In times of cri­sis and pan­demics, ef­fi­cient com­mu­ni­ca­tion must suc­ceed in bring­ing about suf­fi­cient changes in people’s be­hav­ior. An im­por­tant fac­tor here is the fear of in­fec­tion: it can mo­ti­vate pro­tec­tive re­ac­tions as well as trig­ger pan­ic be­hav­ior. We there­fore ex­am­ined the de­ter­mi­nants pre­dict­ing fear of the virus and sub­jec­tive health sta­tus dur­ing the lock­down. The re­sult: In­di­vid­ual psy­cho­log­i­cal vari­ables have much bet­ter pre­dic­tive pow­er than en­vi­ron­men­tal conditions.

To ef­fec­tive­ly man­age a pan­dem­ic, it is im­por­tant to know what fac­tors cause fear but al­so mod­u­late sub­jec­tive health and well-be­ing at the same time. This will al­low us to pre­dict how dif­fer­ent peo­ple and pop­u­la­tions will re­spond to ex­ter­nal threats and con­straints. To this end, we stud­ied 533 Eu­ro­pean subjects.

We used ma­chine learn­ing to iden­ti­fy which vari­ables pre­dict­ed fear and sub­jec­tive health sta­tus dur­ing the lock­down. And our re­sults show that psy­cho­log­i­cal vari­ables such as at­tach­ment style, per­ceived sus­cep­ti­bil­i­ty to in­fec­tion in gen­er­al, and dis­gust with germs had much bet­ter pre­dic­tive pow­er than en­vi­ron­men­tal con­di­tions such as lo­cal spread of the virus, mor­tal­i­ty, and type of lock­down. In turn, we can in­fer how peo­ple per­ceive the threats posed by a pan­dem­ic and iden­ti­fy tar­get groups that are par­tic­u­lar­ly af­fect­ed by the neg­a­tive im­pact of pro­tec­tive measures.

You’ll find the spe­cif­ic re­sults of our in­ves­ti­ga­tion here shortly.