How we know what we say

Jim Cummins, working since the 1970s on bilingual education, drew a distinction that most parenting conversations skip. Conversational fluency in a language — BICS, Basic Interpersonal Communicative Skills — usually develops in two to three years of exposure. Academic fluency in the same language — CALP, Cognitive Academic Language Proficiency — takes five to seven years, often longer. A Korean-American child can sound completely native in spoken Korean at age eight and still struggle to read a 한글 textbook at age twelve. That gap isn't a failure; it's the predictable shape of bilingual development. Cummins also proposed a threshold hypothesis: bilingual children whose two languages are both developed past a certain level show measurable cognitive advantages. Below that threshold, the advantages don't appear. The practical implication for our platform: the bilingual Spec section asks specifically about reading and writing in 한글 (CALP indicators), not just spoken use (BICS indicators), and the Initial Report's heritage-language reasoning bands its observations to that distinction. We don't conflate "speaks Korean" with "can grow up academically in Korean" — Cummins's work is the reason.

Read the source — Cummins, J. (1980). The Cross-Lingual Dimensions of Language Proficiency.

The CDC's "Learn the Signs. Act Early." program, in collaboration with the American Academy of Pediatrics, publishes age-banded developmental milestones across motor, language, cognitive, and social-emotional domains — from two months through five years, with extensions for older children. The framework is explicitly observational rather than diagnostic. Parents are guided to notice what a child is doing in real life, compared to what most children at the same age are doing, and to bring those observations to a pediatrician — never to convert the observations into a label themselves. The 2022 milestone revision raised the bar from "what 50% of children do by this age" to "what 75% of children do," precisely so families catch concerns earlier without false alarms. The platform's Initial Report follows this discipline exactly. Each developmental observation is time-bounded ("right now," "in the past few months") and tied to a concrete behavior the parent reported, not to a diagnostic category. When a pattern suggests a clinical conversation is worth having, the report names that consideration and points the family toward a qualified professional — pediatrician, speech-language pathologist, school psychologist — instead of overstepping. We describe; we don't diagnose. CDC's framing is why.

Read the source — CDC — Learn the Signs. Act Early.

Lev Vygotsky, a Soviet psychologist whose work was rediscovered in the West through the 1980s, proposed that the most productive learning happens in a narrow band: the space between what a child can do alone and what they can do with help from a more capable other. He called it the Zone of Proximal Development, or ZPD. Tasks easier than that range are boring and don't build new capacity; tasks harder than that range overwhelm. The implication, validated repeatedly in intelligent-tutoring-system research and in spaced-practice meta-analyses, is that two priorities at the right difficulty beat five priorities at any difficulty — almost every time. The platform's weekly roadmap caps at two or three priorities for exactly this reason. When a priority doesn't get completed, the roadmap reduces the friction for the next week instead of replacing it; the same priority can roll forward across multiple weeks. The point is to keep the child working in their actual ZPD, not to optimize a completion-rate metric that would punish a parent for picking a meaningful challenge. Vygotsky also stressed that the "more capable other" — a parent, sibling, mentor, or platform — is what lifts the ceiling. Our roadmap items are designed to be scaffold-able by the parent, not to outsource the relationship.

Read the source — Vygotsky, L. S. (1978). Mind in Society — The Development of Higher Psychological Processes.

The World Health Organization's International Classification of Functioning, Disability and Health (ICF) framework was adapted for children by Peter Rosenbaum and colleagues into a six-domain strengths language: Function (what the child can do), Family (the unit they live within), Fitness (physical capability and health), Fun (what brings them joy), Friends (their social connections), and Future (what they can imagine becoming). The framework was originally developed in the context of childhood disability — to push clinicians and families past deficit-focused language toward strength-anchored observation — but it generalizes. Generic praise ("you're so smart") doesn't shift behavior; it shifts the child toward avoiding situations where the praised label might fail. Carol Dweck's research on growth mindset documents this carefully. Anchored observation — "you noticed your brother needed help and you helped him without being asked, that's the Family F-word at work" — does the opposite: it names a concrete behavior in a domain a child can keep developing. The Initial Report's strengths section uses the six F-words categorically. Every domain section in the report names at least one strength tied to a behavior the parent reported, in plain language, with the F-word labeled. Children remember how they were talked about. The framework's job is to make sure how they're talked about helps them grow.

Read the source — Rosenbaum, P., & Gorter, J. W. (2012). The 'F-words' in childhood disability — I swear this is how we should think!

Edward Deci and Richard Ryan, working together since the 1970s, identified three psychological needs that turn out to predict whether any behavior — exercise, study, family ritual — sticks across time: autonomy (a sense of acting from your own choice rather than under pressure), competence (a sense of growing capability in the activity), and relatedness (a sense of connection to people who matter to you in the activity). When all three are present, motivation tends to be intrinsic and durable; the activity continues without external rewards. When one is missing, motivation tends to collapse the moment the external pressure lifts. Decades of follow-up research, including the OECD's PISA-linked motivation studies and intelligent-tutoring-system meta-analyses, have made this the most empirically supported motivation framework in education. The platform's weekly roadmap is built to honor all three needs. Autonomy: the family chooses which priorities to pin or dismiss; the roadmap proposes, never imposes. Competence: every priority is calibrated to the child's developmental band and rolling difficulty, so the child experiences growing capability rather than overwhelm or boredom (this is also where Vygotsky's ZPD shows up). Relatedness: every priority is designed to be done with the parent, not delegated to a screen. A roadmap that ignored these three needs would feel productive for a week and abandoned by week four. Deci and Ryan's work is the reason we don't build it that way.

Read the source — Deci, E. L., & Ryan, R. M. (2000). Self-Determination Theory and the Facilitation of Intrinsic Motivation, Social Development, and Well-Being.

Maria Polinsky and Silvina Montrul, working across the last twenty years, mapped what happens to bilingual children whose heritage language is the minority language in their environment. The pattern is consistent across language pairs (Russian-English, Spanish-English, Korean-English) and consistent across families. Until age four or five, the heritage language often looks like the dominant language. Then school starts. English becomes the language of peers, instruction, identity, and most waking hours. The heritage language doesn't disappear — but it stops developing along its full grammatical and academic trajectory. Polinsky describes this as "incomplete acquisition" — the speaker reaches an asymptote, often around the linguistic competence of a five-to-seven-year-old monolingual peer, and stays there. Montrul frames it as ongoing attrition driven by reduced input. The two framings disagree about mechanism but converge on what reverses the curve: consistent meaningful use at home (not just exposure — production, with stakes and feedback) and emotional positivity around the language. Children whose heritage language is associated with shame, correction, or family conflict drop the language faster than children for whom it's associated with warmth and competence. The platform's bilingual Spec captures both signals — who speaks Korean with the child, in which contexts, with what emotional tone — and the Initial Report's heritage-language reasoning treats those signals as the variables that move outcomes. Telling parents to "speak more Korean" isn't a useful intervention. Asking them to look at the emotional climate around the language is.

Read the source — Polinsky, M. (2018). Heritage Languages and Their Speakers. Cambridge University Press.

Ellen Bialystok at York University has spent four decades measuring what bilingual children do differently from monolingual peers — and importantly, where the differences hold up under scrutiny and where they don't. The strongest finding, replicated across hundreds of studies: bilingual children show advantages on tasks that require attention control and selective inhibition. They're better at ignoring misleading information, switching between rules, and holding two systems in mind simultaneously. The cognitive mechanism Bialystok proposes is straightforward — a bilingual brain manages two language systems that are always active, so the prefrontal machinery for choosing one and suppressing the other gets daily exercise. Where she's honest: the early literature overstated some claims. There's no general "bilinguals are smarter" effect; vocabulary breadth in each individual language is often smaller than in monolingual peers (because the same brain holds twice the linguistic territory). The advantages are specific — attention control, task-switching, and later in life, a meaningful 4-to-5-year delay in the symptomatic onset of dementia. The platform's bilingual Spec treats Korean-American children's bilingualism as a developmental asset along this specific axis. The Initial Report's executive-function reasoning notes that strong attention-control behaviors observed in bilingual children are consistent with Bialystok's findings — without overclaiming.

Read the source — Bialystok, E. (2017). The bilingual adaptation. Psychological Bulletin, 143(3), 233–262.