Numerical routines individualize through a three-stage sequence of vertical and horizontal bonding¶

Claim · All claims
Evidence · 1 study · qualitative r2 · q3 · n=4 children (1 focal case followed in depth, 3 corroborating), ages 2(1)-4(6) across the sample
Currency · newest study 2019 · not yet checked against its sources · not reviewed by a person

A longitudinal teaching experiment tracing one child's numerical development over 18 months (corroborated by three additional cases) found a consistent three-stage sequence — unbonded precursor routines, then separately vertically-bonded quantitative and numerical routines, then horizontally-bonded coalescence into a discourse of cardinality — with the same qualitative sequence, though not the same pace, across all four children studied.

Subclaims¶

q3 i? A focal child's numerical development over 18 months (ages 2[8]-4[2]) fit a three-stage sequence: unbonded precursor routines (arbitrary choosing, disconnected naming), then vertically-bonded routines in which counting or estimation outcomes were connected to comparison choices, then a discourse of cardinality in which counting and quantitative estimation were recognized as equivalent (cross-discursive utterances like "this is four and this is a few" first appeared at this stage). → Lavie & Sfard 2019

q3 i? Three additional children, tracked with the same method and coding scheme, showed the same qualitative sequence of routine types and vocabulary development despite individual variation in pace, supporting the sequence as a general mechanism rather than an idiosyncrasy of one case. → Lavie & Sfard 2019

q3 i? The focal child could count accurately well before using counting to answer a quantity-comparison question — accurate counting alone did not confer quantitative reasoning; only after horizontal bonding (recognizing that counting and estimation answer the same question) did the child spontaneously turn to counting for comparisons, a behavior otherwise reported as rare before ages 5-6. → Lavie & Sfard 2019

Evidence¶

Lavie & Sfard 2019¶

Lavie, I., & Sfard, A. (2019). How Children Individualize Numerical Routines: Elements of a Discursive Theory in Making. Journal of the Learning Sciences, 28(4-5), 419-461. https://doi.org/10.1080/10508406.2019.1646650

q3 · intensive longitudinal teaching experiment (1 focal case, 3 corroborating cases), video-recorded interviews approximately every 4-5 months over 18-24 months, with an explicit coding scheme for routines and vocabulary; not a controlled experiment or RCT, and the authors themselves note their task battery had gaps (e.g., too few small-set comparisons to fully test whether cardinality coalescence happens for small sets before large ones) · i? · the abstract prints no effect size; the full text may · n=4 children (1 focal case followed in depth, 3 corroborating), ages 2(1)-4(6) across the sample · qualitative · r2

Under Commognition, the researcher acted as both teacher and observer in a Vygotskian teaching-experiment design, posing repeated "Where Is There More?" comparison tasks to four children over 18-24 months and video-recording their routines and vocabulary. Vertical and Horizontal Bonding in Numerical Routines names the two mechanisms the authors identify as driving the transition between stages. The focal case (Milo) is documented in the greatest depth; the three additional cases replicate the qualitative sequence (though not its timing), which is the paper's basis for treating the mechanism, not just the timeline, as generalizable.

Discussion¶

This is a theory-building empirical case study, not an intervention trial: there is no comparison condition, and the sample is small (n=4) even though the depth of documentation per case is high. The three-stage model and the bonding mechanism are well-corroborated within this methodological frame — consistent video evidence, an explicit coding scheme, cross-case replication of sequence — but would benefit from larger-scale testing, and the authors are candid that their task battery could not fully test every constituent claim (e.g., the small-set vs. large-set coalescence order). The instructional implication (embed counting in a purposeful "deed," not a decontextualized exercise) is drawn from a single case's accelerated trajectory and should be treated as a plausible hypothesis rather than an established design recommendation.

Learner Variables¶

• Number Sense — outcome: instruction changes it

Related Claims¶

• (none yet linked)
• Number Board Games Improve Numerical Knowledge — related
• Early numeracy mastery is measurable at 54 months across three competencies in a large longitudinal sample — related
• Preschoolers' ANS task performance follows Weber's Law: accuracy decreases as numerical ratio approaches equality, and improves with age over six months — related
• Children as young as 3½ years perform significantly above chance identifying and comparing multidigit numerals — related