of sounds (/b/); rushing; saying the word too fast for the child to hear the sounds; pairing with letters before the child can segment reliably
  • Limited space: draw boxes on a placemat at dinner; use 3 coins on a napkin in a café; finger-poke on the leg in the car (no boxes needed)
  • Cross-domain: fine motor (precision pushing); literacy (most direct PA-to-spelling transfer); working memory (holding the word while segmenting); mathematics (counting phonemes); pre-writing (transitions naturally to invented spelling)
  • Progression: 2-phoneme audio-only → 3-phoneme audio-only → 3-phoneme audio + letter tiles → 4-phoneme blends → child segments to spell → child manipulates phonemes (delete/substitute) using boxes

Sources

  • Elkonin, D. B. (1963/1973). “The psychology of mastering the elements of reading.” In Downing, J. (Ed.), Comparative Reading. Macmillan. Earlier work in Simon, B. & Simon, J. (Eds.), Educational Psychology in the USSR, Routledge
  • Ehri, L. C., Nunes, S. R., Willows, D. M., Schuster, B. V., Yaghoub-Zadeh, Z. & Shanahan, T. (2001). “Phonemic awareness instruction helps children learn to read: Evidence from the National Reading Panel’s meta-analysis.” Reading Research Quarterly, 36(3), 250–287
  • Cunningham, A. E. (1990). “Explicit versus implicit instruction in phonemic awareness.” Journal of Experimental Child Psychology, 50(3), 429–444
  • McCarthy, P. A. (2008). “Using Sound Boxes Systematically to Develop Phonemic Awareness.” The Reading Teacher, 62(4), 346–349
  • Joseph, L. M. (1998). “Word boxes help children with learning disabilities identify and spell words.” The Reading Teacher, 52(4), 348–356
  • Joseph, L. M. (2000). “Developing first graders’ phonemic awareness, word identification, and spelling.” Reading Research and Instruction, 39(2), 160–169
  • Kilpatrick, D. A. (2015). Essentials of Assessing, Preventing, and Overcoming Reading Difficulties. Wiley
  • Common Core State Standards RF.K.2e — add or substitute individual sounds; segment phonemes in spoken words
  • Reading Rockets — Elkonin Boxes (free printable templates and demonstration videos)