Fear without a dose-response: real-time crowdsourced psychometrics show a moderate earthquake’s psychological footprint is decoupled from shaking intensity in an anticipation-primed megacity

Yasemin Didem Aktas , Remy Bossu, Simon Issartel, Robert Steed

Fear shapes how populations respond to earthquakes, yet it is almost always measured retrospectively, once it cannot be separated from the damage. On 23 April 2025 a MW 6.2 earthquake shook Istanbul – a megacity that has anticipated a major rupture for 25 years – at ground motions below design level, causing one death and light damage. Using a validated earthquake-fear instrument delivered in real time on a crowdsourced platform, we obtained 186 responses within hours, each linked to the respondent’s reported intensity and location. Within the sample, fear barely tracks shaking: its correlation with reported intensity is weak (r = 0.24; 6% of variance), and respondents on Istanbul’s European and Asian sides report indistinguishable fear (24.2 versus 24.5; p = 0.85) despite the European side shaking harder. Fear responds to being substantially shaken at all – a threshold – not to how hard the ground shook – a dose. The footprint was a chronic condition made vivid, not an acute reaction: it produced a behavioural response – 100,000 people outdoors, injuries dominated by panic – out of proportion to a one-fatality event. Real-time psychometric measurement through crowdsourcing is feasible at scale and uniquely separates fear from its physical cause.