FOUNDATION”, a novel by Isaac Asimov from the golden age of sci­ence fic­tion, imag­ines a sci­ence called psy­chohis­tory which enables its prac­ti­tioners to pre­dict pre­cisely the behav­iour of large groups of people. The inventor of psy­chohis­tory, Hari Seldon, uses his dis­covery to save humanity from an his­tor­ical dark age.

A fan­tasy, of course. But the rise of mobile phones and social net­works means bud­ding psy­chohis­to­rians do now have an enor­mous amount of data that they can search for infor­ma­tion which might yield more modest pat­terns of pre­dictability. And, as sev­eral of them told the AAAS meeting, they are doing just that.

Song Chaoming, for instance, is a researcher at North­eastern Uni­ver­sity in Boston. He is a physi­cist, but he moon­lights as a social sci­en­tist. With that hat on he has devised an algo­rithm which can look at someone’s mobile-​​phone records and pre­dict with an average of 93% accu­racy where that person is at any moment of any day. Given most people’s reg­ular habits (sleep, com­mute, work, com­mute, sleep), this might not seem too hard. What is impres­sive is that his accu­racy was never lower than 80% for any of the 50,000 people he looked at.


Alessandro Vespig­nani, one of Dr Song’s col­leagues at North­eastern, dis­cussed what might be done with such knowl­edge. Dr Vespig­nani, another moon­lighting physi­cist, studies epi­demi­ology. He and his team have cre­ated a pro­gram called GLEAM (Global Epi­demic and Mobility Model) that divides the world into hun­dreds of thou­sands of squares. It models travel pat­terns between these squares (busy roads, flight paths and so on) using equa­tions based on data as var­ious as inter­na­tional air links and school holidays.

Read the article at The Economist →