1. A Tale of Two Language Systems

These two systems are not assumed to be the same. The child’s first system is a living outcome of language participation. The second is a culturally useful analytical system imposed later for describing and teaching alphabetic literacy.

2. Purpose of OPP

The primary purpose of OPP is diagnostic and epistemic: to reveal the extent to which literacy-conditioned adult categories can obscure the auditory organization that children actually bring to literacy. What instructional systems might eventually do differently is secondary.

OPP is not primarily a theory that “high-frequency sounds make spelling easier.” Its purpose is to expose the mismatch between the adult observer’s literacy-conditioned decomposition of speech and the child’s already-learned auditory organization.

3. What an “Organic Phoneme” Is—and Is Not

It is not a claim that the child internally represents a discrete phoneme. All phonemes are observer abstractions. The child may instead possess a recurrence-shaped field of differential sensitivity, familiarity, expectancy, resonance, or readiness.

4. Recurrence, Hebbian Learning, and Affective Amplitude

The OPP learning claim is best expressed as:

The nervous system is not presumed to count discrete phonemes. It experiences a continuous speech stream in which similar auditory events recur with different degrees of similarity, in different words and positions, across speakers, rates, pitches, coarticulations, and contexts.

Two variables must be separated

For a first broad test, affective amplitude can be provisionally set aside. The working hypothesis is that recurrence distribution is the primary experience-dependent source of differential auditory learning density, while acknowledging that acoustic salience, maturation, articulation, and other variables also contribute.

5. How Invented Spelling Supports OPP

Invented-spelling research provides behavioral evidence that children’s early speech-to-letter responses do not always follow adult phonemic divisions.

• Children may write D for the flap they hear in words such as city or lady.
• Some children treat the beginning of truck as more like CH than the adult decomposition /t/ + /r/.
• Children sometimes omit nasals or liquids in final clusters in ways compatible with distributed vowel coloring rather than a clean sequence of separate phonemes.
• Letter-name-shaped portions of words can attract letter choices that conventional adult scoring treats as equivalent to other choices.

6. Treiman & Wolter: 1,800 Literal Spellings

The supplied public archive contained 1,800 literal spelling trials from 75 preschool children, participant data, and the authors’ original analysis materials.

The reanalysis reproduced the authors’ main letter-name coefficient almost exactly: approximately b = 0.514, corresponding to an odds ratio of about 1.67.

C versus K

For /k/ items, the adult scoring system can treat C, K, and Q as equivalent plausible representations. But children redistributed among those letters depending on the larger heard pattern.

The attraction of K in cake, case, came, cave is especially informative because the conventional spellings use C, while the spoken letter name K = “kay” overlaps a much larger heard beginning.

G versus J

Adult phoneme scoring may treat both G and J as plausible for the same onset category, but the child’s literal choices distinguish G = “gee” as a better fit to the larger beginning of jeep.

7. Information Lost by Adult Phoneme Scoring

The strongest quantification came from restricting analysis to 93 child × item-pair cases in which both responses were already scored as phonemically successful by the adult system.

Within the same child and matched pair, 25 shifted toward the larger-pattern handle and only 8 shifted away. Exact paired p = .0046.

The literal C/K/Q and G/J choices retained about 0.0686 bits of information about whether the child heard the letter-name-shaped or control pattern.

The adult prop_first score retained exactly 0 bits in this subset because all of the differing literal responses had been normalized to the same value: “first phoneme plausibly represented.”

8. Whole-Letter-Name Capture: Evidence and Correction

Literal spellings such as BF for beef, BP for beep, TM for team, and GP for jeep suggest that some children may occasionally use one written letter as a handle for a larger auditory span corresponding to the letter’s spoken name.

Adult decomposition of BF = beef:

Possible child-functional organization:

9. Brazilian “Prephonological” Spellers

The Brazilian study reported 313 children and 766 spelling tests overall, with 129 children and 187 spelling tests classified as “prephonological.”

The classification procedure aligned observer-defined target phonemes with child phonograms and penalized implausible mappings, omitted phonemes, and extraneous letters. Children entered the prephonological group when both whole-word and first-phoneme analyses failed to outperform shuffled chance at the authors’ criterion.

The paper nevertheless demonstrates rich statistical learning: older prephonological spellers produced longer spellings, used letters and digrams more like Portuguese print, and often drew letters from their own names.

10. Broad Developmental Evidence Scaffold

The main OPP argument should not depend on tiny invented-spelling subgroups. It should be cumulative and developmental:

The scientifically safer formulation is not that newborns can literally discriminate and produce every sound in every language, but that infants begin with broad sensitivity to many speech contrasts and that this sensitivity is reorganized by language experience during the first year.

11. Classroom Study: Core Research Goal

The study should be designed as a large, simple, classroom-replicable “Two Language Systems” study, not as dozens of tiny phoneme experiments.

Primary research question

When a child hears a word, does the child’s response follow the adult-defined phonemic partition, or does it reveal larger or differently organized auditory affinities that were learned through natural oral-language participation?

A second question then asks whether those affinities are related to independently estimated natural recurrence density.

A third asks how those pre-existing auditory affinities begin to recruit alphabetic handles as letter names and letter sounds become known.

12. What Would Count as a Confirmable OP?

An “organic phoneme” remains an observer’s label. A recurring auditory pattern would count as a strong OP candidate when three conditions converge:

1. Natural recurrence: the auditory pattern occurs substantially in child-directed oral language across many words and contexts, not merely because it is taught as a letter or phonics unit.
2. Auditory sensitivity without print: children show differential sensitivity to the pattern when no letters, letter names, “first sounds,” or phoneme terminology are present.
3. Recurrence relation: stronger natural recurrence density predicts stronger auditory affinity across patterns, items, or children.

A fourth finding is highly relevant to literacy but not required to establish the organic auditory learning:

4. Alphabetic recruitment: once an appropriate letter or letter name is known, it is preferentially recruited where it can couple to that pre-existing auditory territory.

13. Auditory-First Elicitation Principle

Instead, with no print visible:

“Listen: cake. Which sounds more like cake: came or cook?”

All three words share what an adult calls /k/, but cake and came share the much larger “kay” region.

Likewise:

“Listen: jeep. Which sounds more like jeep: genie or jam?”

And for an internal or cross-orthography example:

“Listen: nature. Which sounds more like nature: paycheck or neighbor?”

The aim is to measure auditory affinity before letters contaminate the response.

14. A–Z Letter-Name OP Elicitation Table

These are candidate OPs, not presumed child-internal phonemes. They are useful because spoken letter names provide independently named auditory patterns that can be searched for inside ordinary language.

15. W as a Negative Control

“Double-you” may have relatively little natural preliteracy recurrence independent of alphabet naming. If corpus analysis confirms that, W becomes scientifically useful.

This gives the study a built-in falsification opportunity rather than making every letter-name pattern automatically count as an OP.

16. Four Ways to Evoke the Same Learning

A. Auditory affinity — primary measure

No print. The child hears one target and two alternatives:

“Cake. Which sounds more like cake: came or cook?”

The child points or answers orally. This is the cleanest measure of the inside-out system.

B. Mispronunciation detection

A puppet or recording says a familiar word naturally or with the candidate auditory region altered. The child is asked only:

“Did the puppet say it right?”

This measures auditory discrimination without requiring explicit segmentation.

C. Natural repetition

Ask the child to repeat the word naturally—not slowly and not “sound it out.” Record the production. This samples the child’s production of the same learned auditory territory without requiring a phonemic explanation.

D. Invented spelling — only after auditory testing

Then ask:

“Write jeep any way that makes sense to you.”

No correction and no sounding-out prompts. Preserve the literal response.

Potentially diagnostic responses include:

• G for jeep because “gee” spans the heard beginning.
• K for cake because “kay” spans the heard beginning.
• Q for cute because “cue” spans the beginning.
• H in a word such as nature if “aitch” is auditorily recruited despite no conventional H there.
• Z at the end of easy if “zee” is recruited.

17. Test Letter Knowledge Last

Only after the auditory session determine:

• Which letter names the child knows.
• Which letter sounds have been explicitly taught.
• Which letters the child can recognize.
• Which letters the child can write.
• What phonics correspondences have already been introduced in instruction.

18. Non-Letter-Name OPs: Bottom-Up Discovery

After testing letter-name candidates, do not simply add a phonics list such as sh, ch, th, oo, ow, ng.

Instead:

If this process rediscovers categories linguists call “sh,” “ng,” “er,” etc., that is acceptable. Their inclusion then comes from the child’s probable oral-language environment, not from a phonics chart.

19. The “Uh” Region as an Example

The “uh” region is a useful candidate because adult linguistic systems may divide similar heard events into schwa, STRUT, reduced vowels, and other categories.

Candidate naturally occurring examples previously discussed include:

• about
• zebra
• what
• glove
• quiet
• orange

The OPP question is not “Which of these contains phoneme /ə/?” It is:

Do children themselves behave as though some of these recurring “uh”-like events occupy a common auditory neighborhood, and does that relationship track how densely those events populate their natural oral-language environment?

Disagreement with a conventional phonemic inventory becomes data rather than error.

20. Generalization Requirements for Every OP Candidate

To prevent a candidate from being “confirmed” by one clever word pair, require evidence across:

• At least 4 different word families.
• At least 2 positions within words where possible.
• At least 2 speakers.
• Different surrounding sounds.
• Different conventional spellings where available.
• Familiar child vocabulary.
• Naturally spoken—not slowed or artificially segmented—words.

For “bee,” for example, use a family such as baby, beach, beetle, ruby rather than relying only on beach and bead.

For “ar,” use a family such as car, star, garden, party, subject to local pronunciation.

21. Dialect and the Child’s Own Pronunciation

Before a target trial, teachers can show a picture and ask:

“What’s this?”

Record the child’s natural production. If the child’s dialect differs from the standard recording or dictionary transcription, that difference must be treated as data.

22. Study Scale, Sampling, and Statistical Power

The user explicitly rejected over-tightening the study into many tiny low-powered tests. The design should therefore use individual OPs as repeated instances of a broader hypothesis.

Recommended scale

• 20–30 classrooms.
• Preschool, kindergarten, and first grade.
• Target 400–600 children.
• Approximately 50–60 OP candidates.
• Each child receives only a balanced subset.

Example:

With about 50 candidates, balanced assignment yields roughly 288 observations per candidate without forcing any one child through an exhausting battery.

Primary hypothesis should be pooled

Do not make the study’s success depend on:

• “Is ‘gee’ significant?”
• “Is ‘ess’ significant?”
• “Is ‘ar’ significant?”

Instead test:

Across naturally recurrent auditory patterns, do children systematically recognize relationships in the speech stream that are not adequately represented by the adult phonemic partition used in literacy instruction?

Then ask whether recognition strength rises with independent natural recurrence density, and whether alphabet knowledge later recruits those regions.

23. Preschool–Kindergarten–First Grade Comparison

The study should examine three developmental groups simultaneously and, ideally, repeat testing in fall, winter, and spring.

A powerful within-child result would be:

The child still judges cake and came as beginning more alike than cake and cook, while increasingly learning to say that all three “start with /k/” and to spell them according to taught conventions.

That would show both systems becoming co-present rather than assuming literacy simply replaces the earlier auditory organization.

24. Primary Outcome Measures

25. Primary Predictions and Falsification Tests

Primary OPP prediction

Before and during literacy learning, children will show systematic auditory distinctions inside categories that the adult phonemic system treats as equivalent, and the strength of those distinctions will be related to the recurrence of corresponding patterns in natural oral language.

Alphabetic recruitment prediction

Once an alphabetic handle becomes available, its probability of being recruited should be partly related to the learning density of the auditory territory with which it can couple.

Developmental prediction

With literacy instruction, children’s explicit answers and spellings should increasingly conform to the adult phonemic/orthographic system while some earlier auditory affinities remain detectable.

Falsification opportunities

• If high-recurrence auditory patterns do not produce stronger auditory affinity than matched low-recurrence patterns, the recurrence-density part of OPP is weakened.
• If children’s auditory choices are fully predicted by adult phoneme identity and no systematic within-category distinctions remain, the “chasm” hypothesis is weakened.
• If alphabet knowledge fully explains all apparent preliteracy effects, the claim of an independently learned oral-auditory substrate is weakened.
• If a low-natural-recurrence pattern such as “double-you” behaves identically to highly recurrent natural patterns before letter learning, the recurrence model is weakened.

26. Teacher-Facing Administration Protocol

Session structure

1. Picture naming / pronunciation check. Confirm the child’s natural label and dialectal pronunciation for target words.
2. Auditory affinity trials. No print, no letters, no “first sound” terminology.
3. Mispronunciation detection trials. Child judges whether a puppet/recording said a familiar word naturally.
4. Natural repetition. Child repeats selected items at normal conversational pace.
5. Invented spelling. Child writes selected words “any way that makes sense.” No correction or sounding-out coaching.
6. Letter knowledge last. Measure known letter names, known taught sounds, letter recognition, writing ability, and instructional exposure.

Teacher script principles

• Use short, neutral prompts.
• Never praise a specific auditory choice as “correct.”
• Never explain the intended sound relation during testing.
• Do not slow, stretch, or segment target words.
• Randomize which comparison appears first.
• Use headphones or consistent speaker playback where possible.
• Preserve literal spellings exactly as produced.

Recommended burden

Keep any single child’s battery to roughly 20–30 auditory comparisons plus a much smaller invented-spelling subset. Use balanced forms across the classroom so the research network accumulates many observations per candidate without exhausting children.

27. Analysis Plan

Level 1: Broad auditory effect

Pool all auditory trials and model the probability of choosing the larger recurrent auditory match versus the adult-phoneme-matched foil.

Level 2: Recurrence density

Add independently computed natural oral recurrence density as a predictor. The key question is whether higher recurrence predicts stronger child auditory preference across candidates.

Level 3: Development

Add age/grade and time point to test whether the auditory system remains detectable while explicit literacy-conditioned behavior increases.

Level 4: Alphabetic recruitment

For children who know the relevant letter name or sound, test whether auditory affinity predicts invented-spelling recruitment of the corresponding alphabetic handle.

Level 5: Information-loss analysis

For each literal spelling, compare:

• the full literal response,
• adult phoneme-equivalent scoring, and
• how much target/condition information is lost when the literal response is compressed into adult categories.

Recommended statistical structure

Use hierarchical or mixed-effects models with random effects for child, item, classroom, and where appropriate speaker. The primary inferential target should be the pooled OPP effect across many candidate auditory regions, with individual OPs treated as replications or moderators rather than independent make-or-break studies.

28. Current Evidentiary Position

29. Working Files from the Reanalysis

• OPP_Treiman_Wolter_item_recurrence.csv
• OPP_Treiman_Wolter_merged_trials.csv
• OPP_Treiman_Wolter_reanalysis.py
• OPP_TW_literal_item_summary.csv
• OPP_TW_flagged_literal_trials.csv
• OPP_TW_participant_literal_contrasts.csv
• OPP_TW_information_loss_summary.csv

These are exploratory working artifacts. The research design above is the prospective study architecture intended to test the broader OPP hypothesis more directly and with much greater statistical power.