When in a tight spot, animals "lie" to their own kind to get what they want, a University of Rochester biologist has found. In work described in the current issue of the Journal of Theoretical Biology, Eldridge Adams shows that within a single species, it is possible for some members to deceive others.

By proving that the weaker are able to deceive the stronger to survive, Adams' findings runs counter to a common belief by biologists that communication within a species must always be reliable and honest.

"We've shown that the communication is not always reliable, and that in theory, you shouldn't expect it to be," says Adams, who began the work as a graduate student at the University of California at Berkeley.

Biologists have long recognized that deception is commonplace in communication between different species. But most believed bluffing among animals of a single species should be rare or impossible.

To demonstrate how widespread deception between members of the same species could be, Adams watched hundreds of confrontations between two-inch, crayfish-like stomatopod crustaceans, whose bluffing was originally observed by behavioral ecologists Roy Caldwell and Rick Steger. Working with co-author Michael Mesterton-Gibbons of Florida State University, Adams modeled a confrontation mathematically and proves in a game theory model that bluffing is possible in a stable communication system.

Adams studied one species of stomatopod crustacean, Gonodactylus bredini, in Panama while doing graduate work under Caldwell. The creatures, known to sushi connoisseurs as mantis shrimp, live in cavities in shallow, tropical waters.

Stomatopods typically compete over these cavities, and when a confrontation looms, they can either flee or fight. If the creature chooses to fight, it often invokes a threat display, holding its appendages out to the side and lifting its head aggressively. If the opponent chooses to fight, the creatures use their appendages to whack each other with considerable force. (Stomatopods are good at this: They regularly beat senseless crabs and other hard sea creatures before cracking the shells open and gulping them down; their striking force is so great that marine biologists have nicknamed them "thumb busters.")

If threats were always honest, or reliable, only the strongest creatures -- those able to back up a threat with deadly force -- would menace others. But both the strongest and the weakest stomatopods threaten. In fact, creatures rely on threats most when they are weakest: immediately after molting, when their new skeleton is still hardening up and they don't even have hard appendages capable of crushing an opponent. "These threats often drive off opponents," says Adams, "yet they're certainly not reliable. These animals are bluffing, and they would be readily killed in a fight."

Biologists have long thought that such widespread bluffing wasn't possible in a stable communication system. If weak animals use threats deceptively, they reasoned, threats would become so common that the communication system would quickly break down because other animals wouldn't pay attention to any threats, whether real or deceptive. Adams and Mesterton-Gibbons have shown that a system can remain stable even though some members bluff successfully.

Adams likens the phenomenon to a game of draw poker between two people. "Each person knows his or her own strength, but not the strength of the opponent. Sometimes the optimal strategy is to threaten the opponent by raising the stakes, even though you have a weak hand."

This work was funded by the Smithsonian Institution and the Packard Foundation. tr