Even be­fore you know you’re sick, it shows

Re­search vol­un­teers pick faces of ill peo­ple out of photo lineup

The Dallas Morning News - - Focus On Science - Au­drey Hen­der­son/The Royal So­ci­ety Karen Ka­plan, Los An­ge­les Times

Peo­ple can judge whether some­one is sick by look­ing at a photo for just a few sec­onds.

That may not sound re­mark­able — un­til you con­sider that the sick peo­ple in the pho­tos were in the very early stages of ill­nesses. They were par­tic­i­pants in a sci­en­tific ex­per­i­ment and had agreed to be in­fected with a bac­terium that would cause an in­flam­ma­tory re­sponse. Their por­traits were taken just two hours af­ter in­fec­tion.

Those pic­tures, along with por­traits of the same peo­ple taken when they were healthy, were flashed in front of dozens of study vol­un­teers in ran­dom or­der. The vol­un­teers had no more than 5 sec­onds to guess whether the per­son in the pic­ture was sick or healthy.

In the first se­ries of rat­ings, 62 vol­un­teers made 2,945 judg­ments about 32 pic­tures (of eight men and eight women, each pho­tographed af­ter be­ing in­jected with an E. coli en­do­toxin and af­ter be­ing in­jected with a placebo).

The vol­un­teers guessed that 1,215 of those pic­tures were of peo­ple who were sick. Those guesses were cor­rect 64 per­cent of the time and in­cor­rect 36 per­cent of the time, ac­cord­ing to a study pub­lished this week in the Pro­ceed­ings of the Royal So­ci­ety B.

Ap­par­ently, three of the 16 par­tic­i­pants were par­tic­u­larly in­scrutable; when they were re­moved from the anal­y­sis, the vol­un­teers guessed right 81 per­cent of the time, the study au­thors found.

On a scale on which 0.5 means guess­ing at ran­dom and 1 means guess­ing right ev­ery time, the vol­un­teers scored a 0.62. In other words, they did sig­nif­i­cantly bet­ter than if they had just flipped a coin to de­cide whether some­one was healthy or sick.

In the sec­ond se­ries of rat­ings, 60 peo­ple looked at the same 32 pho­tos and were asked whether the per­son in each pic­ture looked sick and if they looked tired. The vol­un­teers also rated eight spe­cific “sick­ness cues” on a scale from 1 (no symp­toms) to 7 (very high symp­toms).

The rat­ings re­vealed that the peo­ple in the pho­tos looked more sick and more tired af­ter they got the E. coli in­jec­tion than af­ter they got the placebo. The raters also thought peo­ple had paler

“It would ar­guably be par­tic­u­larly ben­e­fi­cial to iden­tify sick in­di­vid­u­als at an early stage of sick­ness when risk for con­ta­gion is high.”

Re­search team led by John Ax­els­son of Stock­holm Univer­sity

skin, paler lips, more swollen faces, red­der eyes, more hang­ing eye­lids and droop­ier mouths af­ter they got the real in­jec­tion than af­ter they got the fake one.

Fur­ther anal­y­sis showed that all six of those fa­cial cues in­flu­enced raters’ per­cep­tion of a per­son’s health. So did tired­ness. But two other cues that got dif­fer­ent rat­ings in sick peo­ple and healthy peo­ple — glossy skin and patchy skin — did not af­fect judg­ments about health, the re­searchers


It may seem sur­pris­ing that reg­u­lar peo­ple could rec­og­nize sick­ness in oth­ers so soon af­ter in­fec­tion, but the study au­thors said this ca­pa­bil­ity has a clear evo­lu­tion­ary pur­pose.

“It would ar­guably be par­tic­u­larly ben­e­fi­cial to iden­tify sick in­di­vid­u­als at an early stage of sick­ness when risk for con­ta­gion is high,” wrote the team led by John Ax­els­son of Stock­holm Univer­sity and the Karolin­ska In­sti­tute

in Stock­holm.

In fact, they noted, the ex­per­i­ments with pho­tos prob­a­bly un­der­state the ex­tent of this skill: “In real-life cir­cum­stances, we would ex­pect hu­mans to have a higher sen­si­tiv­ity due to the pos­si­bil­ity of in­te­grat­ing other cues (e.g. gait, body odor and speech).”

But this judg­ment isn’t per­fect — es­pe­cially since many of the cues as­so­ci­ated with ill­ness are also present in peo­ple who are merely tired or sad. It seems in­evitable

that hu­mans some­times avoid oth­ers “who pose no threat of con­ta­gion,” the re­searchers wrote, adding that this might help ex­plain a phe­nom­e­non known as “dis­abil­ity stigma” and other kinds of prej­u­dices.

“Such be­hav­ioral ten­den­cies would have been fa­vored by se­lec­tion pres­sures to avoid false-neg­a­tive re­sponses when scan­ning the en­vi­ron­ment for im­mi­nent in­fec­tious threats,” they con­cluded.

Newspapers in English

Newspapers from USA

© PressReader. All rights reserved.