Published November 15, 2026 | Version v1

Weight Rose, Perception Did Not: A Test of Hidden Capture, the Signature Prediction of the Correction Freeze Problem

Authors/Creators

  • 1. Phenikaa University

Description

The Correction Freeze Problem (CFP) ends in a chain: drift, freeze, divergence, feedback, hysteresis and hidden capture. Its final link, a self-reinforcing shift that completes while perception reports no change, is the one prediction in the chain that models of tipping and critical transitions do not make. This report tests it in U.S. body weight, a state that rose over two decades and is socially reinforced, and in which perception is measured separately from the state itself. Using the National Health and Nutrition Examination Survey (NHANES), measured overweight among adults aged 20 to 74 rose from 63.8% in 1999 to 2000 to 73.6% in 2017 to March 2020. Over the same period, the share who considered themselves overweight moved from 56.5% to 57.0%. Holding the 1999 to 2000 perception standard fixed, 64.4% would have been expected. The freeze fraction, the share of the expected rise in perceived overweight that never appeared, was 0.93 (95% CI 0.68 to 1.30), which meets the rule for hidden capture fixed before the run. At the same measured weight, people became steadily less likely to see themselves as overweight, while their reports of their own weight stayed accurate to within about one kilogram. A first run with a shorter baseline, 2011 to 2012, pointed the same way but was not decisive. The result is the first support in this series for the unique link of CFP's signature.

Files

CFP_HiddenCapture_NHANES.pdf

Files (433.7 kB)

Name Size
md5:95b40ead994200888b9b86e581413ee0
433.7 kB Preview Download