LIFE SCIENCES – Ran­dom­ness

Even pre­tend­ing to con­struct ran­dom se­quences is a young per­son’s game, AN­DREW MASTER­SON re­ports.

Cosmos - - Digest -

Code-break­ing, at its heart, in­volves look­ing at lots of num­bers and try­ing to spot tell-tale pat­terns that may pro­vide a key to de­cryp­tion.

Spies long ago re­alised that the best way to cre­ate truly ran­dom en­cryp­tion is to ask a com­puter to do it. This is be­cause hu­mans are rather poor at cre­at­ing lists of ran­dom num­bers, even when they are con­sciously try­ing to do so.

“Some early cryp­tog­ra­phers as­sumed they could gen­er­ate huge amounts of ran­dom keys by hap­haz­ardly tap­ping away at a type­writer,” wrote physi­cist Si­mon Singh in his his­tory of es­pi­onage, The Code Book (Fourth Es­tate, 1999).

“This might be a quick way of gen­er­at­ing a [cipher] key, but the re­sult­ing se­quence has struc­ture and is no longer ran­dom.”

Look­ing for an in­sight into why ran­dom­ness – or ap­par­ent ran­dom­ness – is so chal­leng­ing, a team of re­searchers from the Lab­o­ra­toire de Recherche Sci­en­tifique in Paris, France, re­cently de­vised a se­ries of tests and ap­plied them to a pool of 3,400 peo­ple, aged be­tween four and 91.

Each of the par­tic­i­pants was asked to com­plete a se­ries of tasks, all of which re­quired the con­struc­tion of lists that would ap­pear ran­dom when viewed by an ob­server.

The tasks in­cluded mak­ing up a set of 12 coin-toss re­sults, and in­vent­ing the num­bers shown by 10 hy­po­thet­i­cal die throws. The re­sults, pub­lished in the jour­nal PLOS Com­pu­ta­tional Bi­ol­ogy, re­veal that age 25 is the peak for achiev­ing truly ran­dom-look­ing lists.

“This ex­per­i­ment is a kind of re­verse Tur­ing test for ran­dom be­hav­iour, a test of strength be­tween al­go­rithms and hu­mans,” says co-au­thor Hec­tor Ze­nil.

The find­ings add weight to pre­vi­ous stud­ies that sug­gest – per­haps coun­ter­in­tu­itively – that con­struct­ing prop­erly ran­dom lists re­quires very high-level cog­ni­tive pro­cesses. Ran­dom­ness may be con­nected to other cog­ni­tive ac­tiv­ity, par­tic­u­larly cre­ativ­ity.

So if you need to jot some­thing truly ran­dom, what should you do? Singh’s ad­vice: “The best ran­dom keys are cre­ated by har­ness­ing nat­u­ral phys­i­cal pro­cesses, such as ra­dioac­tiv­ity, which is known to ex­hibit truly ran­dom be­hav­ior.”


Newspapers in English

Newspapers from Australia

© PressReader. All rights reserved.