On getting it right by being wrong: A case study of how flawed research may become self-fulfilling at last

Abstract

Significance Understanding how humans process time series data is more pressing now than ever amid a progressing pandemic. Current research draws on some fifty years of empirical evidence on laypeople’s (in-)ability to extrapolate exponential growth. Yet even canonized evidence ought not to be trusted blindly. As a case in point, I review a seminal study that is still highly (even increasingly) cited, although seriously flawed. This case serves as both a reminder of how readily even experienced, well-meaning researchers underestimate exponential dynamics, and an admonition for subsequent researchers to critically read and evaluate the research they cite in order to catch and correct errors quickly rather than carry them forward over decades.