The Cost of Learning Judgment: Retrieval–Judgment Structure and Adolescent Learning Motivation in AI-Age China

Generative artificial intelligence has driven the cost of obtaining information close to zero without lowering the cost of judging it to a comparable degree. This article separates learning cost into retrieval cost (R) and judgment cost (J) and asks why, as learning becomes cheaper, the motivation of a subset of Chinese adolescents falls. AI compresses the front end of R while the unit cost of the steps after R barely changes; the fall in J is derivative—it comes from having fewer candidates to judge, not from AI taking over judgment—and the marginal cost of J rises again once AI returns a large volume of plausible but doubtful candidates. Judgment is not internally flat: the scale of the knowledge structure that must be mobilised subsumes the distinction between lower- and higher-order judgment, so the larger the mobilisation, the dearer and the less substitutable the step. On the exchange side, motivation depends on expected return (E), the credibility of delivery (T), and perceived fairness (F), minus the two costs; T is the credibility of an intertemporal commitment and cannot be folded into E, which is a probability of obtaining. China’s high-stakes standardized assessment regime has long priced only answers that are retrievable, reproducible, and scorable, while the credential–establishment–hukou–housing–marriage composite return has depreciated amid credential inflation, contracting establishment positions, and weak signal governance; with a single evaluation scale and a mismatch in timing, withdrawal is confirmed by behaviour and closes a loop into learned helplessness. The article sets out an R–J framework and seven testable propositions; the exit threshold and the measurement of mobilisation are left as interfaces.

Authors

Publication Details

Journal
EdArXiv (OSF Preprints)
Published
2026-11-17
DOI
https://doi.org/10.35542/osf.io/cuygs
Primary Topic
Education, Achievement, and Giftedness
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

The Cost of Learning Judgment: Retrieval–Judgment Structure and Adolescent Learning Motivation in AI-Age China

Yilin Li
3 links
EdArXiv (OSF Preprints)
Education, Achievement, and Giftedness
preprint

The Cost of Learning Judgment: Retrieval–Judgment Structure and Adolescent Learning Motivation in AI-Age China

Yilin Li
preprint en

Abstract

Generative artificial intelligence has driven the cost of obtaining information close to zero without lowering the cost of judging it to a comparable degree. This article separates learning cost into retrieval cost (R) and judgment cost (J) and asks why, as learning becomes cheaper, the motivation of a subset of Chinese adolescents falls. AI compresses the front end of R while the unit cost of the steps after R barely changes; the fall in J is derivative—it comes from having fewer candidates to judge, not from AI taking over judgment—and the marginal cost of J rises again once AI returns a large volume of plausible but doubtful candidates. Judgment is not internally flat: the scale of the knowledge structure that must be mobilised subsumes the distinction between lower- and higher-order judgment, so the larger the mobilisation, the dearer and the less substitutable the step. On the exchange side, motivation depends on expected return (E), the credibility of delivery (T), and perceived fairness (F), minus the two costs; T is the credibility of an intertemporal commitment and cannot be folded into E, which is a probability of obtaining. China’s high-stakes standardized assessment regime has long priced only answers that are retrievable, reproducible, and scorable, while the credential–establishment–hukou–housing–marriage composite return has depreciated amid credential inflation, contracting establishment positions, and weak signal governance; with a single evaluation scale and a mismatch in timing, withdrawal is confirmed by behaviour and closes a loop into learned helplessness. The article sets out an R–J framework and seven testable propositions; the exit threshold and the measurement of mobilisation are left as interfaces.

EdArXiv (OSF Preprints)
Education, Achievement, and Giftedness
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon