"Long before it's in the papers"
August 03, 2010

RETURN TO THE WORLD SCIENCE HOME PAGE

For ants, one playbook fits many situations

Oct. 9, 2006
Special to World Science  

Ants of the spe­cies Tem­notho­rax cur­vi­spi­no­sus fol­low just one bas­ic set of rules for choos­ing a new home for their group.

But by mak­ing a few sim­ple changes, they can ad­just that play­book to fit al­most any sit­u­a­tion rang­ing from cas­u­al real-es­t­ate brows­ing to to­tal ur­gen­cy, a study has found.

Sci­en­tists are in­ter­est­ed in the “al­go­rithms,” or step-by-step rules, by which or­gan­isms make de­ci­sions, ei­ther in­di­vid­u­al or col­lec­tive­ly.

Life’s adapt­a­bil­ity de­pends in part on the fact that such al­go­rithms are easily re­peat­a­ble, yet ad­justable to dif­fer­ent sit­u­a­tions, wrote the researchers who conducted the study. As­pects of the hu­man brain al­so show such prop­er­ties, added the sci­en­t­ists, Ste­phen Pratt of Prince­ton Uni­ver­si­ty in Prince­ton, N.J. and Da­vid Sump­ter of the Uni­ver­si­ty of Ox­ford, U.K. 

Many ants use an al­go­rithm like that of T. cur­vi­spi­no­sus to search for a new home, ac­cord­ing to Pratt and Sump­ter.

Of an ant col­o­ny’s sev­er­al hun­dred mem­bers, a few con­stant­ly scout for new re­si­den­ces. The re­search­ers in­ves­ti­gat­ed how col­o­nies choose a new home when their old nest is de­stroyed, ver­sus when it’s mere­ly in­fe­ri­or to oth­ers near­by.

In the al­go­rithm, when a scout found a new nest, it de­cid­ed how good it was and ad­ver­tised it to the rest of the col­o­ny by lead­ing more ants there. Scouts kept track of how many oth­er ants liked the nest to help de­cide wheth­er it should be­come their new home. 

In a cri­sis, scouts simp­ly searched more, be­gan ad­ver­tis­ing sites soon­er, and waited for few­er nest­mates to agree with them be­fore start­ing to move, the re­search­ers found.

The ants thus “tune the param­e­ters of a sin­gle de­ci­sion al­go­rithm to re­spond adap­tive­ly to two dis­tinct prob­lem­s,” they wrote in this week’s ear­ly on­line edi­tion of the research journal Pro­ceed­ings of the Na­tion­al Aca­de­my of Sci­ences. 

“A tun­a­ble al­go­rithm rep­re­sents a gen­er­al means for com­plex bi­o­log­i­cal sys­tems to solve mul­ti­ple prob­lems with­out need­ing spe­cif­ic so­lu­tions for each one,” they added. T. cur­vi­spi­no­sus col­o­nies typ­i­cal­ly live in rock crevices or hol­low nuts.

Pratt and Sump­ter said the find­ings may help bi­o­lo­gists un­der­stand the de­sign of sys­tems as di­verse as bac­te­ri­al col­o­nies and hu­man brains.

* * *

Ants of the species Temnothorax curvispinosus follow just one basic set of rules for choosing a new home for their group. But by making a few simple changes, they can adjust that playbook to fit almost any situation ranging from casual new-home shopping to total urgency, researchers have found. Scientists are interested in the “algorithms,” or step-by-step rules, that organisms use to make decisions, either individually or collectively. Life’s adaptability depends in part on the fact that such algorithms are straightforward, yet easily adjustable to different situations, wrote the scientists, Stephen Pratt of Princeton University in Princeton, N.J. and David Sumpter of the University of Oxford, U.K. Some aspects of the human brain may also show such properties, they added. Many ants use an algorithm like that of T. curvispinosus to search for a new home, according to Pratt and Sumpter. Of an ant colony’s several hundred members, a few are always scouting for new homes. The researchers investigated how ant colonies choose a new home when their old nest is completely destroyed versus when it is merely inferior to others nearby. In the algorithm, when a scout found a new nest, it decided how good it was and advertised it to the rest of the colony. Scouts kept track of how many other ants liked the nest to help decide whether it should become their new home. In a crisis, scouts simply searched more, began advertising sites sooner, and waited for fewer nestmates to agree with them before starting to move, the researchers found. The ants thus “tune the parameters of a single decision algorithm to respond adaptively to two distinct problems,” they wrote in findings published in this week’s early online edition of pnas. “A tunable algorithm represents a general means for complex biological systems to solve multiple problems without needing specific solutions for each one.” T. curvispinosus colonies typically live in rock crevices or hollow nuts. Pratt and Sumpter said the findings may help understand the design of biological systems as diverse as bacterial colonies and human brains.