Kahneman, Sibony, and Sunstein distinguish between two forms of error in human judgment: bias (systematic error in one direction) and noise (random variability in judgments that should be identical). While bias has received extensive attention in the behavioural science literature, noise has been largely neglected despite being at least as costly. The book documents the pervasive presence of noise in medical diagnoses, legal sentences, financial forecasts, and institutional decisions — and proposes a framework for reducing it through what the authors call 'decision hygiene'.
The relevance to the information environment is less direct than Kahneman's Thinking, Fast and Slow but significant. Noise in intelligence assessment, threat evaluation, and institutional decision-making creates vulnerabilities that adversaries can exploit — inconsistent responses to the same information signal create opportunities for manipulation and plausible deniability. More broadly, understanding the distinction between bias and noise is important for practitioners designing counter-disinformation interventions: the corrective strategies for systematic bias (where people consistently err in one direction) are different from the corrective strategies for noise (where people err randomly). A rigorous and important contribution to the science of human judgment.
Cogitavi commentary
Kahneman, Sibony, and Sunstein distinguish between two forms of error in human judgment: bias (systematic error in one direction) and noise (random variability in judgments that should be identical). While bias has received extensive attention in the behavioural science literature, noise has been largely neglected despite being at least as costly. The book documents the pervasive presence of noise in medical diagnoses, legal sentences, financial forecasts, and institutional decisions — and proposes a framework for reducing it through what the authors call 'decision hygiene'.
The relevance to the information environment is less direct than Kahneman's Thinking, Fast and Slow but significant. Noise in intelligence assessment, threat evaluation, and institutional decision-making creates vulnerabilities that adversaries can exploit — inconsistent responses to the same information signal create opportunities for manipulation and plausible deniability. More broadly, understanding the distinction between bias and noise is important for practitioners designing counter-disinformation interventions: the corrective strategies for systematic bias (where people consistently err in one direction) are different from the corrective strategies for noise (where people err randomly). A rigorous and important contribution to the science of human judgment.